Jury Assignment

Data Analytics & R

SUBMITTED BY
DEPARTMENT OF FASHION TECHNOLOGY 2018-22
VAISHALI KCANNAN (BFT/18/398)
NAYONIKA DAS (BFT/18/574)
01
INTRODUCTION
 Introduction
R is a language and environment for statistical computing
and graphics. It is a GNU project which is similar to the S
language and environment which was developed at Bell.
It includes:
• An effective data handling and storage facility,
• A suite of operators for calculations on arrays, in

WELCOME •
particular matrices
A large, coherent, integrated collection of intermediate
MESSAGE tools for data analysis
• Graphical facilities for data analysis and display either on
screen or on hardcopy
• A well-developed, simple and effective programming
language which includes conditionals, loops, user-defined
recursive functions and input and output facilities.
Steps of Data Analysis with R

Data analysis with R is done in a series of steps;

programming, transforming, discovering, modeling
and communicate the results.

• Program: R is a clear and accessible programming

tool

• Transform: R is made up of a collection of

libraries designed specifically for data science.

• Discover: Investigate the data, refine your

hypothesis and analyze them

• Model: R provides a wide array of tools to

capture the right model for your data.

• Communicate: Integrate codes, graphs, and

outputs to a report with R Markdown or build
Shiny apps to share with the world.
 Importance of R
1. R is a well-developed, simple and effective
programming language. Which includes
conditional loops; user defined recursive
functions and input and output facilities.
2. R provides graphical facilities for data analysis
and display.
3. R is a very flexible language. It does not
necessitate that everything should be done in R
itself. It allows the use of other tools, like C and
C++ if required.
4. R has an effective data handling and storage
facility.
5. R provides an extensive, coherent and integrated
collection of tools for data analysis.
6. R also includes a package system that allows the
users to add their individual functionality in a
manner that is indistinguishable from the core of
R.
7. R is actively used for statistical computing and
design. It has brought about revolutionary
improvements in big data and data analytics.
02
Advantages &
Disadvantages
of R
 Advantages of R

1. Open Source
2. Exemplary Support for Data Wrangling
3. The Array of Packages
4. Quality Plotting and Graphing
5. Highly Compatible
6. Platform Independent
7. Eye-Catching Reports
8. Machine Learning Operations
9. Statistics
10. Continuously Growing
 Disadvantages of R

1. Weak Origin
2. Data Handling
3. Basic Security
4. Complicated Language
5. Lesser Speed
 Real Life Applications of R
• Facebook – Facebook uses R to update status and its social network graph. It
is also used for predicting colleague interactions with R.
• Ford Motor Company – Ford relies on Hadoop. It also relies on R for
statistical analysis as well as carrying out data-driven support for decision
making.
• Google – Google uses R to calculate ROI on advertising campaigns and to
predict economic activity and also to improve the efficiency of online
advertising.

• Microsoft – Microsoft uses R for the Xbox matchmaking service and also as a
statistical engine within the Azure ML framework.
• Mozilla – It is the foundation behind the Firefox web browser and uses R to
visualize web activity
• National Weather Service – The National Weather Service uses R at its River
Forecast Centers. Thus, it is used to generate graphics for flood forecasting.
03
R I n Fa s h i o n S t u d i e s &
C h a l l e n g e s Fa c e d
 R in Fashion Studies

• Fashion industry is an extremely competitive and dynamic market. Trends and styles change with
the blink of an eye. A collection or trend takes thousands of man-hours from the most creative
minds and the lucrative outcome depends on the simple judgements from fashion pundits,
bloggers and celebrities.

• Data Science can be used here on historical data to predict the trends which will be potentially
saving a lot of time and money. Data Scientists can use concepts from predictive algorithms,
visual search, capturing structured data from photographs, natural language processing and many
more. The importance of data has been gradually acknowledged by fashion professionals to
improve sales and margins because fashion brands and retailers need to develop, manufacture,
and sell styles that resonate with consumers
 R in Fashion Studies

