AI Meets BI: Artificial Intelligence and

Business Intelligence 1st Edition

Lakshman Bulusu
Visit to download the full and correct content document:
https://textbookfull.com/product/ai-meets-bi-artificial-intelligence-and-business-intellig
ence-1st-edition-lakshman-bulusu/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Oracle Business Intelligence with Machine Learning :

Artificial Intelligence Techniques in OBIEE for
Actionable BI 1st Edition Rosendo Abellera

https://textbookfull.com/product/oracle-business-intelligence-
with-machine-learning-artificial-intelligence-techniques-in-
obiee-for-actionable-bi-1st-edition-rosendo-abellera/

Clinical Neurotechnology meets Artificial Intelligence

Philosophical Ethical Legal and Social Implications 1st
Edition Orsolya Friedrich

https://textbookfull.com/product/clinical-neurotechnology-meets-
artificial-intelligence-philosophical-ethical-legal-and-social-
implications-1st-edition-orsolya-friedrich/

AI and UX: Why Artificial Intelligence Needs User

Experience Gavin Lew

https://textbookfull.com/product/ai-and-ux-why-artificial-
intelligence-needs-user-experience-gavin-lew/

Artificial Intelligence XXXVII 40th SGAI International

Conference on Artificial Intelligence AI 2020 Cambridge
UK December 15 17 2020 Proceedings 1st Edition Max
Bramer
https://textbookfull.com/product/artificial-intelligence-
xxxvii-40th-sgai-international-conference-on-artificial-
intelligence-ai-2020-cambridge-uk-
Artificial Intelligence XXXVI 39th SGAI International
Conference on Artificial Intelligence AI 2019 Cambridge
UK December 17 19 2019 Proceedings Max Bramer

https://textbookfull.com/product/artificial-intelligence-
xxxvi-39th-sgai-international-conference-on-artificial-
intelligence-ai-2019-cambridge-uk-
december-17-19-2019-proceedings-max-bramer/

AI for the Good: Artificial Intelligence and Ethics

(Management for Professionals) Stefan Vieweg

https://textbookfull.com/product/ai-for-the-good-artificial-
intelligence-and-ethics-management-for-professionals-stefan-
vieweg/

Playing Smart: On Games, Intelligence, and Artificial

Intelligence 1st Edition Julian Togelius

https://textbookfull.com/product/playing-smart-on-games-
intelligence-and-artificial-intelligence-1st-edition-julian-
togelius/

Artificial Intelligence By Example Develop machine

intelligence from scratch using real artificial
intelligence use cases Denis Rothman

https://textbookfull.com/product/artificial-intelligence-by-
example-develop-machine-intelligence-from-scratch-using-real-
artificial-intelligence-use-cases-denis-rothman/

Beginning Game AI with Unity: Programming Artificial

Intelligence with C# Sebastiano M. Cossu

https://textbookfull.com/product/beginning-game-ai-with-unity-
programming-artificial-intelligence-with-c-sebastiano-m-cossu/
AI MEETS BI
Artificial Intelligence and
Business Intelligence
AI MEETS BI
Artificial Intelligence and
Business Intelligence

Lakshman Bulusu
Rosendo Abellera
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2021 by Taylor & Francis Group

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-0-367-33260-0 (Hardback) 978-0-367-64381-2 (Paperback)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to
publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials
or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material repro­
duced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any
form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming,
and recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copy­
right.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400.
CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have
been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identifica­
tion and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
 Trademarks Covered in This Book

BIS3® is a registered trademark of Business Intelligence Software Service Solution.

Facebook™ is a trademark of Facebook, Inc.
IBM® and IBM Watson® are registered trademarks of International Business Machines Corporation.
Keras is neural network–based deep learning library under the open source umbrella. It is licensed by MIT.
Microsoft® is a registered trademark of Microsoft Corporation.
MNIST is a database of images of handwritten digits created from National Institute of Standards and Technology
(NIST) data sets.
Oracle® is a registered trademark of Oracle Corporation
Python® is a registered trademark of the Python Software Foundation.
Qteria® is a BIS3® company.
SparkCognition™ is a trademark of SparkCognition.
Tableau® is a registered trademark of Tableau Software.
TensorFlow® is a registered trademark of Google, Inc.
Dedication

I dedicate this book to the memory of my parents, Professor B.S.K.R. Somayajulu and Smt. B. Sita, for
their untiring efforts in nurturing me to become who I am today and their inspiring acumen and advice
that have been a constant beacon throughout my career.
I also dedicate this book to my uncle, Professor B.L. Deekshatulu, and my aunt, the late Smt. B.
Kameswari, for their blessings, which have enabled me to rise in life, both professionally and personally.
And, finally, I dedicate this book to all the readers of my previous books and their invaluable feed­
back, which has gone a long way in shaping this book.
— Lakshman Bulusu

vii
Contents

Dedication vii

Contents ix

Preface by Lakshman Bulusu xiii

Preface by Rosendo Abellera xv

Acknowledgments xvii

About the Author xix

Chapter 1 Introduction 1
In This Chapter 1
1.1 Introduction 1
1.2 Traditional Ways of Enabling BI 3
1.3 Three Generations of BI 3
1.4 How Business Fits into BI—From Business Data to Business
Decision Making 7
1.5 Summary 8

Chapter 2 AI and AI-Powered Analytics 9

In This Chapter 9
2.1 Introduction 9
2.2 AI and Its Rise in the Modern Enterprise 9
2.3 How AI Changes the Enterprise BI Landscape 10
2.4 The Analytics Sphere 10
2.5 BI-Enabled vs. AI-Powered Analytics 13
2.6 From Data to Intelligent Decision Making to Insightful Decisions 13
2.7 AI-Powered Techniques for BI 15
2.8 Summary 17

ix
x AI Meets BI

Chapter 3 Industry Uses Cases of Enterprise BI—A Business Perspective 19

In This Chapter 19
3.1 Introduction 19
3.2 Classifying Commodity Saleable Grade Based on Its Attributes 20
3.2.1 Descriptive Analytics 20
3.3 Predicting Commodity Prices in Advance 33
3.3.1 Predictive Analytics 33
3.4 Recommender Systems to Suggest Optimal Choices Based on Score 46
3.5 Automatic Image Recognition 46
3.6 Summary 48

Chapter 4 Industry Use Cases of Enterprise BI—The AI-Way of Implementation 49

In This Chapter 49
4.1 Introduction 49
4.2 Classifying Commodity Saleable Grade Based on Their Attributes 50
4.2.1 The AI-Methodology—Descriptive Analytics—Using the
Random Forest Machine Learning Algorithm and Comparing
It with Decision Trees 50
4.2.2 The BI-Enablement and Its Impact on the Enterprise 112
4.3 Predicting Commodity Prices in Advance 113
4.3.1 The AI-Methodology—Predictive Analytics—Using Neural
Networks 113
4.3.2 The BI-Enablement and Its Impact on the Enterprise 121
4.4 Recommender Systems to Suggest Optimal Choices Based on Score 122
4.4.1 The AI-Methodology—Prescriptive Analytics by
Recommending Viable Sources Based on Score Prediction
(using DNNs) and Predictive Analytics by Predicting “How
the Product Fares in the Market” Based on the
Recommended Choices 122
4.4.2 The BI-Enablement and Its Impact on the Enterprise 182
4.5 Automatic Image Recognition 195
4.5.1 The AI-Methodology—Prescriptive and Augmented
Analytics—Using Deep Learning–Based Convolutional
Neural Networks for Classification 195
4.6 Summary 207
4.7 References 208

Chapter 5 What’s Next in AI Meets BI? 209

In This Chapter 209
5.1 Introduction 209
5.2 AI-Powered Cognitive Computing 209
5.3 Security and Governance in AI-Powered BI 211
 Contents xi

