Professional Documents
Culture Documents
Data Intelligence and Cognitive Informatics: Proceedings of ICDICI 2021 (Algorithms For Intelligent Systems) I. Jeena Jacob (Editor)
Data Intelligence and Cognitive Informatics: Proceedings of ICDICI 2021 (Algorithms For Intelligent Systems) I. Jeena Jacob (Editor)
https://ebookmeta.com/product/business-intelligence-6th-
international-conference-cbi-2021-beni-mellal-morocco-
may-27-29-2021-proceedings-1st-edition-mohamed-fakir/
https://ebookmeta.com/product/intelligent-tutoring-systems-18th-
international-conference-its-2022-bucharest-romania-
june-29-july-1-2022-proceedings-scott-crossley/
https://ebookmeta.com/product/data-science-and-intelligent-
systems-proceedings-of-5th-computational-methods-in-systems-and-
software-2021-vol-2-radek-silhavy/
https://ebookmeta.com/product/data-engineering-for-smart-systems-
proceedings-of-ssic-2021-priyadarsi-nanda/
https://ebookmeta.com/product/networking-intelligent-systems-and-
security-proceedings-of-niss-2021-smart-innovation-systems-and-
technologies-237/
https://ebookmeta.com/product/computational-logistics-12th-
international-conference-iccl-2021-enschede-the-netherlands-
september-27-29-2021-proceedings-martijn-mes-editor/
I. Jeena Jacob
Selvanayaki Kolandapalayam Shanmugam
Robert Bestak Editors
Data Intelligence
and Cognitive
Informatics
Proceedings of ICDICI 2021
Algorithms for Intelligent Systems
Series Editors
Jagdish Chand Bansal, Department of Mathematics, South Asian University,
New Delhi, Delhi, India
Kusum Deep, Department of Mathematics, Indian Institute of Technology Roorkee,
Roorkee, Uttarakhand, India
Atulya K. Nagar, School of Mathematics, Computer Science and Engineering,
Liverpool Hope University, Liverpool, UK
This book series publishes research on the analysis and development of algorithms
for intelligent systems with their applications to various real world problems. It
covers research related to autonomous agents, multi-agent systems, behavioral
modeling, reinforcement learning, game theory, mechanism design, machine
learning, meta-heuristic search, optimization, planning and scheduling, artificial
neural networks, evolutionary computation, swarm intelligence and other algo-
rithms for intelligent systems.
The book series includes recent advancements, modification and applications
of the artificial neural networks, evolutionary computation, swarm intelligence,
artificial immune systems, fuzzy system, autonomous and multi agent systems,
machine learning and other intelligent systems related areas. The material will be
beneficial for the graduate students, post-graduate students as well as the
researchers who want a broader view of advances in algorithms for intelligent
systems. The contents will also be useful to the researchers from other fields who
have no knowledge of the power of intelligent systems, e.g. the researchers in the
field of bioinformatics, biochemists, mechanical and chemical engineers,
economists, musicians and medical practitioners.
The series publishes monographs, edited volumes, advanced textbooks and
selected proceedings.
All books published in the series are submitted for consideration in Web of
Science.
Data Intelligence
and Cognitive Informatics
Proceedings of ICDICI 2021
Editors
I. Jeena Jacob Selvanayaki Kolandapalayam Shanmugam
Department of Computer Science Assistant Professor of Computer Science,
and Engineering Department of Mathematics and Computer
GITAM University Science
Bangalore, India Ashland University
Ashland, OH, USA
Robert Bestak
Department of Telecommunication
Engineering
Czech Technical University in Prague
Prague, Czech Republic
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Singapore Pte Ltd. 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
The ICDICI 2021 conference is solely
dedicated to all the editors, reviewers, and
authors of the conference event.
Foreword
vii
viii Foreword
authors and conference participants for contributing their novel research results to
the conference. Special thanks to Springer publications.
P. Ebby Darney
Conference Chair – ICDICI 2021
SCAD College of Engineering
and Technology
Tirunelveli, Tamil Nadu, India
Preface
With a deep satisfaction, I write this foreword to welcome you all to the International
Conference on Data Intelligence and Cognitive Informatics [ICDICI 2021] held in
Tirunelveli, India, on July 16–17, 2021.
The theme of ICDICI 2021 is Data Intelligence, a research topic that is gaining
quick research traction in both industries and academia due to its research relevance
to the emerging societal and economic issues in the areas like healthcare, transporta-
tion, industries, education, and so on. The well-established research track records
on intelligent data systems mandate the integration of artificial intelligence tech-
niques and processes make ICDICI an excellent venue for exploring the cognitive
foundations for emerging data systems.
With respect to the potential hard work of the ICDICI 2021 conference committee,
I would like to express my appreciation and gratitude to all the technical program
committee members, international and national advisory board members, and review
committee members, who have made this conference a successful and possible one.
Finally, I would like to extend my warm thanks to all the keynote speakers, session
chairs, and fellow researchers, who have willingly shared their research experience
and knowledge with all the readers of this extended conference proceedings.
I hope that this proceeding of ICDICI 2021 will further stimulate research in data
mining and intelligent systems and provide practitioners with advanced algorithms,
techniques, and tools for deployment. I feel honored and privileged to serve the
significant recent developments in the field of intelligent systems and data intelligence
to you through this exciting program.
ix
Acknowledgments
We are deeply obliged to all the contributors of this ICDICI 2021 conference and all
the reviewers for their informative, cogent, and timely reviews of papers submitted to
the conference, and also the SCAD College of Engineering and Technology staff and
international technical reviewer’s community for their professional and thoughtful
expertise to shape this conference event.