The unique challenges that fashion retailers face: -

• While digital transformation has affected every retailer, fashion retailers are amongst the hardest
hit verticals facing specific challenges which complicate the use of data analytics and demand
forecasting. This is because fashion retailers work with dynamic assortments and products with
short lifecycles.
• Fashion and apparel retailers also deal with additional complexity in distribution of sizes, colors,
and styles — all of which is complicated further by shifting seasonality. This often results in
inventory distortion that leads to lost sales and costly markdowns.
• The nature of fashion itself makes selling online a more difficult endeavor. The variations in fit and
color between brands and styles causes an increase in product returns. The cost of returns alone
can make a retailer question the profitability of online stores.
04
PREDICTIVE AND PRESCRIPTIVE
ANALYTICS FOR FASHION
 Predictive and Prescriptive Analytics for Fashion

 
1. Predictive Analytics
Predictive analytics in fashion can be most easily summed up by the
example of demand forecasting. This is an AI driven forecasting process
that automatically accounts for dozens of factors (previous sales, price
elasticity, product cannibalization, events, etc.) to very accurately predict
the demand of each SKU at each store. This allows retailers to cut costs on
inventory (by only stocking products with demand) and maximize sales (by
slashing out-of-stocks on inventory that typically sells out).

2. Prescriptive Analytics

Prescriptive analytics, on the other hand, takes all of the predictions made by
predictive analytics, runs them through algorithms and AI built for fashion
retail, and automatically generates recommendations for retailers. For
example, a prescriptive analytics platform may predict when to preemptively
transfer inventory from one store to another to reduce markdowns and
increase sales, and automatically generate an inventory transfer request .
05
RECENT ADVANCEMENT IN FASHION STUDIES
USING R PROGRAMMING
 1. Behavioral and Contextual Targeting

• Data analytics can change a top-down segmentation approach to a bottom-up segmentation in the
fashion industry. Gruzbarg proposed the four dimensions to make “behavioral and contextual
targeting” possible by using data analytics as follows: -
• Demographics like gender, age, or income to understand who the customer is.
• Historic/behavioral data including purchase transactions, cart abandoned, etc. to understand
what the customer did in the past and can be used to predict their future behavior.
• Streaming/contextual data such as a customer’s current digital behaviors, web pages they are
viewing, emails they just open, ads they are clicking on to know in what product the customer is
interested and whether s/he is currently in shopping mode.
• Predictive analytics using AI-powered software to predict how the customer is likely to respond in
the future by identifying invisible patterns in the previous three dimensions.

Using a data-driven segmentation approach will help fashion brands and retailers develop more
personalized products and marketing strategies.
 2. The Battle of Sexes
A key point for all designers is the choice between selling women’s or men’s clothing, whether it’s accessories, high
couture or everyday clothes. Every designer has a different target demographic so understanding who is more likely to
purchase their products is imperative for the success of the business. Big data clarifies how and when consumers
make their purchases, how big those purchases typically are and how many items are bought. This data can help in
deciding which lines are more relevant for the business success and choosing between women’s and men’s wear could
be a deciding factor.

3. Streamlining Product Lifecycle

Fashion brands can measure customer reaction to samples and ideas, then adjust their product accordingly. Big
data helps brands manage the whole feedback and iteration process with ease.

The manufacturing process itself is also becoming more streamlined and data-oriented. IT infrastructure,
operation and production systems can now communicate with each other through the Internet of Things (IoT).

In an intelligent factory, all the separate machines and systems required to create a product are connected with
each other, rather than being connected via a central controlling unit. Nowadays, production components can
communicate issues such as low or faulty stock directly to loading systems, which can then react autonomously
and fix the issue.
 4. Favourite Colour Trends

Current global trends do drive the fashion industry but what fashion companies also need to know is which
colours are favoured by their customers. This is where big data steps in revealing the most popular colour
trends among consumers so the design can easily be changed and adapted according to the demand.

Big data shows the range of colours that were selling best so out of that, the company can choose the
spectrum that is the most popular. The use of big data is simply – it’s comparing colours sold by a particular
business and guides the designer and the production by keeping in mind the desires of customers.

5. AI- Based Merchandising Applications

AI-based merchandising applications helps to create executable solutions for how to ensure maximum sales
by optimizing product assortment, efficient allocation and a great customer experience.
 6. Transforming Runway Fashion Into Retail Merchandise

In reality the fashion that we often see on runways is rarely donned by the ordinary customers; because, the
dresses and outfits that are showcased on the ramp are a bit OTT, thus altered before being placed in the
stores. So, big data aids in deciphering which attires are going to be successful, and which will fail down the
line. So, use the power of big data prudently and reap benefit, unimaginable across the global retail stores.