5.4 The Trust Factor in AI-Powered BI 212

5.5 Summary 213

Index 215
Preface

This book focuses on the primary aspect of “AI meets BI” and details the intersection of AI, data sci­
ence, machine learning, deep learning, and cognitive computing and how these technologies can assist
and/or automate BI and decision making, resulting in more informed business insights. It stands out
as the first publication to address the integration of artificial intelligence tools and technologies from
the multitude of such tools and technologies available today to boost and enhance business intelligence.
It covers various AI-powered analytics for BI-enabled decision making. Specifically, it discusses four
industry use cases and their implementation, using AI-enabled machine learning and deep learning
techniques along with their integration into enterprise BI platforms, by making use of the abovemen­
tioned technologies. It also discusses how all types of data, including Big Data, can be scaled to enter­
prise level and thereby enhance customer experience. Details are discussed from a business perspective
and the AI-way of implementation.
The intended audience for this book includes business and data analysts who need to interpret the
output of AI-powered BI systems from a business perspective; data scientists; data engineers; AI and
BI practitioners, designers, programmers, and technical/project managers involved in AI/BI solution
implementations; academia and research scientists involved in teaching/training of AI/BI models; and,
last but not least, teachers and students of data science/AI/BI courses at the graduate level and/or in
career-based learning courses. It is also intended for anyone involved in instructor-led or online training
in AI and BI who will benefit from augmenting their courses with proven AI techniques for BI that will
prepare the students for on-the-job solutions and certifications.
From data sources to deployment models (of AI), enterprise BI has most recently entered its third
phase—one powered by AI. There are many aspects that address renewing focus on the capabilities of
today’s artificial intelligence. However, in light of how the corporate world touches our everyday lives,
artificial intelligence–enhanced business intelligence would now seem to have the utmost importance
in affecting how we work on a daily basis. It is for this reason that this book helps to pave the techno­
logical path of AI meets BI. In addition, this topic, never before offered, will provide important details
about new ways and tools that will lead us into the future of business intelligence—a future enhanced
with opportunities never before effectively available and only imagined for artificial intelligence.
Beginning with an outline of the traditional methods for implementing BI in the enterprise and
how BI has evolved into using self-service analytics, data discovery, and, most recently, AI-powered BI,
Chapter 1 lays out the three typical architectures of the first, second, and third generations of BI in the
enterprise. It touches upon the “business” paradigm in the context of BI in the enterprise and how both
BI and business analytics complement each other to provide a holistic BI solution—one that empowers
all types of users, from business analysts to end users.

xiii
xiv AI Meets BI

Chapter 2 outlines AI and AI-powered analytics in some depth, describing the various types of ana­
lytics using an analytics sphere and highlights how each of these can be implemented using AI-enabled
algorithms and deep learning models. In particular, it touches upon AI and its rise in the modern
enterprise. It provides a compare-and-contrast discussion of BI-enabled and AI-powered analytics and
shows how businesses can leverage these techniques to tailor business processes to provide insight into
how enterprises can gain a competitive edge. Chapter 2 also provides information on how AI techniques
can be used to work upon the data, not vice versa, including the leveraging of data augmentation, data
discovery, augmented analytics, and other such techniques. It demonstrates how the implementation
of such AI-based techniques can be done right in the BI platform, and how to integrate the results into
the same.
Chapters 3 and 4 form the crux of the book. They discuss, in great detail, the four industry use cases.
Chapter 3 focuses on the business perspective of these use cases. It describes how an enterprise
can access, assess, and perform analytics on data by way of discovering data, defining key metrics that
enable the same, defining governance rules, and activating metadata for AI/ML recommendations. It
highlights how business users can explain key variables that serve as drivers for the business process in
context, perform interactive correlations to determine variable importance, associate results with time,
and, most important, how to predict outcomes based on data using AI-powered deep learning mod­
els—business analytics anytime and anywhere—and on all kinds of data and in real time. In addition,
Chapter 3 highlights how enabling power users to create visuals based on these and other embedded
metrics for deep-dive analysis in an interactive fashion is key for an enterprise to remain competitive.
Four use cases that describe descriptive and predictive analytics in terms of predicting likelihood as well
as forecasting and predicting measures for the future, along with prescriptive analytics, are outlined.
Chapter 4 takes a very in-depth approach in describing the implementation specifics of each of these
four use cases by way of using various AI-enabled machine learning and deep learning algorithms and
shows how the results enable actionable BI. The key component of predictive capabilities that was miss­
ing in the second generation of BI is brought to light, and it explains how it can be executed by way of
AI-powered BI. Complete code for each of the implementations is provided, along with the output of
the code supplemented by visuals that aid in BI-enabled decision making.
Chapter 5 touches upon what’s next in AI meets BI. It begins with a brief discussion of the cogni­
tive computing aspects of AI as well as augmented analytics and automated and autonomous BI. It then
outlines how security and governance can be addressed in AI-powered BI. It ends with a note on the
ethical aspects of BI and the trust factor involved in AI-powered decision making.
BI has evolved from its initial landscape of OLAP-based reporting and analysis to ad hoc query­
ing, operational BI, and self-service analytics to AI-powered business analytics and big data discovery.
And many of the industry’s leading BI vendors, such as Tableau®, Microsoft®, Oracle®, IBM®, etc., have
already started offering AI-powered capabilities to enable all kinds of users (technical and nontechnical)
to take advantage of these new technologies and trends and derive intelligent insights that can lead to
better, smarter, faster BI. This book comes at a time when this changing landscape for BI is in place,
and the pragmatics of AI in the production and democratization/monetization of AI, as well as their
implementation specifics from a BI perspective, are entering the mainstream of enterprises worldwide.
It provides specifics regarding the most widely used AI-based technologies and their role in enterprise
BI, thus positively reflecting the real-world scenarios that make AI-powered BI a reality.
— Lakshman Bulusu
Preface

Business Intelligence is dead. Long live Business Intelligence—with Artificial Intelligence, that is!
With the emergence of Artificial Intelligence (AI) in the business world, a new era of Business
Intelligence (BI) has been ushered in to create real-world business solutions using analytics. These ana­
lytical capabilities, which were previously either not available or just difficult to use, have now become
feasible and readily usable. For BI developers and practitioners, we now have new tools and technolo­
gies to complete and enhance the job that we set out to do—that is, provide our users (or customers)
with the systems and solutions needed to provide effective decision making that stems from reliable and
accurate information and intelligence, which, in turn, leads to valuable, actionable insights for business.
Oftentimes in the past, BI was stymied by bad or incomplete data, poorly architected solutions,
or even just outright incapable systems or resources. The industry has come a long way in combat­
ting these obstacles, and now, with the advent of AI, BI has renewed hope for true effectiveness. This
is a long-awaited phase for practitioners and developers, and moreover, for executives and leadership
depending on knowledgeable and intelligent decision making for their respective organizations.
More than 20 years ago, after serving in military intelligence and with a passion for software devel­
opment, I began my career in BI. I met Lakshman about 10 years ago. By that time, he was already an
established, published author, and he introduced me to his world of book writing. I quickly embraced
his expertise, and, together, we began this partnership of writing about a subject of our choice that
we both so deeply enjoyed—that is, BI and analytics. Therefore, when it became evident that AI was
becoming more prevalent and, more importantly, feasible to augment BI, we quickly jumped at the
opportunity to write about this game changer. Together, we authored the first published book on
Oracle BI and AI with machine learning.* It was received quite well. An audience began to form around
it, and many even reached out for help in implementing the system and using it to provide a solution
to their AI problems. In that book, we used a single case study to illustrate how AI could be effectively
applied to BI.
With its success, Lakshman suggested and pushed for us to go beyond a single vendor and a single
case study. Thus, this book begins that journey, as it presents numerous ways that AI can be applied
to BI. Our hope is that it opens up a dialog about how we can best proceed and further augment and
enhance BI to effectively fulfill its promise of successfully implementing advanced and predictive ana­
lytics in various applications for a multitude of industries—especially, with vital case studies that may
affect many people.