Overall thanks go to all the conference program and local organizing committee
members, who have gone out in their way to make this a successful event.
Finally, we offer our sincere thanks to conference chair, co-conference chair,
and organizing secretary for their continuous effort in the organization, preparation,
and handling of ICDICI conference administration. Further appreciation is also due
to the editors of Springer publications, high standards of editorial productions of
conference proceedings.
xi
Contents
xiii
xiv Contents
Robert Bestak received the Ph.D. degree in Computer Science from ENST Paris,
France (2003) and M.Sc. degree in Telecommunications from Czech Technical
University in Prague, CTU, Czech Republic (1999). Since 2004, he has been Assis-
tant Professor at the Department of Telecommunication Engineering, Faculty of
Electrical Engineering, CTU. He took part in several national, EU, and third-party
research projects. He is Czech Representative in the IFIP TC6 organization and
Vice-Chair of working group TC6 WG6.8. He serves as Steering and Technical
Program Committee Member of many IEEE/IFIP conferences (Networking, WMNC,
NGMAST, etc.), and he is Member of the editorial board of several international jour-
nals (Computers & Electrical Engineering, Electronic Commerce Research Journal,
etc.). His research interests include 5G networks, spectrum management, and big
data in mobile networks.
xix
Chapter 1
NLP-Based Resume Screening and Job
Recruitment Portal
1 Introduction
Associations get a huge number of resumes for each job ad and utilize dedicated
screeners to narrow down the qualified applicants. However, resume screening is
considered as an important and time-consuming element of the decision-making
process. Different perspectives should be considered while screening resumes, for
instance, aptitudes of the contender, work understanding, assignment, and degree. To
the extent work is considered, picking the right competitor for the enrollment cycle
from a huge pool of candidates has been a focal issue. In the original employing
model, the associations would publicize their openings on papers and television. The
up-and-comers would send in their resumes by methods of post and their resumes
would be arranged physically. Once shortlisted, the recruiting group would call the
candidates for additional rounds of meetings. Clearly, this was a drawn-out proce-
dure. Nevertheless, the endeavors started growing hence did the employing needs.
Subsequently, the associations started reallocating their enrolling cycle. Enrolling
consultancies showed up. These associations required the candidates to transfer their
resumes on their sites specifically arranged. The workplaces would then experience
the coordinated data and shortlist competitors for the association. This cycle had a
huge detriment.
The approach discussed in the paper is that we have built up a product/framework
which can check the resumes and afterward sort them according to the organization’s
prerequisite. This framework is created utilizing the methods—Natural Language
Processing (NLP). Natural Language handling is a subfield of man-made reasoning.
It focuses on the factual examination of information to enable PC frameworks to
learn “independently” without being explicitly modified. It is the driver behind huge
numbers of the mechanical developments we find in the working environment today
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 1
I. J. Jacob et al. (eds.), Data Intelligence and Cognitive Informatics,
Algorithms for Intelligent Systems,
https://doi.org/10.1007/978-981-16-6460-1_1
2 R. Kadam et al.
just as a driver of the fourth modern insurgency. It also proposes that the framework
fuses all the highlights that ought to be actualized so as to make the recruiting
cycle snappy and proficient. Normal language handling will be utilized to investigate
different qualities or conditions. When the resumes are screened, it is coordinated
with the undisclosed range of abilities, and with the assistance of a calculation, a score
is assigned to each element as needed for figuring the estimations of the resume. At
that point, the positioning will be done dependent on scores determined.
Section 2 cursorily describes the existing systems that are available. After that,
Sect. 3 describes the entire system design that has been proposed to solve this
problem. Section 4 includes a wide overview on the implementation of this underway
project. Finally, Sect. 5 includes the conclusion and the future scope.
2 Literature Review
application is changed over into text design which is then taken care of into the
framework where only required information will be extracted to retrieve the best-
coordinating resume. Prior to presenting the resume, it is once shown to the candidate
to guarantee that the qualities submitted are substantial. They have utilized Machine
Learning for training the dataset and have used section-based segmentation for infor-
mation extraction utilizing Natural Language Processing (NLP). Their framework
utilizes semi-supervised learning to accomplish high precision. It helps the recruiters
to sift through the most forthcoming up-and-comer based on their resumes. They have
utilized the Spacy framework for NLP. Avoiding fraud with such a system is very
difficult as it is the job description with which the resume is compared and it is easy
for the candidate to bluff skills and get selected for the interview [3]. Ayishathahira
C. H. in “Combination of Neural Networks and Conditional Random Fields for effi-
cient resume parsing” states that their framework classifies a resume into 3 fragments
and extracts 23 fields. Their division model is made utilizing Convolutional Neural
Network (CNN) architecture. For data extraction, they utilized Conditional Random
Fields (CRFs) which are contrasted with the Bidirectional-LSTM-CNN model that
helps in named entity acknowledgment utilizing deep learning ideas, both are utilized
for sequence labeling. Hidden Markov Model (Well) and Support Vector Machine
(SVM) algorithms are utilized for data extraction. Their framework splits the resumes
into segments. It initially changed the resume into pdf, and then that pdf into HTML.
HTML labels were then taken out and sentence end algorithms were applied to mark
the finish of a sentence or passage. Inside Out Between End Start (IOBES) property
was utilized to recognize the right named entity for each word. 91% exactness was
accomplished with CNN and 43% precision was accomplished with BI-LSTM. They
are using a strict format of resumes which makes screening of resumes with different
formats, difficult [4].