7. Capitalizing On A Continuous Feedback Loop

Customer preferences are sent to clothing designers, while machine learning analyses the written feedback
that customers leave after receiving their clothes.

Designers have saved a lot of money, time and effort the company and its by using the underlying data they
collect to forecast trends based on customer preferences, rather than making the products and sending them
out to retailers only to have them lose money.
06
G E O S PAT I A L A N A LY T I C S I N
O M N I C H A N N E L R E TA I L
 Technologies in Virtual Reality
 
1. Challenge
Unfortunately, retailers often make the wrong decisions about which stores to close, thus inadvertently hurting
their business further. They also overlook valuable opportunities to expand their market presence and unlock
growth. The main reason is that they’re using outdated metrics: many retailers continue to use a combination of
trend analysis and “four-wall economics” to assess store performance—that is, they’re still primarily taking into
account the sales and profits that the store generates within its four walls, without considering its impact on
other channels.
2. Solution
They’re taking an omnichannel view of store performance—allowing each store to “get credit” for all the sales in
which it played a role, whether those sales happened offline or online. In doing so, retailers are getting a more
accurate picture of each store’s total economic value and making better decisions about their omnichannel
presence.
3. Outside the four walls
Advances in data and analytics can help a retailer quantify both a store’s halo effect (positive) and its
cannibalization effect (negative)—in other words, how a store’s existence influences the performance of the
retailer’s other sales channels.
07
D ATA S C I E N C E I N R E TA I L
 Data Science In Retail

1.RECOMMENDATION ENGINES
2. FRAUD DETECTION
3. PERSONALIZED MARKETING
4. PRICE OPTIMIZATION
5. INTELLIGENT CROSS-SELLING AND UPSELLING
6. INVENTORY MANAGEMENT
7. CUSTOMER SENTIMENT ANALYSIS
8. FORETELLING TRENDS THROUGH SOCIAL MEDIA
9. WIFI SIGNALS
10. CUSTOMER LIFETIME VALUE PREDICTION
08
F U T U R E O F FA S H I O N W I T H R
 BIG DATA TO SOLVE THE COUNTERFEIT MERCHANDISE

‘Replica fashionwear’ or ‘Pirated merchandise’ is one of the biggest problems that has plagued the
fashion industry. The huge numbers have garnered the attention of industry leaders who are now
using big data to solve the problem of counterfeit merchandise. Companies are now using pattern
recognition coupled with big data to protect the integrity of their company. Cognitive Prints, a suite of
AI tools can be used for pattern recognition where it will scan through huge amounts of data looking
for similarities. This in addition to fighting piracy helps in building a brand exclusivity and recognition
 DEALING COMPETITION WITH DATA
One of the biggest concerns of most fashion retailers is to understand their competitor's strategy and
to be ready to outshine them. With appropriate big data tools, fashion retailers can get real-time
insights on what their competitors are creating and how their campaign is performing. Armed with
these insights, they can strategize their campaigns better than their peers

DATA WILL UPGRADE THE PERFORMANCE OF CLOTHING

Smart Fashion is the new wave of tech development in the fashion industry and it is completely driven
by the consumer desire for data. CES 2016 was stacked with wearables that track health and fitness
performance, and numerous other factors. These products do everything from track body heat, record
heart rates, adjustable bra support, and charging via solar power. While some of these products aren’t
fully realized or fashionable yet, it’s only a matter of time before newer, more marketable iterations
become available.

BIG DATA TO HELP IDENTIFY INFLUENCERS

Predictive technology will be doing more than identify key colors, cuts, and styles. It will also help
brands identify influencers whom they can ally to prior to the latter’s ‘boom.’
09
A P P L I C AT I O N S O F D ATA
A N A LY T I C S & R I N Z A R A
 ZARA’S BIG DATA APPROACHES
Data is captured from POS terminals, e-commerce sales, customer surveys, PDA devices and RFID tags
on the clothing. Store personnel are trained to capture customer preferences in their PDAs on every
detail: buttons, zippers, color, cut and more. They enter and upload the feedback nightly and regional
managers isolate their region’s preferences and tastes.