* Abellera, R. and Bulusu, L. (2017). Oracle Business Intelligence with Machine Learning, 1st ed. New York (NY):
Apress.

xv
xvi AI Meets BI

For instance, it was about 5 years ago that we, as a company, began to develop a software applica­
tion with the goal of utilizing genomics for prescriptive treatments. This goal was only possible with
the use of AI coupled with BI. It was rather an early attempt, as we used the first version (1.0, in fact)
of Microsoft®’s machine learning tool. We struggled to implement it with this early tool and to con­
vince investors of the capabilities, or even feasibility, of our vision for healthcare. This was back in the
2014, and in software development years, that seems like a lifetime ago. Many new developments and
improvements in AI have materialized since then. Now it is 2020, and with a global pandemic, I can’t
help but think of the many ways that the new technologies in AI and machine learning could have pos­
sibly aided and been effectively used to accurately predict, prescribe, and prevent the unfortunate events
that have transpired. If we had somehow been able to clearly foresee the future, we certainly would have
focused on that goal. Perhaps that could be the ultimate use and application for AI, as it satisfies the
demands of BI as it pertains specifically to healthcare in matters of life and death. For those case stud­
ies, providing a solution would definitely be a worthwhile cause. In that field, we’ve only really begun
to scratch the surface.
The possibilities are truly endless. New capabilities are now at hand as AI meets BI. For our future,
we can only hope.
— Rosendo Abellera
Acknowledgments

I thank my immediate family for their cooperation and patience during the time that I wrote this book.
I thank my brother Dr. B. Rama for his guidance on predictive analytics, which helped me to write
related parts of this book.
I thank Mr. Linesh Dave and Mr. Praveen Sharma for judging my book proposal and giving the
go-ahead to this book project.
And I thank Theron R. Shreve (Director) and Marje Pollack (copy editor and typesetter) at
Derryfield Publishing Services, as well as the publisher, CRC Press/Taylor & Francis Group, for the
support, feedback, reviews, and production work they provided to enable this book project to see the
light of day.
— Lakshman Bulusu

xvii
About the Authors

Lakshman Bulusu is a veteran IT professional and data scientist, with 28

years of experience in the IT industry. He has worked at major industry
verticals in the retail, banking, pharma/health care, insurance, media,
telecom, and education fields. He currently consults for a major banking
client in the New York/New Jersey metropolitan area. He has exper­
tise in RDBMS technologies, including Oracle®, MS SQL Server, and
Vertica and their related technologies. He is also well versed in the latest­
and-greatest technologies such as artificial intelligence, data science, and
business intelligence.
Lakshman also serves as Vice President of Research at Qteria.com.
When not at his job, he lectures at various technical schools, user group conferences and summits,
and data science meetings. He also devotes his free time to writing poetry in English and in his native
language, Telugu.
He holds a Masters credential from Oracle Corporation and Master certifications from Brainbench,
and he is an OCP-certified Professional. He is a Barnes & Noble Educator and holds a Certification of
Appreciation from Barnes & Noble.
He has authored 10 books spanning topics such as Oracle, Oracle BI, Open Source DW/BI, and
Oracle and machine learning, as well a host of papers in technical magazines and journals.
— Lakshman Bulusu

With a long and proven track record of success in the Business Intelligence
(BI) industry throughout several decades, Rosendo Abellera is a subject
matter expert and expert practitioner in business intelligence and analytics.
As a career consultant, he has serviced numerous leading global commer­
cial clients and major US government organizations. A strategist as well as a
hands-on developer, he architected and implemented complete, holistic, and
data-centric decision-making systems and solutions from the ground up—
from complex data warehouses for Revenue Recognition to advanced dash­
boards for Financial Analytics.
Rosendo has held key management positions in establishing the Business
Intelligence and Analytics practices of several global consulting organizations. Moreover, he founded
and established BIS3®, a successful consulting firm started more than 10 years ago, specializing in what

xix
xx AI Meets BI

was then Oracle®’s new BI strategy with OBIEE and Essbase—both of which are subjects of his previ­
ously published books (jointly with Lakshman Bulusu). In an additional role for the company, he is the
chief architect of Qteria®, an AI and machine learning platform.
Rosendo is a veteran of the US Air Force and the National Security Agency (NSA), where he served
worldwide as a cryptologist and linguist for several languages. With these beginnings in the intelligence
community, he provides unique insight and knowledge for utilizing data as a critical asset.
— Rosendo Abellera
Chapter 1
Introduction

In This Chapter

1.1 Introduction
1.2 Traditional Ways of Enabling BI
1.3 Three Generations of BI
1.4 How Business Fits into BI—From Business Data to Business Decision Making
1.5 Summary

1.1 Introduction

This chapter begins with an outline of the conventional methods for implementing BI in an enterprise
and highlights how BI has evolved into using self-service analytics and, most recently, AI-powered BI.
It touches upon the three generations of BI and how DW/BI solutions have changed the landscape of
business decision making for the better. Finally, it ends with notes on how the commonly used term
“business” aligns with and can be fit into a comprehensive BI solution that is not only state of the art
but also one that continually and automatically evolves on metrics or key performance indicators (KPIs)
that the BI solution outputs, which, in turn, can be refueled into the same architecture as input, lead­
ing to more insightful decisions. The primary aspect of “pairing AI with BI” on a one-on-one basis
as well as its details, including the intersection of AI, data science, machine learning, deep learning,
cognitive computing, and how this can lead to the automation of BI and decision-making processes
resulting in more informed business decision making, is briefly covered in Section 1.3. There are many
topics with renewed focus on the capabilities of today’s artificial intelligence. On the surface, admit­
tedly, it may not seem as interesting, perhaps, as artificial intelligence, which is igniting new interest,
for example, in the areas of self-driving cars or robotics. However, in light of how the corporate world
touches our everyday lives, artificial intelligence enhancing business intelligence would now seem to
have the utmost importance in affecting how we work on a daily basis. It is for this reason that Chapters
2 through 5 of this book provide the coverage needed to define a technological path that is filled with
new methods and tools that will bring us to business intelligence’s future—a future never before effec­
tively available and only imagined for artificial intelligence.

1
Figure 1.1 A typical architecture of a traditional (first-generation) EDW/BI implementation.
 Introduction 3

1.2 Traditional Ways of Enabling BI

Any BI-enabled ecosystem involves the integration of all of the tools and technologies—from
data preparation (including data quality) to data integration (via ETL/ELT or data virtualiza­
tion) to data warehousing (including data lakes that act as sources to data warehouse) to data
presentation (via data marts, data warehouses, and/or BI dashboards) to data visualization, and
from decision support to decision making—presented by today’s plethora of artificial intelli­
gence tools and technologies to boost and enhance business intelligence. In addition, data gover­
nance needs to be applied to each of these phases to meet regulatory requirements etc.
Data virtualization and data lakes entered the EDW/BI scenario as part of the so-called Data
Integration 2.0 technologies.
Figure 1.1 shows a typical architecture of a comprehensive EDW/BI solution. As can be seen from
this figure, there are three major components of the EDW/BI solution:

1. Data Integration—This consists of collecting data from multiple sources and in multiple for­
mats, including mostly structured or relational data that can be massaged into a common format
that can be then be integrated into the subsequent data warehousing component.
2. Data Warehousing—This consists of staging the data from the data integration component into
a landing zone that is responsible for massaging the data. On the basis of the data sources and the
type of data, a logical dimensional model is constructed that clearly lays out the dimensions and
facts based on business processes. This results in a STAR schema wherein a fact is surrounded by
multiple dimensions in a star format. This logical model forms the core of the EDW that stores
data in a semantic model for inter-day and as-is and as-of reporting. Multiple Data Marts are
created off of the EDW to cater to reporting on a departmental level of the enterprise. Each data
mart is modeled as a STAR schema.
3. Business Intelligence (BI) and Reporting Component—This consists of the dashboarding and
reporting of data (based on canned reports using a BI tool that is integrated with the EDW to
enable business analysts and user interaction of the data. These users are given access to the data
to use predefined KPIs.

Although the above architecture shows EDW/BI at an enterprise level, it had limitations, such
as no intra-day reporting (Operational BI), no real capability to enable users to create their
own KPIs via interactive dashboarding (self-service analytics), and ad hoc reporting. The next
section outlines how the EDW/BI landscape evolved to result in second- and third-generation
EDW/BI solutions.

1.3 Three Generations of BI

The first generation of BI was based on a primarily OLAP-based solution, which supported mostly
relational data formats and used ETL/ELT for data integration. The EDW was designed based on the
STAR-schema, which was also relational and could be extended to Multidimensional Online Analytical
Processing (MOLAP). Figure 1.1 depicts a typical first-generation BI architecture. First-generation
reporting and BI were limited to nightly reporting based on batch loads and support for real-time data
integration but no ad hoc reporting. In addition, operational BI or, in other words, analysis of data as it
was created was not feasible, and BI capabilities did not extend to enable users to create their own KPIs
or self-service analytics.
These needs were addressed by the second generation of BI architecture, which introduced analytics
as part of BI and, additionally, took advantage of Data Integration 2.0, which came with robust data
4 AI Meets BI

integration technologies such as data virtualization and data integration pipelines and EDW enhance­
ments such as data lakes (that housed raw data) and could be used as a source for EDW.
The three main capabilities of second-generation BI included ad hoc querying, operational BI, and
self-service analytics. It could also leverage data virtualization, which enabled data integration at scale.
Second-generation BI also took care of BI at scale by enabling the use of data lakes and data agility.