Dena F. Mujtaba discusses “Ethical considerations in AI-based recruitment”
where they chipped away at reasonableness in AI and simulated intelligence-based
frameworks. Training data, Label definitions, Feature Selection, Proxies, Masking
were the reasons for inclination that they decided. For Moderating the issues that
were brought about by these reasons they did pre-processing where the information
was checked where the data was checked if it was biased or unbiased, if it was biased
then they made or used another unbiased dataset. At that point, In-Processing was
done, where they checked if the framework was working appropriately, and lastly,
they did post-processing where the output that was accomplished was matched with
the expected output. All through the cycle, it made a point to experience any incli-
nation that was occurring all the while. They used many tools like Lucid, Lime, Fair
ML, Fair test, etc. Working with biased and unbiased sets may not be always accurate,
it might give us uncertain results [5]. Ki-Young Shin in “A job applicant’s resume
verification method using a social network analysis” enhances the effectiveness of
a recruitment system by evaluating the feasibility of Facebook data. Netvizz was
utilized to gather network information and Gephi was utilized for the investigation
of information and they got 70.08% of the recall rate. Facebook information was
utilized to gain proficiency with the client’s experience and nature of a candidate
4 R. Kadam et al.
from daily life logs. They utilized 40 subjects for testing. Their research recom-
mends a technique to utilize Social Network Services (SNS) information to confirm
the resume of job candidates. Experience Sampling techniques were led to decide
how much the client’s life was reflected in Facebook data. They made use of social
media to learn the candidate’s personality and verify their achievements through
social media platforms like Facebook, LinkedIn, etc. In such a case, it will become
difficult to verify all candidates who have applied because not everybody has an
account on all social media platforms and many do not even update their profiles [6].
The Smart Job Recruitment Automation application helped to overcome the manual
hiring process of freshers or college students to give every student a chance to easily
find a job for themselves. Real environment experience was focused on to develop
this system. This system is especially made for one of the engineering institutes
of Nepal, Pulchowk Campus. They made use of Job posts and candidate resumes
for comparison and score calculation. However, it is possible for the candidate to
bluff skills by looking at the job post or job description in order to get selected.
Fake application can be a problem here [7]. Seghal says that Job Portal-A Web
Application for Geographically Distributed Multiple Clients is a portal that is able
to capture job requirements based on industry needs. It is a system that contains
knowledge and information relevant to the company and the candidate’s needs. It
is a knowledge-based system where the company is provided with the knowledge
related to their domain with respect to selection of desired candidates. Similarly, it
provides candidates with relevant knowledge. They have used search algorithms to
implement searching in the system. The system, however, does not provide a ranked
list of candidates to the company therefore the company has to make some efforts to
choose the desired candidates [8].
Alghamla states that Predicting the Suitability of IT Students’ Skills for the
Recruitment in Saudi Labor Market is a project that focuses on prediction and recom-
mendation of jobs to IT professionals in the kingdom. Data from the various online
job portals is extracted with the help of which machine learning algorithms are
created in order to develop this web-based application that predicts the suitability of
IT students’ skills for the recruitment process. This system also offers general recom-
mendations of the highest needed IT skills in KSA. There is a lack of research in this
field and therefore the system is not as accurate as it would be [9]. Burukayo talks
about Jobs256’ Mobile App Linking Job Seekers to Job Opportunities, which is an
android application that acts as a bridge between the applicants and the companies.
It was specially designed for significant use in places where there is limited network
connectivity. Their offline sync feature enables the users to access job opportunities
even when they are offline. The app crawls pre-programmed online job portals in
Uganda by using the crawl engine that reads the content of each site’s Really Simple
Syndication (RSS) file and then synthesizes them to get related information which is
aggregated, categorized, and delivered to the application users. Users who subscribe
to particular job are notified at a desired time through push notifications with the help
of Google Cloud Messaging Service. It is an android app and will be available android
users only, people who do have smartphones can miss this opportunity. The scope
of this project is with the boundaries of Uganda and not beyond it [10]. Zaroor talks
1 NLP-Based Resume Screening and Job Recruitment Portal 5
about A Job Post and Resume Classification system is a hybrid approach that exploits
an integrated knowledge base in order to run the classification task. JRC matches
resumes to the job descriptions that fall under the same or relevant categories. They
have used Automatic Integrated Knowledge-based Occupational Category Classifi-
cation of Resumes and Job Postings and employed a Section-based Segmentation
heuristic by exploiting Natural Language Processing (NLP), Concept-relatedness
techniques, and regular expressions. It generates classification rules from a set of
pre-classified job openings using the Dictionary of Occupational Titles (DOT). The
main disadvantage of this system is that DOT does not cover the occupational infor-
mation relevant to the modern workplace [11]. Soni explains that Natural Language
Processing for the Job Portal Enhancement is a system that defines the NLP process
that will be used for screening and matching purposes. It describes how the features
are extracted from the resumes using NLP and how matching is done using regular
expression and lemmatization in order to extract important features that should be
taken into consideration for ranking and score calculations. It is a basic procedure to
extract words and count them or apply a score calculation algorithm to it [12].
3 Proposed System
Our proposed system assists recruiters in selecting the right candidate from the pool
of candidates, therefore, reducing workload. Our system incorporates all the features
that should be implemented in order to make the hiring process quick and efficient.