ZARA’S BIG DATA SUPPLY CHAIN MANAGEMENT

(Big Data Help Reduce the Time Elapse between Order and Distribution)
10
A P P L I C AT I O N S O F D ATA
A N A LY T I C S & R I N E C H O L O O K
 ECHO LOOK
Echo Look features a depth-sensing camera, built-in LED lighting, and computer vision based
background blur. The device also comes with Style Check software. Echo Look can give customers
personalized recommendations, helps customers organize their closet and build personal lookbooks by
categories like season, weather, occasion, and more, and provides fashion inspirations. The Internet
connected camera is designed to take high-quality photos and videos when the user says, “Alexa, take
a picture” or “Alexa, take a video.”
11
A P P L I C AT I O N S O F D ATA
A N A LY T I C S & R I N S T Y L E S N A P
 STYLESNAP

StyleSnap can improve customer engagement by helping them get fashion inspiration in real time.
StyleSnap, which is available on the Amazon app, allows customers to upload a picture or screenshot
of a fashion look they like and get immediate recommendations of similar items available from
12
A P P L I C AT I O N S O F D ATA
A N A LY T I C S & R I N B U R B E R R Y
 BURBERRY
Burberry implemented data analytics with customer feedback to better understand customers and
offer personalized services through the omnichannel (app, website, email, and in stores), which helps
enrich customers’ experiences and builds a better connection between the company and their
customers.
CLIENTELING AND CUSTOMER 360
Burberry focused on a new customer-centric service, “Clienteling”. Clienteling can be defined as a
business practice that analyzes customer information and offers personalized and memorable service.
Burberry launched a data-driven program called “Customer 360” to help them achieve clienteling, to
offer exceptional customer service and increase sales. This program is designed to identify ideal
products and services that meet customer demands by synthesizing interactions with customers from
all the channels. This program invites customers to voluntarily digitally share their buying history,
shopping preferences, and experiences through a number of loyalty and rewards programs.
13
A P P L I C AT I O N S O F D ATA
A N A LY T I C S & R I N G A P
 GAP
Art Peck put forth tremendous efforts toward enabling customers to seamlessly shop omnichannels.
Omnichannel shopping helps Gap Inc. attract customers and gives the company more access to
customers’ data (Israeli & Avery, 2018). At the same time, data analytics can make contributions to
building a quick response system because of its great capability to extract useful information from
related data, which can provide a reliable basis for decision-making. In order to shorten the cycle from
design to sale and find the right trends, Peck started to push the company to utilize a data mining
approach to extract the market trend for the next season’s collections instead of relying solely on one
person’s vision.
13
A P P L I C AT I O N S O F D ATA
A N A LY T I C S & R A L P H L A U R E N
 RALPH LAUREN
WEARABLE TECHNOLOGY TO GET BETTER INSIGHTS
Wearable technology, which can be defined as the integration of various micro sensors into wearable
items for the purpose of collecting biometric data for monitoring and improving fitness and wellness.
These devices give wearers real-time insights into their wellness activities through wireless
communications. Ralph Lauren stepped into the wearable technologies market with a trial of the
PoloTech Shirt in conjunction with the 2014 US Open. Inside the PoloTech shirt are sensors attached to
silver threads that measure the wearer’s movement, as well as their heart and breathing rates,
number of steps taken and calories burned. With the accompanying app, users can monitor their data
and access custom workouts the app has designed specifically for them, based on their readings.
14
CONCLUSION
Conclusion

• Analyzing what is the next breakthrough in the fashion world is just what you need to make it big in the
fashion business.
• In order to stay relevant amidst growing competition, it is important for fashion designers to come up with
unique products.
• However, trends are hard to analyze using traditional monitoring techniques.
• By integrating big data with AI and ML, retailers can use live insights to create the most innovative fashion
trends.
• Big data will also help you analyze customer behavior to suggest a product price that will ignite demand
while locking a rise in overall profits.
• Fashion brands now will start using data science training to consistently create winning products. It’s
important to engage the loyal customer, create new customers, be a trendsetter and most importantly break
a brand perception by adopting big data and data science at the earliest.
Conclusion
 SUBMITTED BY

Nayonika Das (BFT/18/574)

Vaishali Kcannan (BFT/18/398)