With the ability to perform analytics on data, enterprises could find the whys of what occurred in the
past using such data or sometimes predict the future using this information (business analytics part)
in addition to getting to know what happened and how it happened in the past based on the data
(business intelligence part). This meant that causation and correlation are part of the analyses and
are both needed to perform the analyses. In addition, both are not the same. The second generation
of BI supports business intelligence and business analytics, leading to a holistic view of business and
the ability to make decisions that are more insightful. This could also pave the way for predicting the
future using these analytics and intelligence such as how a particular decision could evolve if and when
the data evolves. In addition, a key component of the second generation of BI was to put busi­
ness analytics in the hands of the business analyst or end user, thus leading to self-service
analytics. Accordingly, BI tools were enhanced to include these new trends as new ways to integrate
into already-existing BI ecosystems. From a business analyses perspective, BI happens before Business
Analytics, and while the former can discern what the criteria are that lead to business competition, the
latter allows you to discern why such criteria lead to business competition of your enterprise. And both
BI and BA fall under the larger umbrella of business analyses.
By using analytics, an enterprise can design metrics that aid in decision making, which, in turn,
enables competition—also called competitive intelligence. In comparison, BI uses metrics that measure
the effectiveness of a business goal within your enterprise. Metrics are like atomic measures that track
business processes.

Figure 1.2 depicts a typical second generation of BI architecture.

Third-generation BI is based on AI meets BI (artificial intelligence meets business intelligence), and
it enables various AI-powered analytics for BI-enabled decision making. It is unique in the sense that it
leverages the intersection of AI, data science, machine learning, deep learning, and cognitive comput­
ing and how these technologies can assist and/or automate BI and decision making. The major aspects
of AI-powered BI are as follows:

• Real-time EDW and BI.

• Automated, augmented, and autonomous BI, including augmented analytics that uses a com­
bination of human and machine intelligence and the automation of content-and-context
involved for relevance and importance of BI-enabled insights and accuracy and transparency of
the same across the enterprise level. Simply put, it enables automated and extended data analyses
and analytics.
• Big Data Discovery using AI-powered search and its integration into the BI landscape that is
efficient and facilitates informed decision making at the enterprise level and scale via transformed
(big) data analytics.
• AI-powered EDW/BI has eased the ability to harness data in varied formats, such as structured,
semi-structured, and unstructured and real-time data. It has also handled the rate at which such
data was created, as well as the huge volume of the same, while simultaneously enabling valida­
tion of the data and maintaining the quality of such data.
• Agile analytics.
• Enterprise-level scalability and how customer experience can thereby be enhanced.
Figure 1.2 A typical second-generation EDW/BI architecture.
Figure 1.3 A typical third-generation AI meets BI architecture.
 Introduction 7

• A holistic EDW/BI solution that can be used by both technical and non-technical users at all
levels of the enterprise.
• Intelligent automation of business and operational processes.
• Delivery of BI-as-a-service for mobility and on-the-go customer analytics.
• Data monetization at the enterprise level.
• Acceleration and enhancement of competitive BI. As an example, BI can have deep neural net­
work solutions embedded within its implementation stack to realize competitive results. Use cases
of such an implementation include image tagging and image-recognition solutions.

Using AI for BI leverages how to use an AI model in context from business by recommending a particular
AI model/machine learning algorithm based on the use case inputs and outputs. It details the aspects
that have not yet been deeply explored in the BI landscape, such as the relevance and importance of BI-
enabled insights and accuracy and transparency of the same across the enterprise/corporate industry level.

Figure 1.3 depicts a typical third-generation BI architecture that is powered by AI.

Third-generation BI is of specific advantage to business analysts, data analysts, data scientists, and
data engineers as well as end users. As stated in the beginning of this section, BI has evolved from its
initial landscape of OLAP-based reporting and analysis (first generation of BI) to ad hoc querying,
operational BI, and self-service analytics (second generation of BI) to AI-powered business analytics
and big data discovery (third generation of BI). And many of the industry’s leading BI vendors, such
as Tableau®, Microsoft®, Oracle®, IBM®, etc., have already started offering AI-powered capabilities to
enable all kinds of users (technical and non-technical) to take advantage of these new technologies and
trends and to derive intelligent insights that can lead to better, smarter, faster BI.

1.4 How Business Fits into BI—From Business Data to

Business Decision Making
Every enterprise has a business landscape that has certain objectives or goals to meet to remain competi­
tive in the industry. To this end, it is composed of a number of business processes that define how the
enterprise is organized to meet its goals. For example, a product manufacturing enterprise has the objec­
tives of maximizing its sales, profit, and its supply chain, which help it stay ahead of the competition.
And each of these objectives comprises one or more business processes that are, in turn, measured by
metrics that track their status. In addition, there are KPIs that track and measure how well the enter­
prise is meeting a certain business objective. These KPIs measure the effectiveness of a business goal
and constitute what is called the business intelligence component of the enterprise. BI needs data that is
current as well as historical to determine the KPI measurements based on which analyses are done that,
in turn, determine the “what happened” and “how it happened” aspects. In addition, these two aspects
determine whether and how well the enterprise has met its goals. To this end, data visualization based
on this data enables a business user and/or end user to perform “what” and “how” analyses that eventu­
ally aid in determining how well a particular business objective was met and provide an idea of what
can be done to further improve the KPI that led to enterprise growth. As an example, visual charts that
depict the trend analysis and dashboards that group charts based on combinations of data and drill-up/
drill-down/drill-across capabilities can aid business users to discern the correlation between data effects.
But BI alone cannot determine the why of data analysis that is very important to actually make deci­
sions based on the reasoning. This is where business analytics enters the picture. As a second aspect
of business analyses, business analytics helps in the decision-making process by aiding the prescriptive
8 AI Meets BI

part (business action to be taken) of a solution as well as, sometimes, the predictive piece of the solution.
Both the business decision made and the business action taken help in determining whether the deci­
sions made and subsequent actions taken can help. Furthermore, enabling business users to create their
own analytics based on the data presented goes a long way in insightful decision making that drives the
enterprise to competitive success. Coming back to the product manufacturing enterprise example, busi­
ness analytics can help in determining what decisions can be taken when a particular product evolves
by way of forecasting sales growth in such a scenario. It is in this predictive analytics scenario that AI
can make a difference by way of intelligent decision support.

Business and business users play a pivotal role by describing the enterprise goals and business processes,
the KPIs, and the metrics necessary for BI as well as the decisions and the subsequent actions to be
taken. Business can also define the type of data visualizations required for analysis as well as those that
enhance the decision-making process.