A website will be created where the recruiters can interact with their applicants
by posting job openings for which the applicants can apply. The companies will
post their job descriptions which will all be visible to the applicant suitable for the
desired post. These companies will even submit an undisclosed skillset which will
be hidden from the user which will be used for scoring and ranking applicants. With
the help of a resume screening algorithm, the resume will be analyzed and necessary
features will be extracted like the candidate’s qualification, work experience, and
skills. Undisclosed skillset is used to match skills that a job offer expects from its
applicants and verifies if the candidate is suitable and then provides interview calls
to the fitting candidates in order to make the recruitment process fair. The applicant
would be displayed with all the job openings suitable to his profile, to which he
can apply if interested. The applicants will upload their resumes on their portal and
can apply to job openings. The evaluator can screen applications for the selected
job profile. Once the resumes are matched with the respective skill set, they are
scored and ranked accordingly. The companies will then receive the ranked list of
applicants and can choose any number of candidates. At the time of registration, the
HR responsible for posting job profiles is verified by the admin, and the candidates
registering on our website are given a declaration form which has to be checked
before completing the registration process. This declaration form specifies that, if
any false information is provided by the user, they will be blocked and will be denied
from accessing and applying to any of the jobs posted on our website. The company
6 R. Kadam et al.
can report to our website if any applicant is caught with false information in their
resumes at the time of interview (Fig. 1).
Our proposed website has three users which will run the website as per their
desired functionalities. The three users are as follows:
1. Admin: The Admin can verify the company’s users and can also block fraud
applicants reported by the company to avoid fraud.
2. Recruiter: The recruiter is the company responsible for posting the job profile,
undisclosed skillset, and who starts screening of applied resumes; upon which
he/she will get ranked results of applicants.
3. Applicant: The candidate who will post his/her resume and apply for interested
job openings suitable for his/her profile.
3.1 Admin
Admin will receive a list of company users who tried to register with our website
depending on which it will physically communicate with the company to get details
of their HR employers and then will validate company users to avoid fraud job
openings. Admin will also get a list of reported applicants from the company side
whom they found to display fraud information while physical evaluation, to which
the admin will evaluate physically to be sure and then finally block the applicant
if found false. These features will help our system to avoid fraud users as much as
possible (Fig. 2).
1 NLP-Based Resume Screening and Job Recruitment Portal 7
3.2 Recruiter
The recruiter will have to first register on the website by entering his/her email ID,
company name and set a password. This request will be displayed to the admin
according to which he/she will manually communicate with the company to check
the authenticity of the user and will ask the company to give credentials of the user to
authenticate the recruiter. These steps will help to avoid fraud recruiters and provide
genuine job openings for applicants.
Then, the recruiter will log into the website and view its dashboard from where it
can perform two operations:
1. Post a job opening by submitting expected salary, degree, minimum year of
experience, job description, and undisclosed skillset to the portal. These job
openings will then be visible to suitable applicants, to which they might apply.
These job openings will also be displayed to the recruiter so that he can have
track of the openings that he/she has posted.
2. Start screening process, where all the applicants interested for the job opening
will be screened and ranked. Each resume along with an undisclosed skill set
would be passed through feature extraction code to extract skills, education
and experience of each candidate and the one expected by the company. Which
8 R. Kadam et al.
are then matched and each resume is allotted a score, according to which all
applicants are ranked and displayed to the recruiter.
After screening, the recruiter gets to see ranked applicants with respect to their
score. The recruiter also gets to see contact details, degree, current salary, and year
of experience of the applicant depending on which he/she can further communicate
with applicants that he/she wants to (Fig. 3).
3.3 Applicant
The applicant or the candidate can register on the website successfully by entering
his email, degree, current salary, year of experience, and password. The applicant
can login by entering his/her credentials. The applicant will view his dashboard after
logging in; on dashboard applicants can view the suitable job openings posted by the
companies. Applicant can view openings if following conditions are satisfied:
1. Applicant’s degree matches with expected degree of job opening.
2. The current salary of the applicant should be less than the salary offered by the
opening.
3. Candidate is having experience more than or equal to what is expected by
opening.
The applicants can perform following operations:
1. Apply for interested job openings.
2. Upload his/her resume on the portal.
3. Update his/her profile, like updating his degree or current salary or years of
experience.
The applicant will be notified if he/she has been selected by the company side on
their email (Fig. 4).
To extract features from a resume, we first extract the file name depending on which
we determine whether the file is a doc file or pdf file, depending on which we extract
text from the file. From extracted text, we try to find the following user:
1. Name of the applicant is extracted by implementing pattern search of Spacy.
2. Email ID of applicant is extracted by using regular expression.
3. Mobile number of the applicant is extracted by using regular expression.
4. For extracting experience of an applicant we first tokenized the entire text, then
removed stop words and lemmatized, then to which we parse through regular
expression where we searched for word “experience” in chunks and then parse
the text ahead of it.
5. For extracting skills of an applicant, we created a “skill.csv” file containing
different possible skills that an applicant can possess, which was extracted to
a DataFrame. Then from extracted text we removed stop words, and then each
mono-gram, bi-gram and trigram in text is matched with each skill from the
DataFrame, and if a match is found then that skill is appended to the output
dictionary.
10 R. Kadam et al.
To extract features from the undisclosed skill set, we try to extract skills in a
similar way as we extract them from resumes (Fig. 5).
To calculate scores we pass both undisclosed skill set and resumes to extract skills.
For each match of skill the candidate is allotted five marks and for each extra year
of experience than that expected candidate is allotted ten marks.
where S is the score that would be allotted to each resume, X is the number of skills
matched between applicants resume and job openings undisclosed skill set and Y
is the difference between the candidates year of experience and expected year of
experience by job opening.
1 NLP-Based Resume Screening and Job Recruitment Portal 11
Y = A − B, where A ≥ B (2)
where A is the year of experience that a candidate possesses and B is the minimum
year of experience that the job opening is expecting (Fig. 6).
4 Implementation Details
Features from the resume are extracted with the help of natural language processing.
Extracted features like name, mobile number, email ID, experience, and skills are
added to a fixed dictionary format for comparison between undisclosed skillset and
each resume. Figure 7 shows output of feature extraction from a resume.
12 R. Kadam et al.
Features from undisclosed skillset are extracted again in similar format as they
are extracted from resume. Figure 8 shows the skills that company has mentioned in
the undisclosed skillset that is considered for scoring candidates (Fig. 9).
New applicants who are interested in finding a new job or first job can easily
register with our website. They have to enter their mandatory details as Name, Email
ID, current degree of qualification, current salary that is cost to company in lakh per
annum unit, years of work experience, and password. A new applicant user will also
get a declaration form which will state that if he/she is found to produce false data
in his resume or profile, then companies can add them to block list and then if on
manual evaluation of admin he found that the block listed candidate has entered false
data, then that user would be permanently banned from portal and all his applications
would be nullified in Fig. 10 the registration form is shown which a new applicant
user has to fill while registering.
Candidates can login to their dashboard by entering their email ID and password.
On dashboard they first get to see their personal details, which they can modify
by updating the profile option. They can also upload or update their resume on the
proposed system. They can upload their resume of any format and of file type like pdf
or docx. Then, the major part which candidate gets to see is the set of job openings that
are suitable for his profile. Candidates can view the name of the company posting the
job opening, degree which he is expected to possess, salary which he will be offered
and a job description. Depending on his interest he can apply for any job opening
visible to him. The following Fig. 11 shows the candidate’s dashboard.
Applicants can add their resumes or even update the file of their resume on our
portal by filling form of Fig. 12.
New company users can register on our site by entering their email ID, company’s
name, and password as shown in Fig. 13. After which the admin will get requests
from the company users who want to register with our system, according to which the
Another random document with
no related content on Scribd:
Pineria, 442
Pinna, 449;
shell, 254
Pinnoctopus, 385
Pinnotheres, 62
Pinoceras, 398
Pirena, 417
Pirenella, 416
Piropsis, 424
Pirula—see Pyrula
Pisania, 424
Pisidium, 453;
smell, 195;
ova, 146;
P. pusillum, distribution, 282
Pitys, 327
Placobranchus, 432
Placostylus, 322, 323–325, 359, 442;
radula, 233
Placuna, 448;
P. placenta used for windows, 101
Placunanomia, 448
Placunopsis, 448
Plagioptycha, 347–351, 441
Plagioptychus, 456
Planaxis, 417
Planispira, 311, 312, 319, 441
Planorbis, 27, 247, 439;
monstrosity, 93;
eye, 181;
P. albus, distribution, 282
Platyceras, 76, 412
Platydoris, 434
Platypoda, 411
Platyschisma, 413
Plaxiphora, 403
Plecochilus, 442
Plecotrema, 439
Plectambonites, 505
Plectomya, 459
Plectopylis, 303, 305, 314, 316;
aperture, 63
Plectostylus, 358, 442
Plectotropis, 305, 306, 310, 311, 314–318, 441
Plectrophorus, 298
Plesiastarte, 451
Plesiotriton, 420
Pleurobranchaea, 431;
jaws, 212
Pleurobranchoidea, 431
Pleurobranchus, 245, 428, 431;
warning coloration, 73;
jaws, 212;
radula, 230
Pleurocera, 340, 417