1.5 Summary

This chapter discussed the traditional method of data warehousing and BI and then outlined the three
generations of BI by way of a typical architecture based on each. Finally, it touched upon how “busi­
ness” fits into the BI land analytics landscape. The next chapter describes AI and AI-powered analytics
and their role in BI-enablement.
Another random document with
no related content on Scribd:
 And then after Young had come Karel, the son of rich parents and
well-placed socially in Braleigh. He was young, well-informed, a snob
of sorts, although a gentle one. The only world he knew was that in
which his parents had been reared. Their ways had been and always
would be his, conservatism run mad. At thirty the only place to go in
summer was Macomber Beach, and in winter the only place to be
was in Braleigh. There he could meet his equals twenty times a day.
They went to the same homes, the same hotels, the same parties
the year round. It was all the life he wanted, and it was all the life she
would have been expected to want. But by then she was being
hopelessly held by this greater vision and something within had said:
“No, no, no!”
“You were making over my ermine cape. Is it finished?”
And Loring! He, for a change, was a physician there in Braleigh
and lived with his sister in Lankester Way, near her home, only hers
was in a cheaper street. He was young and good-looking but
seemed to think only of his practice, how it was to make him and
achieve her perhaps, although it had all seemed so commonplace
and practical to her. He was so keen as to his standing with the best
people, always so careful of his ways and appearance, as though his
life depended upon it. He might have married more to his social and
financial advantage but he had wanted her. And she had never been
able to endure him—never seriously tolerate his pursuit.
“Yes, if you would alter these sleeves I might like it.”
Whenever he saw her he would come hustling up. “My, but it’s nice
to see you again, Ulrica. You are always the same, always charming,
always beautiful—now don’t frown. Have you changed your mind
yet, Ulrica? You don’t want to forget that I’m going to be one of the
successful men here some day. Please do smile a little for me. I’ll be
just as successful as Joyce or any of them.”
“And is it just success you think I want?” she had asked.
“Oh, I know it isn’t just that, but I’ve had a hard time and so have
you. I know it wouldn’t do any good to offer you only success, but
what I mean is that it makes everything so much easier. With you I
could do anything—” and so he would ramble on.
“To McCafferey’s, the Post Street entrance.”
But the shrewd hard eyes and dapper figure and unvaried
attention to his interests had all bored and after a time alienated her,
since her ideal seemed to dwarf and discolor every one and
everything. Was there not something somewhere much bigger than
all this, these various and unending men, she had asked herself,
some man not necessarily so successful financially but different?
She had felt that she would find him somewhere, must indeed if her
life was to mean anything to her. Always her great asset, her beauty,
had been looked upon as the one thing she must keep for this other.
And so it had gone, man after man and flirtation after flirtation. It had
seemed as though it would never end. Even after she had
transferred her life to the great city, to work, to go upon the stage if
need be, there were more of these endless approaches and
recessions; but, like the others, they had come and gone, leaving
only a faint impression. Not until that day at Althea’s party in the
rooming-house in which they both lived had she found the one who
touched her.
And then—
“And now to the Willoughby.”
It was late afternoon and just as she was returning from her task of
seeking work in connection with the stage that they met. There he
was in Althea’s room, tall, spare, angular, slightly sallow and
cloisterish, his heavy eyebrows low above his sunken eyes as
though he sought to shut himself in to himself, and with those large
dark eyes fixed ruminatively and yet somewhat uncertainly upon all,
even her when she came. And from Althea she gathered that he was
a painter of strange dark landscapes and decorations which many of
those who knew seemed to think were wonderful but which as yet
had achieved no recognition at all. Worse, he was from the Rockies,
a sheep-rancher’s son, but had not been able to endure ranching.
His future was still very far before him, and, as one could sense, he
was so innocent of any desire to be put forward; he seemed half the
time to be a—dream. By some strange freak of luck he was still there
when she entered, sitting in a corner not entirely at ease, because,
as he told her later, he was strange to such affairs and did not know
when to go.
The brightness of the buildings in the spring sun!
And she had looked at his hands, at his commonplace clothes,
and then, a little troubled by his gaze, had withdrawn hers. Again
and again her eyes sought his or his hers, as though they were
furtively surveying each other; as though each was unable to keep
his eyes off the other. And by degrees there was set up in her a
tremendous something that was like music and fear combined, as
though all at once she had awakened and comprehended. She was
no longer the complete master of herself, as she had always
imagined, but was now seized upon and possessed by this stranger!
In brief, here he was, her dream, and now she could do nothing save
gaze nervously and appealingly—for what? Those dark, sombre
eyes, the coarse black hair and sallow skin! Yes, it was he indeed,
her love, her star, the one by whose mystic light she had been
steering her course these many years. She sensed it. Knew it. He
was here before her now as though saying: “Come.” And she could
only smile foolishly without speaking. Her hands trembled and her
throat tightened until she almost choked. “I never saw any one more
beautiful than you,” he had said afterwards when they talked, and
she had thrilled so that it was an effort not to cry out. And then he
had sighed like a child and said: “Talk to me, about anything—but
don’t go, will you?”
The air—the air—this day!
And so, realizing that he valued her for this one gift at least, her
beauty, she had sought now to make him understand that she was
his without, however, throwing herself beggingly before him. With her
eyes, her smile, her every gesture, she had said: “I am yours! I am
yours! Can’t you see?” At last, in his shy way, he had seemed to
comprehend, but even then, as he afterwards confessed, he could
not believe that anything so wonderful could follow so speedily upon
contact, that one could love, adore, at sight. She had asked where
he lived and if she might come and see his work, and with repressed
intensity he had said: “I wish you would! I wish you could come to-
day!” It had made her sad and yet laugh, too, for joy.
That single tree blooming in this long, hard block!
There and then, with only the necessary little interludes which
propriety seemed to demand, and with longing and seeking on the
part of each, had begun that wondrous thing, their love. Only it
seemed to have had no fixed beginning,—to have been always—just
been. For the day she had called him up his voice had so thrilled her
that she could scarcely speak. She had still felt she had known him
for so long. How could that have been?
“I was afraid you might not call,” he had said tremulously, and she
had replied: “And I was wondering if you really wanted me to.”
And when she sought him out in his studio she had found it to be
such a poor mean room over a stable, in a mean street among a
maze of mean streets, and yet had thought it heaven. It was so like
him, so bare and yet wonderful—a lovely spiritual mood set over
against tawdry materials and surroundings.
“Drive me through the East Side, Fred.”
Better, she had found him painting or perhaps merely pretending
to. He had on that old long gray linen duster which later became so
familiar a thing to her. And to one side of him and his easel on a
table were some of the colors of his palette, greens and purples and
browns and blues. He had said so softly as he opened the door to
her: “My painting is all bluffing to-day. I haven’t been able to think of
anything but you, how you might not come, how you would look—”
And then, without further introduction or explanation, under the north
light of his roof window filtering down dustily upon them, he had put
his arms about her and she her lips to his, and they had clung
together, thinking only of each other, their joy and their love. And he
had sighed, a tired sigh, or one of great relief after a strain, such a
strain as she herself had been under.
That one little cloud in the sky!
 And then after a time, he had shown her the picture he was
painting, a green lush sea-marsh with a ribbon of dark enamel-like
water laving the mucky strand, and overhead heavy, sombre, smoky
clouds, those of a sultry summer day over a marsh. And in the
distance, along the horizon, a fringe of trees showing as a filigree.
But what a mood! Now it hung in the Wakefield Gallery—and—
(Harry had helped to place it there for her!) But then he had said,
putting his brushes aside: “But what is the use of trying to paint now
that you are here?” And she had sighed for joy, so wonderful was it
all.
The crowds in these East Side streets!
Yet what had impressed her most was that he made no apology
for the bareness and cheapness of his surroundings. Outside were
swarming push carts and crowds, the babble of the great foreign
section, but it was all as though he did not hear. Over a rack at the
back of the large bare room he had hung a strip of faded burnt
orange silk and another of clear light green, which vivified what
otherwise would have been dusty and gray. Behind this, as she later
discovered, were his culinary and sleeping worlds.
And then, of course, had come other days.
But how like that first day was this one, so fresh and
bright!
There was no question here of what was right or wrong,
conventional or otherwise. This was love, and this her beloved. Had
she not sought him in the highways and the byways? At the close of
one afternoon, as she was insisting that she must continue her
search for work, now more than ever since neither he nor she had
anything, he had said sadly: “Don’t go. We need so little, Ulrica.
Don’t. I can’t stand it now.” And she had come back. “No,” she had
replied, “I won’t—I can’t—not any more, if you want me.”
And she had stayed.
And that wondrous, beautiful love-life! The only love-life she had
ever known.
 But just the same she had seen that she must redouble her efforts
to make her way, and had. Six hundred dollars she had brought to
the city was nearly all gone, and as for Vivian, his allotment was
what he could earn, a beggar’s dole. During the days that followed,
each bringing them closer, he had confessed more and more of the
difficulties that confronted him, how hard it was to sell his wares. And
she—it was needful for her to reopen the pages of her past. She had
not been happy or prosperous, she told him; fortune might have
been hers for the taking but she could not endure those who came
with it. Now that she had the misery of her soul’s ache removed she
must find something to do. The stage was her great opportunity. And
plainly his life was one which had always been and must be based
on the grudged dole that life offers to those who love its beauty and
lift their eyes. So few, as yet, knew of his work or had been arrested
by it. Yet if he persisted, as she felt,—if that wondrous something in
his work which had attracted the sensitive and selective did not fail—
The hot, bare redness of the walls of these streets, so
flowerless, so bleak, and yet so alive and human!
But all too well she understood that his life, unless changed by her,
would ever be the meagre thing it had been. Beauty was his, but no
more,—a beauty of mind and of dreams and of the streets and the
night and the sea and the movements of life itself, but of that which
was material he had nothing. That was for those whom she had
been unable to endure. Only by a deft synthesis of those wondrous
faculties which concern beauty was he able to perceive, respond to,
translate the things which he saw and felt, and these were not of
matter. Rather, they were epitomes, his pictures, of lands and skies
and seas and strange valleys of dreams, worlds in miniature. But
what transmutations and transferences! She was never weary of the
pictures he made. Nor was she ever weary of the picture he made
before his easel, tenuous and pale and concerned, his graceful
hands at work with the colors he synthetized. The patience, the
stability, the indifference to all but that which was his to do!
“Into Bartow Street, Fred.”
 And in him, too, was no impatience with life for anything it might
have failed to provide. Instead, he seemed ever to be thinking of its
beauties and harmonies, the wonder of its dawns and sunsets, the
colors and harmonies of its streets, buildings, crowds, silences.
Often of a morning when it was yet dark he would arise and open a
door that gave out onto a balcony and from there gaze upon the sky
and city. And at any time it was always an instinct with him to pause
before anything that appealed to either of them as beautiful or