Pleuroceridae, origin, 17
Pleurodonta, 348;
aperture, 63
Pleuroleura, 433
Pleuromya, 458
Pleurophorus, 451
Pleurophyllidia, 433;
breathing organs, 159;
radula, 230
Pleuropyrgus, 357
Pleurotoma, 426, 426;
slit, 263, 265
Pleurotomaria, 266, 373, 376, 407, 407;
prices given for recent, 122;
slit, 156;
radula, 226
Plicatula, 450
Pliny the elder, on use of snails, 118, 120
Plocamopherus, 434
Plochelaea, 425
Plutonia, 298, 440
Pneumoderma, 158, 437, 438
Poecilozonites, 352, 440
Poisonous bite of Conus, 65;
poisonous oysters, 114;
mussels, 117
Polycera, 434;
radula, 230
Polycerella, 434
Polyconites, 456
Polydontes, 346–351, 347, 441
Polygona, 424
Polygyra, 340, 345–353, 441;
aperture, 63
Polygyratia, 246, 263, 357, 442
Polymita, 346–351, 347, 441
Polyplacophora, 9, 401 f.;
radula, 228
Polytremaria, 266, 407
Pomatia, 285, 293, 295, 441
Pomatias, 288, 289, 292 f., 302, 413
Pomatiopsis, 415
Pomaulax, 409
Pompholyx, 250, 341, 439
Ponsonbya, 332
Poromya, 459;
branchiae, 168
Porphyrobaphe, 27, 356, 442
Position of Mollusca in Animal Kingdom, 4
Potamides, 16, 416
Potamomya, 15
Potamopyrgus, 325, 326, 415
Poterioceratidae, 394
Praecardium, 459
Prasina, 449
Prices given for rare shells, 121
Primitive mollusc, form of, 245;
types of, 7
Prisogaster, 409
Pristiloma, 341, 440
Proboscidella, 497, 504
Productidae, 497, 500, 504, 508
Productus, 492, 501, 502, 504;
stratigraphical distribution, 508
Promachoteuthis, 389
Proneomenia, 404;
breathing organs, 154;
nervous system, 203;
radula, 229
Prophysaon, 341, 441;
habits, 44
Propilidium, 405
Proserpina, 21, 355, 410
Proserpinella, 354, 410
Proserpinidae, relationships, 21
Prosobranchiata, 9, 404 f.;
breathing organs, 154
Prosocoelus, 451
Protective coloration, 69 f.;
in snails, 70;
in Nudibranchs, 71 f.;
in other Mollusca, 74
Protegulum, 509
Protobranchiata, 447;
branchiae, 166
Protoma, 417
Protremata, 511
Provocator, 376, 425
Psammobia, 456
Pseudachatina, 328–330, 443
Pseudedmondia, 452
Pseudobalea, 350
Pseudo-deltidium, 498, 511
Pseudodon, 295, 307, 452
Pseudolamellibranchiata, 167, 449
Pseudoliva, 424
Pseudomelania, 417
Pseudomilax, 296, 440
Pseudomurex, 423
Pseudopartula, 323
Pseudosubulina, 440
Ptenoglossa, 224, 411
Pterinaea, 449
Pteroceras, 256, 262, 418
Pteroctopus, 384
Pterocyclus, 266, 267, 300, 316, 414;
tube, 157
Pterodonta, 418
Pteropoda, 7, 434;
breathing organs, 158;
foot, 200;
radula, 230
Pterotrachaea, 421;
foot, 200;
radula, 227
Ptychatractus, 424
Ptychoceras, 399
Ptychodesma, 452
Pugilina, 424
Pulmonata, 10, 22, 151, 185, 438;
origin, 17, 19;
breathing organs, 160;
nervous system, 203
Pulsellum, 444
Punctum, 441
Puncturella, 265, 406
Pupa, 289, 296, 325–357, 442;
P. cinerea, hybrid union, 129
Pupidae, radula, 233
Pupilla, 442
Pupillaea, 406
Pupina, 157, 266, 309, 318–327, 414
Pupinella, 318, 414
Purpura, 423;
operculum, 269;
erosion, 276;
P. coronata, 367;
lapillus, feeding on Mytilus, 60;
on oysters, 111;
protective coloration, 69;
variation, 90;
egg-capsules, 124;
time of breeding, 129;
distribution, 363 n.
Purpuroidea, 423
Pusionella, 426
Pygocardia, 451
Pygope, 497
Pyramidella, 422
Pyramidellidae, 262
Pyrazus, 50, 416
Pyrgina, 330
Pyrgula, 415
Pyrochilus, 441
Pyrolofusus, 423
Pyrula (= Pirula), 419, 420;
spawn, 125;
operculum, 269
Pythina, 453
Quenstedtia, 456
Quoyia, 260, 417
Sabatia, 430
Sactoceras, 394
Sagda, 348–351, 441
Sageceras, 398
Salasiella, 353, 440
Salivary glands, 237
Sandford, on strength of Helix, 45
Sandwich islanders, use of shells, 99
Sanguinolaria, 456
Sarepta, 447
Sarmaticus, 409
Satsuma, 314, 316, 441
Saxicava, 447, 457
Saxidomus arata, money made from, 97
Scalaria, 247, 263, 411;
radula, 224
Scaldia, 452
Scalenostoma, 422
Scaliola, 415
Scaphander, 428, 429, 430;
radula, 231;
gizzard, 238
Scaphites, 399, 399
Scaphopoda, 444;
defined, 6;
breathing organs, 160;
nervous system, 205;
radula, 236
Scaphula, 14, 305, 448
Scarabus, 18, 278, 439, 439
Scharff, R., on food of slugs, 31;
on protective coloration in slugs, 70
Schasicheila, 347, 351, 354, 410
Schismope, 266, 407
Schizochiton, 187, 402, 403
Schizodus, 448
Schizoglossa, 325, 440
Schizoplax, 403
Schizostoma, 413
Schloenbacia, 398
Scintilla, 175, 453
Scissurella, 265, 407;
radula, 226
Sclerochiton, 403
Scrobicularia, 15, 164, 453;
siphons, 164
Sculptaria, 333
Scurria, 405
Scutalus, 356, 442
Scutellastra, 405
Scutus, 245, 406, 406
Scyllaea, 433;
jaws, 212;
stomach, 239
Segmentina, 320
Selenites, 339, 341, 440
Selenitidae, radula, 231
Selenochlamys, 296
Self-fertilisation, 42–44
Semele, 453
Semicassis, 420
Semper, K., on habits of Limnaea, 34;
of Helicarion, 45, 67;
on mimicry, 67;
on parasitic Eulima, 79;
on development of Limnaea, 84, 94;
on sexual maturity in snails, 129;
on Onchidium, 187
Sepia, 381, 385–387, 389;
egg-capsules, 127;
glands, 136;
jaws, 214;
radula, 236;
alimentary canal, 238;
ink-sac, 241;
hectocotylus, 389
Sepiadarium, 389
Sepiella, 389
Sepiola, 389;
glands, 136;
radula, 236
Sepioloidea, 389
Sepiophora, 388
Sepioteuthis, 390;
hectocotylus, 139
Septaria, 337, 338, 410
Septibranchiata, 145, 167, 459;
branchiae, 166
Septifer, 274, 449
Sequenzia, 420
Sergius Orata, 104
Serrifusus, 424
Sesara, 305, 440
Sex, differences of, 133
Shell, 244 f.;
internal, 174;
shape of bivalve, 445
Shell-gland, primitive, 132
Shells as money, 96 f.;
as ornament, etc., 98 f.;
various uses of, 98 f.;
prices given for rare, 121;
sinistral, 249
Shores of N. Asia, no littoral fauna, 2
Showers of shells, 47
Sigaretus, 186, 245, 267, 411;
foot, 198
Sight, 180
Silenia, 459;
branchiae, 168
Silia, 425
Siliqua, 274, 457
Siliquaria, 248, 418
Simnia, 419
Simpulopsis, 345, 350, 442
Simpulum, 420
Simroth, on recent forms of Helix, 22;
on food of slugs, 31;
on crawling of Helix, 45
Singular habitat, 48
Sinistral shells, 249
Sinistralia, 424
Sinusigera, 133
Sipho, 424
Siphonalia, 424
Siphonaria, 18, 431;
classification, 19;
breathing organs, 151, 152
Siphonarioidea, 431
Siphonodentalium, 444
Siphonostomata, 156
Siphonotreta, 493, 496, 504;
stratigraphical distribution, 507, 508
Siphons, 173;
in burrowing genera, 165;
branchial, 155
Sistrum, 75, 423;
radula of S. spectrum, 79, 222
Sitala, 301, 304, 310, 314–319, 333, 440
Skärgard, Mollusca of the, 13
Skenea, 415
Skenidium, 505, 508
Slit, in Gasteropoda, 265, 406
Slugs, habits and food of, 30 f.;
bite hand of captor, 33;
in bee-hives, 36;
in greenhouses, 36;
protective coloration, 70;
eaten in England, 120
Smaragdia, 21
Smaragdinella, 430
Smell, sense of, 192
Smith, W. Anderson, quoted, 98, 111, 114, 191
Snails as barometers, 50;
plants fertilised by, 102;
cultivation for food, 118 f.;
used for cream, 119;
as medicine, 120;
banned by the Church, 121
Solariella, 408;
radula, 225
Solarium, 264, 412, 413;
radula, 224
Solaropsis, 343, 353–357, 442
Solecurtus, 165, 457
Solen, 171, 446, 457;
vision, 190;
habits, 45
Solenaia, 452
Solenomya, 275, 448
Solenotellina, 456
Solomon islanders, use of shells, 98
Somatogyrus, 415
Sophina, 305
Spallanzani, experiments on Helix, 163
Spat, fall of, 113
Spatha, 294, 331, 336, 452
Spekia, 333
Spermatophore, in Cephalopoda, 137;
in Helix, 142
Spermatozoa, forms of, 136
Sphaerium, 453
Sphenia, 456
Sphenodiscus, 398
Sphyradium, 442
Spines, use of, 64
Spiraculum, 266, 414
Spiraxis, 442
Spirialis, 249
Spirifera, 468, 501, 505;
stratigraphical distribution, 507, 508, 511, 512
Spiriferidae, 501, 505, 508
Spiriferina, stratigraphical distribution, 507, 508
Spirobranchiata, 464
Spirotropis, 426;
radula, 218, 219
Spirula, 247, 386, 387, 388
Spirulirostra, 380, 386, 388
Spondylium, 500
Spondylus, 257, 446, 450, 450;
ocelli, 191;
genital orifice, 242
Spongiobranchaea, 437
Spongiochiton, 403
Sportella, 453
Starfish eat oysters, 110
Stearns, R. E. C., on tenacity of life, 38
Stegodera, 306
Stenochisma, 505;
stratigraphical distribution, 507, 508
Stenogyra, 324, 442;
S. decollata, 279;
food, 34;
smell, 194;
Goodallii, 279;
octona, sudden appearance, 47
Stenogyridae, radula, 234
Stenopus, 440;
habits, 45
Stenothyra, 415
Stenotis, 416
Stenotrema, 340, 441
Stephanoceras, 399
Stepsanoda, 358
Stilifer, 76, 77, 79, 422
Stiliferina, 76, 422
Stiliger, 432
Stilina, 76
Stoastoma, 348–351, 410
Stoloteuthis, 389
Stomach, 239
Stomatella, 408
Stomatia, 408
Stomatodon, 302, 417
Strebelia, 353, 440
Strength of Helix, 45
Strephobasis, 417
Strepsidura, 424
Streptaulus, 414
Streptaxis, 302, 306, 309, 314–331, 343, 357–359, 440;
variation, 87
Streptoneura, 203, 404
Streptosiphon, 424
Streptostele, 329, 338, 440
Streptostyla, 343–355, 353, 440
Stricklandia, 505;
stratigraphical distribution, 507, 508