interesting. And in his eye was never the estimating glint of one who
seeks to capture for profit that which is elemental and hence
evanescent, but only the gaze of the lover of beauty, the worshiper of
that which is profitable to the soul only.
The very street! The very studio!
Although she was ignorant of the spirit or the technique of art she
had been able to comprehend it and him, all that he represented as
a portion of beauty itself, the vast and supernal beauty toward which
the creative forces of life in their harsh and yet tender ways seem
impelled at times.
Had she not understood very well that it was as beauty that she
appealed to him, at first anyhow, an artistry of face and form plus a
certain mood of appreciation or adoration or understanding which
was of value to him? How often had he spoken of her lavender-
lidded eyes, the whiteness and roundness of her arms, the dark gold
of her hair, the sombre unrevealing blue of the iris of her eyes! Here
strange it was that these seemed to enthrall and hold him at times,
leaving him, if not weak, at least childlike in her hands. He had never
seemed to weary of her and during all their days together she could
feel his unreasoning joy in her.
His one-time yellow curtains exchanged for green ones!
That she had proved and remained irresistible to him was
evidenced by his welcoming and gratified eyes, the manner in which
he paused to survey her whenever she came near, seeming to re-
estimate her every least attribute with loving interest. Indeed, he
seemed to need her as much as she needed him, to yearn with an
intense hunger over her as a thing of beauty,—he who to her was
strength, beauty, ideals, power, all the substance of beauty and
delight that she could crave.
Yes, here was where they had come to gaze at the
towers of the bridge beyond.
And so for over a year it was that they clung together, seeking to
make of their lives an ideal thing. Only it was after she came into his
life that he had begun to worry—and because of what? It was no
hardship for him to live upon what he could make, but now that she
had come, with her beauty and her beauty’s needs, it was no longer
the same. As soon as she appeared he had seemed to sense his
inefficiency as a creator of means. Bowdler, the wealthy dealer, had
once told him that if he pleased him with something it might be worth
five hundred dollars. Five hundred dollars! But when he took a
painting to Bowdler he said he was overstocked, had too many of his
things on hand—the very things that to-day—(now that Vivian was
gone) —were selling for as much as ten and twelve thousand! And a
single one of all those now being sold would have made them both
happy for a whole year or more!
He had called this tree her parasol!
And she had been able to do so little for him! Realizing how little
life had done for him she had decided then and there that all her
efforts must be bent toward correcting this injustice. Life owed him
more. And so it was that at last she had turned to the stage and
sought earnestly, day after day and week after week, only to obtain
very little of all she needed to make them happy, a small part in one
of Wexford’s many productions, he of the comedies and farces and
beauty shows. Yet after some effort she had made him admit that
she was distinctive and that he could use her. But then had come
that long wait of nearly three months before the work began! And in
the meantime what labor, the night and day work of rehearsals and
appearances, the trying to get back to him each afternoon or night.
And he had been so patient and hopeful and helpful, waiting for her
after late hours of rehearsal to walk home with her and encouraging
her in every way. And yet always there was a tang of something
unreal about it all, hopes, as she so truly feared, that were never to
be realized, dreams too good to come true. The hours had flown.
The very pressure of his hand had suggested paradise, present and
yet not to be.
She must be returning now. It was not wise for her to sit
here alone.
And while those three months were dragging their slow days she
had borrowed what money she could to keep them going. She had
even borrowed from her mother! Yet they had been happy,
wandering here and there, he always rejoicing in the success which
her work promised to bring her. The studios facing the great park
where she now lived at which they had looked, he seeming to think
they were not for such as he. (The creatures who really dwelt there!)
Yes, she must be going. His train was due at four.
And then at last, her trial period was over, Wexford had
complimented her and her salary had been increased. She had
begun to buy things for Vivian and his studio, much as he protested.
But best of all for her the hope of better days still to come, greater
fame for herself and so better days for Vivian, a real future in which
he was to share—money,—comfort for them both.
“To the apartment, Fred.”
And then—In spite of all her wishes and fears, had come the
necessity for her to go on the road with the show. And owing to their
limited means he was compelled to remain behind. Worse, despite
the fact that each knew that every thought was for the other, the
thought of separation tortured them both. Wherever she was there
was the thought of him, alone, at his easel and brooding. And herself
alone. It had seemed at times as though she must die unless this
separation could be ended.
If only Harry were not coming into her life again!
But it was not ended—for weeks. And then one day, after a brief
silence had come the word that he had been ill. A wave of influenza
was sweeping the city and had seized him. She was not to worry.
But she did worry—and returned immediately, only to find him far
along the path which he was never to retrace. He was so ill. And
worse, a strange despondency based on the thought that he was
never to get well, had seized him. He had felt when she left, or so he
said, that something were sure to happen. They might not ever,
really, be together again. It had been so hard for him to do without
her.
He had added that he was sorry to be so poor a fighter, to bring
her back from her work. Her work! And he ill!
The immense wall these hotels made along the park!
And then against the utmost protest of her soul had come the end,
a conclusion so sudden and unexpected that it had driven despair
like metal into her very soul. Hour after hour and under her very
eyes, her protesting if not restraining hands and thoughts, he had
grown weaker. Though he knew, he seemed to wish to deny it, until
at last his big dear eyes fixed upon her, he had gone, looking as
though he wished to say something.
This wretchedly wealthy West Side!
It was that look, the seeking in it, that wishing to remain with her
that was written there, that had haunted her and did still. It was as
though he had wished to say: “I do not want to go! I do not want to
go!”
And then, half-dead, she had flung herself upon him. With her
hands she had tried to draw him back, until she was led away. For
days she was too ill to know, and only his grave—chosen by
strangers!—had brought it all back. And then the long days! Never
again would life be the same. For the first time in her life she had
been happy. A bowl of joy had been placed in her eager fingers, only
to be dashed from them. Yes, once more now she was alone and
would remain so, thrust back upon herself. And worse, with the
agonizing knowledge of what beauty might be. Life had lost its lustre.
What matter if others told of her beauty, if one or many sought to
make her life less bare?
This stodgy porter always at the door in his showy braid!
Why might not such as he die instead?
 But then her mother and sister, learning of her despair, had come
to her. Only since there was nothing that any pleasure, or aspect of
life could offer her, the days rolled drearily,—meaninglessly. And only
because of what was still missing in her mother’s life, material
comfort, had she changed. It had been with the thought of helping
her mother that after a year she had returned to the city and the
stage, but exhausted, moping, a dreary wanderer amid old and
broken dreams.
By degrees of course she had managed to pick up the threads of
her life again. Who did not? And now nature, cynical, contemptuous
of the dreams and longings which possess men, now lavished upon
her that which she and Vivian had longed for in vain. Fame? It was
hers. Money? A Score of fortunes had sought her in vain.
Friendship? She could scarcely drive it from the door. She was
successful.
But what mattered it now? Was it not a part of the routine, shabby
method of life to first disappoint one—sweat and agonize one—and
then lavish luxury upon one,—afterwards?
“I want nothing. And if any one calls, I am not in.”
And so it was that after a time Harry descending upon her with his
millions, and seeking solace for himself through her sympathy, she
had succumbed to that—or him—as a kindly thing to do. He too had
confessed to a wretched dole of difficulties that had dogged his early
years. He too had been disappointed in love, comfort—almost
everything until too late. In his earliest years he had risen at four in a
mill-town to milk cows and deliver milk, only later to betake himself,
barefooted and in the snow, to a mill to work. Later still he had
worked in a jewelry factory, until his lungs had failed. And had then
taken to the open road as a peripatetic photographer of street
children in order to recover his health. But because of this work—the
chemistry, and physics of photography—he had interested himself in
chemistry and physics—later taking a “regular job,” as he phrased it,
in a photographic supply house and later still opening a store of his
own. It was here that he had met Kesselbloom, who had solved the
mystery of the revolving shutter and the selenium bath. Financing
him and his patents, he had been able to rise still more, to fly really,
as though others were standing still. The vast Dagmar Optical and
Photographic Company. It was now his, with all its patents. And the
Baker-Wile Chemical Company. Yes, now he was a multimillionaire,
and lonely—as lonely as she was. Strange that he and she should
have met.
“No, I will not see any one.”
So now, through her, he was seeking the youth which could be his
no more. Because of some strange sense of comradeship in misery,
perhaps, they had agreed to share each other’s unhappiness!
“You say Mr. Harris telephoned from the station?”
Yes, as he had told her in his brooding hours, at fifty it had
suddenly struck him that his plethora of wealth was pointless. As a
boy he had not learned to play, and now it was too late. Already he
was old and lonely. Where lay his youth or any happiness?
And so now—nearly icy-cold the two of them, and contemning life
dreams—they were still facing life together. And here he was this
day, at her door or soon would be, fresh from financial labors in one
city and another. And returning to what? With a kind of slavish and
yet royal persistence he still pursued her—to comfort—as well as to
be comforted, and out of sheer weariness she endured him. Perhaps
because he was willing to await her mood, to accept the least crumb
of her favor as priceless. The only kinship that existed between them
was this unhappy youth of his and her sympathy for it, and his
seeming understanding of and sympathy for the ills that had beset
her. Supposing (so his argument had run) that the burden of this
proposed friendship with him were to be made very light, the lightest
of all burdens, that upon closer contact he proved not so hopeless or
dull as he appeared, could she not—would she not—endure him?
(The amazing contrarieties and strangenesses of things!) And so
friendship, and later marriage under these strange conditions. Yet
she could not love him, never had and never would. However it
might have seemed at first—and she did sympathize with and
appreciate him—still only because of her mother and sister and the
fact that she herself needed some one to fall back upon, a support in
this dull round of living, had caused her to go on as long as she had.
How deserted that wading-pool looked at evening, with
all the children gone!
And now at this very moment he was below stairs waiting for her,
waiting to learn whether she had smiled or her mood had relaxed so
that he might come up to plead afresh for so little as she could give
—her worthless disinterested company somewhere!
Well, perhaps it was unfair to serve one so who wished nothing
more than to be kind and who had striven in every way for several
years now to make himself useful if not agreeable to her, and yet—
True, she had accepted of his largess, not only for herself but for her
mother; but had she not had things of her own before that? And had
she not been content? Was it charity from her or from him?
And still—
Those darkening shadows in the sky in the east!
And yet it was always “Ulrica” here and “Ulrica” there. Did she so
much as refer to an old-time longing was it not he who attempted to
make amends in some way or to bring about a belated fulfilment?
Vivian’s painting now in the museum, the talk as to his worth, his
monument but now being erected—to whom, to whom were those
things due—this belated honoring of her darling—?
“Oh, well, tell him to come up. And you may lay out my
green evening dress, Olga.”
 XII
THE VICTOR