Strigatella, 425
Stringocephalidae, 506, 508
Stringocephalus, 492, 497, 498, 500, 501, 506;
stratigraphical distribution, 507, 508
Strobila, 340, 345–353
Strobilops, 442
Strombidae, habits, 64;
penis, 136
Strombina, 423
Strombus, 69, 200, 252, 418;
mimicking Conus, 69;
operculum, 78, 269;
pearls from, 101;
metapodium, 199;
stomach, 239
Strophalosia, 504;
stratigraphical distribution, 507, 508
Stropheodonta, 497, 505, 508
Strophia, 343–355, 442;
S. nana, 278
Strophochilus, 358, 441
Strophomena, 499, 505;
stratigraphical distribution, 507, 508
Strophomenidae, 500, 505, 508
Strophostoma, 248, 414
Structure of shell, 252
Struthiolaria, 99, 418;
radula, 216
Styliola, 437
Stylodonta, 339, 441
Stylommatophora, 11, 181, 439;
origin, 19
Subemarginula, 406
Submytilacea, 451
Subularia, 422
Subulina, 332, 352, 442
Subulites, 420
Succinea, 325, 327, 358, 433;
jaw, 211;
S. putris, parasite of, 61
Succineidae, 443;
radula, 234
Sudden appearance of Mollusca, 46
Suessia, stratigraphical distribution, 507
Sulphuric acid, 237
Surcula, 426
Sycotypus, 424
Synaptocochlea, 408
Syndosmya, 453
Syringothyris, 500, 508
Syrnola, 422
Syrnolopis, 332, 333
Systrophia, 356, 357
Tachea, 441
Taenioglossa, 223, 411
Taheitia, 414
Talona, 457
Tanalia, 304, 417
Tancredia, 453
Tanganyicia, 332, 415
Tanganyika, L., fauna of, 12
Tanysiphon, 454
Taonius, 391, 391
Tapes, 454
Taste, 179
Tebennophorus, 143, 340, 440
Tectarius, 413
Tectibranchiata, 10, 429
Tectura, 305, 405
Tectus, 408
Teeth in aperture of the shell, 63
Teinostoma, 247, 408
Teinotis, 407
Telescopium, 252, 416
Tellina, 440, 453, 453
T. balthica, variation, 84
Tellinacea, 453
Telotremata, 511
Tenacity of life, 37
Tenison-Woods, on red blood, 171;
on shell-eyes, 189
Tennent, Sir J. E., on musical sounds produced by Mollusca, 50
Tennentia, 304, 314, 338, 440
Terebellum, 418;
jumping powers, 64
Terebra, 246, 263, 426, 426;
radula, 219
Terebratella, 468, 487;
distribution, 486;
fossil, 506;
stratigraphical distribution, 508
Terebratula, 467, 468, 487;
size, 484;
distribution, 485, 486;
fossil, 492, 499, 506;
stratigraphical distribution, 506, 507, 508
Terebratulidae, 487;
fossil, 500, 505, 506;
stratigraphical distribution, 507, 508
Terebratulina, 466, 479, 487;
larva, 482;
distribution, 486;
fossil, 506;
stratigraphical distribution, 508;
form of shell, 510
Teredina, 457
Teredo, 262, 457, 458;
nervous system, 206;
intestine, 241
Tergipes, 432
Terquemia, 450
Testacella, 22, 52, 440;
habits, etc., 49, 51 f.;
pulmonary orifice, 160;
eyes, 186;
radula, 231;
anus, 241
Testicardines, 466, 487;
muscles, 476;
fossil, 497, 504;
external characters, 497;
internal characters, 499;
attachment of muscles, 501;
stratigraphical distribution, 508
Testis, 135
Tethyidae, 216
Tethys, 432
Tetrabranchiata, 391 f.
Thala, 425
Thalassia, 319
Thalotia, 408
Thapsia, 329
Thaumasia, 349, 442
Thaumastus, 356, 442
Thecacera, 434;
radula, 229
Thecidiidae, 487;
fossil, 501, 506, 508
Thecidium, 475, 479, 480, 483, 487;
fossil, 506, 508
Thecosomata, 435
Thelidomus, 346–351, 350, 441
Theora, 453
Therasia, 441
Thersites (Helicidae), 322, 325
Thersites (Fasciolariidae), 424
Thetis, 454
Thracia, 245, 459
Thread-spinning, 29
Thridachia, 432
Thyca, 76, 79
Thyrophorella, 330, 440
Thysanoteuthis, 390
Tiedemannia, veliger, 132
Tiphobia, 332, 333, 417
Titicaca, L., Mollusca of, 25
Todarodes, 390
Tomichia, 414
Tomigerus, 334, 356, 358, 442
Tomocyclus, 354
Tomostele, 330, 440
Tonicella, 403
Tonicia, 403;
eyes, 188
Torellia, 411
Torinia, 413;
radula, 224;
operculum, 269
Tornatellina, 278, 319, 323–327, 338, 358, 443
Tornatina, 250, 430
Torquilla, 442
Toucasia, 455
Touch, sense of, 177
Toxoglossa, 218, 426
Trachia, 314
Trachyceras, 397
Trachydermon, 403
Trachyteuthis, 389
Tralia, 439
Transovula, 419
Trematis, 492, 493, 504;
stratigraphical distribution, 507, 508
Trematonotus, 407
Tremoctopus, 384;
radula, 236;
hectocotylus, 137
Trevelyana, 434
Trichia, 316
Trichotropis, 275, 411
Tricula, 302
Tridacna, 273, 455
Triforis, 416;
radula, 224
Trigonellites, 397
Trigonia, 15, 254, 269, 448;
jumping powers, 65;
distribution, 370
Trigonochlamys, 296, 440
Trigonostoma, 426
Trimerella, 495, 504, 508, 511
Trimerellidae, 493, 494, 496, 504;
stratigraphical distribution, 507, 508
Trinacria, 448