Some excerpts from an article on the late J. H.

Osterman, by C. A. Gridley, Chief Engineer of the
Osterman Development Company. This article appeared
in the Engineering Record, for August last.

“M Y admiration for the late J. H. Osterman was based on his

force and courage and initiative, rather than upon his large
fortune and the speed with which he had accumulated it after he had
passed the age of forty. Mr. Osterman was not always a pleasant
person to be near. Not that he was given to violent rages, but in the
prosecution of his various enterprises he had the faculty of giving
one the impression that but a fraction of his thoughts was being
revealed and that he was sitting apart and in judgment upon one, as
it were, even while he talked. He had the habit of extracting the most
carefully thought-out opinions of all those about him, and when all
had been said of shaking his head and dismissing the whole matter
as negligible, only to make use of the advice in some form later. At
such times he was apt to convince himself, and quite innocently, I
am sure, that his final opinion was his own.
“In so far as I could judge from hearsay and active contact with
him for a period of something like fourteen years, Mr. Osterman was
one who required little if any rest and at all times much work to keep
him content. His was an intense and always dominant personality.
Even after he had passed the age of sixty-five, when most men of
means are content to rest and let others assume the strenuous
burdens of the world, he was always thinking of some new thing to
do. It was only the week before he died, stricken while walking upon
his verandah, that he was in my office with a plan to subsidize the
reigning authorities of a certain minor Asiatic state, in order that
certain oil and other properties there might be developed under
peaceful conditions. A part of this plan contemplated a local army to
be organized and equipped and maintained at his expense. Of a
related nature was his plan for the double-screw platform descents
and exits for the proposed New York-New Jersey traffic tunnel, which
he appears to have worked out during the spring which preceded his
sudden demise and plans for which he was most anxious to have
this department prepare in order that they might be submitted to the
respective states. It is hardly needful to state, since the fact is
generally known, that those plans have been accepted. Of a related
nature were those Argentine-Chilean Trans-Andean railway projects
so much discussed in the technical engineering as well as the trade
papers of a few years since, and which recently have been jointly
financed by the two governments. Only the natural tact and
diplomacy of a man like Mr. Osterman, combined with his absolute
genius for detecting and organizing the natural though oftentimes
difficult resources of a country, would have been capable of making
anything out of that very knotty problem. It was too much identified
with diplomacy and the respective ambitions and prejudices of the
countries involved. Yet it was solved and he succeeded in winning
for his South American organization the confidence and friendship of
the two governments.”

II

The facts concerning the founding and development of the fortune

of the late J. H. Osterman, as developed by C. B. Cummings,
quondam secretary to Mr. Osterman, special investigator for E. X.
Bush, of counsel for the minority stockholders of the C. C. and Q. L.,
in their suit to compel the resale of the road to the original holders
and the return of certain moneys alleged to have been illegally
abstracted by J. H. Osterman and Frank O. Parm, of Parm-Baggott
and Company, and by him set forth in his reminiscences of Mr. Bush
and the Osterman-Parm-C. C. & Q. L. imbroglio.
 1. The details of the Osterman-De Malquit matter were, as near as
I have been able to gather, or recall, since I was Mr. Osterman’s
secretary until that time, as follows: De Malquit was one of the many
curb brokers in New York dabbling in rubber and other things at the
time Osterman returned from Honduras and executed his very
dubious coup. The afternoon before De Malquit killed himself—and
this fact was long held against Mr. Osterman in connection with his
sudden rise—he had come to Osterman’s office in Broad Street, and
there, amid rosewood and mahogany and an unnecessary show of
luxury which Osterman appeared to relish even at that time, had
pleaded for time in which to meet a demand for one hundred
thousand dollars due for ten thousand shares of Calamita Rubber,
which Osterman then entirely controlled and for which he was
demanding the par or face value. And this in spite of the fact that it
had been selling on “curb” only the day before for seven and one-
fourth and seven and three-fourths. De Malquit was one of those
curb brokers whom Osterman, upon coming to New York and
launching Calamita (which was built on nothing more solid than air),
had deliberately plotted to trap in this way. Unwitting of Osterman’s
scheme, he had sold ten thousand shares of Calamita on margin at
the above low price without troubling to have the same in his safe, as
Osterman well knew. That was what Osterman had been counting on
and it had pleased him to see De Malquit, along with many others
just at that moment, in the very same difficult position. For up to that
day Calamita, like many another of its kind, had been a wildcat
stock. Only “wash sales” were traded in by brokers in order to entrap
the unwary from the outside. They traded in it without ever buying
any of it. It had to fluctuate so that the outsider might be induced to
buy, and that was why it was traded in. But when it did fluctuate and
the lamb approached, he was sold any quantity he wished, the same
being entered upon their books as having been sold or bought on his
order. When a quotation sufficiently low to wipe out the margin
exacted had been engineered among friends, the lamb was notified
that he must either post more money to cover the decline or retire. In
quite all cases the lamb retired, leaving the broker with a neat profit.
This was the very situation upon which Osterman had been
counting to net him the fortune which it eventually did, and overnight
at that. Unknown to the brokers, he had long employed agents
whose business it was to permit themselves to be fleeced for small
sums in order that these several brokers, growing more and more
careless and finding this stock to be easy and a money-maker,
should sell enough of it without actually having it in their safes to
permit him to pounce upon them unexpectedly and make them pay
up. And so they did. The promoters of the stock seeming indifferent
or unable to manage their affairs, these fake sales became larger
and larger, a thousand and finally a ten thousand share margin sale
being not uncommon. When the stage was set the trap was sprung.
Overnight, as it were, all those who at Osterman’s order had bought
the stock on margin (and by then some hundreds of thousands of
dollars’ worth had been disposed of) decided that they would not
lose their margins but would follow up their cash with more and take
the stock itself, holding it as an investment. It then became the duty
of these brokers to deliver the stock within twenty-four hours or take
the consequences—say, a petition in bankruptcy or a term in jail.
Naturally there was a scrambling about to find any loose blocks of
the stock. But these had been carefully garnered into the safe of Mr.
Osterman, who was the sole owner of the stock, and they were
compelled to hurry eventually to him. Here they were met by the
genial eye of the cat that is expecting the mouse. They wanted
Calamita, did they? Well, they could have it, all they wanted ... at par
or a little more. Did that seem harsh, seeing that it had been selling
only the day before for seven and three-fourths or seven-eighths?
Sorry. That was the best he could do. They could take it at that price
or leave it.
Naturally there was a panic among those who were short. The
trick was obvious, but so was the law. Those who could, pocketed
their losses without undue complaint and departed; some who could
not, and were financially unimportant, decamped, leaving
Osterman’s agents to collect as best they might. Only one, Mr. De
Malquit, finding himself faced by complications which he could not
meet, took his own life. He had unfortunately let himself in for much
more money than had the others and at a time when he could least
stand the strain.
 The day De Malquit came to see Osterman, he was behind his
desk expecting many, and because, as I afterwards learned from Mr.
Osterman himself, Mr. De Malquit had been so wary, making
agreements at first, which made it hard to trap him, Osterman saw
him as one of those who had made it most difficult for him to win,
and therefore deserving to be sheared the closest. “One hundred per
share, take it or leave it,” was his only comment in reply to Mr. De
Malquit’s statement that he found himself in a bit of a hole and would
like to explain how a little time would see him through.
“But, Mr. Osterman,” I recall De Malquit replying, “I haven’t so
much now, and I can’t get it. These shares were being quoted at
seven and three-fourths only yesterday. Can’t you let me off easier
than that, or give me a few months in which to pay? If I could have
six months or a year—I have some other matters that are pressing
me even more than this. They will have to come first, but I might pull
through if I had a little time.”
“One hundred, on the nail. That’s the best I can do,” Osterman
replied, for I was in the room at the time. And then signaled me to
open the door for him.
But at that Mr. De Malquit turned and bent on him a very troubled
look, which, however, did not move Mr. Osterman any. “Mr.
Osterman,” he said, “I am not here to waste either your time or mine.
I am in a corner, and I am desperate. Unless you can let me have
some of this stock at a reasonable price I am done for. That will bring
too much trouble to those who are near and dear to me for me to
care to live any longer. I am too old to begin over again. Let me have
some of it now. To-morrow will be too late. Perhaps it won’t make
any difference to you, but I won’t be here to pay anybody. I have a
wife who has been an invalid for two years. I have a young son and
daughter in school. Unless I can go on—” He turned, paused,
swallowed, and then moistened his lips.
But Mr. Osterman was not inclined to believe any broker or to be
worked by sentiment. “Sorry. One hundred is the best I can do.”
At that De Malquit struck his hands together, a resounding smack,
and then went out, turning upon Osterman a last despairing glance.
That same night De Malquit killed himself, a thing which Osterman
had assumed he would not do—or so he said, and I resigned. The
man had really been in earnest. And, to make matters worse, only
three months later De Malquit’s wife killed herself, taking poison in
the small apartment to which she had been forced to remove once
the breadwinner of the family was gone. According to the pictures
and descriptions published in the newspapers at the time she was,
as De Malquit had said, an invalid, practically bedridden. Also,
according to the newspapers, De Malquit had in more successful
days been charitably inclined, having contributed liberally to the
support of an orphan asylum, the Gratiot Home for Orphans, the
exterior appearance of which Osterman was familiar with. This fact
was published in all of the papers and was said to have impressed
Osterman, who was said always to have had a friendly leaning
toward orphans. I have since heard that only his very sudden death
three years ago prevented his signing a will which contained a
proviso leaving the bulk of his great fortune to a holding company
instructed to look after orphans. Whether this is truth or romance I do
not know.
2. The case of Henry Greasadick, another of Mr. Osterman’s
competitors, was similar. Mr. Greasadick has been described to me
as a very coarse and rough man, without any education of any kind,
but one who understood oil prospecting and refining, and who was
finally, though rather unfortunately for himself, the cause of the
development by Osterman of the immensely valuable Arroya Verde
field. It is not likely that Greasadick would ever have made the
fortune from this field that Osterman and his confrères were destined
to reap. However, it is equally true that he was most shabbily treated
in the matter, far more so than was De Malquit in regard to his very
questionable holdings and sales. The details of the Arroya Verde
field and Greasadick are as follows: Greasadick has been described
to me as a big, blustery, dusty soul, uncouth in manners and speech,
but one who was a sound and able prospector. And Osterman, it
appears, having laid the foundation of his fortune by treating De
Malquit and others as he had, had come west, first to the lumber
properties of Washington and Oregon, where he bought immense
tracts; and later, to the oil lands of California and Mexico, in which
state and country he acquired very important and eventually (under
him) productive holdings. Now it chanced that in his wanderings
through southern California and Arizona he came across
Greasadick, who had recently chanced upon a virgin oil field which,
although having very little capital himself, he was secretly attempting
to develop. In fact, Greasadick had no money when he discovered
this oil field and was borrowing from L. T. Drewberry, of the K. B. &
B., and one or two others on the strength of his prospects. It also
appears that Drewberry it was who first called the attention of
Osterman to Greasadick and his find and later plotted with him to
oust Greasadick. Osterman was at that time one of three or four men
who were interested in developing the K. B. & B. into a paying
property by extending it into Arizona.
At any rate, Greasadick’s holdings were one hundred miles from
any main line road, and there was very little water, only a thin trickle
that came down through a cut. True, the K. B. & B. was about to
build a spur to Larston in order to aid him, but Larston, once the line
was built to it, was fourteen miles away and left Greasadick with the
problem of piping or hauling his oil to that point. Once he heard of it
Osterman saw at a glance that by a little deft manœuvring it could be
made very difficult for Greasadick to do anything with his property
except sell, and this manœuvring he proceeded to do. By buying the
land above Greasadick’s, which was a mountain slope, and then
because of a thin wall of clay and shale dividing the Arroya Verde, in
which lay Greasadick’s land, from the Arroya Blanco, which was
unwatered and worthless, being able to knock the same through, he
was able to divert the little water upon which Greasadick then
depended to do his work. Only it was all disguised as a landslide—
an act of God—and a very expensive one for Greasadick to remedy.
As for the proposed spur to Larston—well, that was easy to delay
indefinitely. There was Drewberry, principal stockholder of the K. B. &
B., who joined with Osterman in this adroit scheme. Finally, there
was the simple device of buying in the mortgage given by
Greasadick to Drewberry and others and waiting until such time as
he was hard-pressed to force him to sell out. This was done through
Whitley, Osterman’s efficient assistant, who in turn employed another