Social Computing (CS60017)

Autumn semester 2020-21

Announcements


  • Assignment 3 declared -- see details below.

  • Class Test 3 on Thursday, November 19, during the class hours. Syllabus -- Ten selected papers

  • Class Test 2 on Friday, November 06, during the class hours. Syllabus -- Community Detection, Fairness and bias, Fake News / rumour / claim identification and verification

  • Schedule of paper presentations by students has been declared -- presentations to start from Oct 28 -- see details below. Also given are some rules and guidelines for the paper presentations.

  • Class Test 1 on Friday, October 16, during the class hours. Syllabus -- Structural properties of large networks, Network centrality.

  • Guest lecture by Dr. Abhijnan Chakraborty in the classes on Oct 7 and Oct 13

  • Assignment 2 declared -- see details below.

  • Assignment 1 declared -- see details below.

  • Every registered student should join the Google mailing list social-computing-2020@googlegroups.com. You need to login with your Google account (e.g., Gmail account) and use this link: https://groups.google.com/forum/#!forum/social-computing-2020. All announcements about the course will henceforth be made on this website or through this mailing list.

  • Every registered student should create an account on the Moodle system of CSE department. This system will be used for submission and grading of assignments. If you do not have an account already on the CSE department Moodle, create a new account for yourself following the given procedure. Login to the system, and follow the link "Autumn Semester (2020-21)". Choose the course "CS60017_A20-21 Social Computing". Join this course as "Student"; use Student Enrolment Key: STD01.

  • First class on Wednesday, September 02, at 11 am. Join the class Social-Computing-2020A on MS Teams (IITKGP domain) using the Team Code uv5obvr

  • The course requires a good knowledge of algorithms and data structures, graph algorithms, probability and statistics, Natural Language Processing and Machine Learning (see Prerequisites below). This is a research-oriented course that would require students to understand several CS research papers. Programming assignments need to be done using Python/C++. It is advisable to take this course only if you have the necessary background.

  • There are too many applicants for the Social Computing course this year. While I would like to accommodate as many students as possible, I have capacity constraints -- I will be able to accommodate only 100 students in the class. For those students who need faculty approval for registration through ERP, the applications will be accepted on basis of CGPA of the students. There is no need to email me about registration to the course; I will not be able to respond to individual emails.


Instructor

Saptarshi Ghosh (Email: saptarshi @ cse . iitkgp . ac . in)

Course Timings and Mode of Teaching

Slot G3: Wednesday 11:00--11:55 | Thursday 12:00--12:55 | Friday 08:00--08:55
Slot G(A): Tuesday 20:00--21:30

Classes will either be online over MS Teams (IITKGP domain), or pre-recorded videos will be uploaded to this website.
Students are required to join the class Social-Computing-2020A on MS Teams (IITKGP domain) using the Team Code uv5obvr

Rules for students while attending online classes:
  • Please keep your video off and microphone muted, unless specifically asked to do otherwise.
  • If you have any question about the topic being taught, type your question in the CHAT. Please do NOT start talking suddenly. The teacher/presenter will intermittently answer the questions typed in the chat.
  • If a question requires further discussion, then the teacher/presenter will ask the corresponding student (who wrote that question) to speak. Only then should the corresponding student unmute himself/herself and speak. After the discussion, please mute yourself again.
  • In case the teacher/presenter gets disconnected, please wait for at least 10 minutes for the teacher/presenter to re-connect.
I will attempt to record the classes taken via MS Teams. The recordings will be available on MS Teams itself. Note that class recording vidoes on MS Teams expire in 20 days (w.r.t. the date they are created); hence videos need to be downloaded before they expire.

Recordings may not be available for guest lectures, or even for regular classes in case of technical snag, or if I forget to record some session. Availability of recordings is not a valid reason for not attending regular classes. In case you miss a class and its recording is not available for any reason, I would not be able to do anything about this.

Pre-requisites for the course

  • Data structures and algorithms
  • Graph algorithms
  • Probability and Statistics
  • Basics of Machine Learning
  • Basics of Natural Language Processing
  • Programming in Python/C++ (there will be programming-based assignments)
  • The course will involve understanding of several research papers, and students will have to present at least one research paper in online class.

Here are some sample papers that you should be able to understand and present in the course: paper on community detection | paper on sentiment analysis | paper on hate speech detection

Course evaluation [tentative]

  • Online tests/quizzes: 40%
  • Take-home programming assignments (3-4 in number): 40%
  • Presenting research paper(s): 20%
Note: The evaluation plan is subject to change based on institute/department policies and other related factors.

Every assignment, quiz / online test should be attempted individually by each student, unless otherwise specified. We will use standard plagiarism detection tools over the assignment submissions. Plagiarism in any form -- copying from other students or from online resources -- will be severely penalized.

Teaching Assistants

  • Abhisek Dash (assignmentad @ gmail . com)
  • Paheli Bhattacharya (paheli.cse.iitkgp @ gmail . com)
  • Soham Poddar (sohampoddar @ kgpian.iitkgp.ac.in)
  • Shounak Paul (shounakpaul95 @ gmail . com)
  • Manjusha M (manjusha123shyny @ gmail . com)


List and grouping of students

List of students from ERP

Student groups and schedule of paper presentations

Rules and guidelines for paper presentations

Assignments

Assignment 1 (Network properties): Problem statement
For any queries, contact TA Soham Poddar
Deadline: 27 September 2020, 23:59 pm IST
Evaluation by TAs: Sl. no. 01-39: Paheli | 40-78: Shounak | 79 onwards: Manjusha

Assignment 2 (Network centrality): Problem statement
For any queries, contact TA Soham Poddar
Deadline: 21 October 2020, 23:59 pm IST
Evaluation by TAs: Sl. no. 01-39: Manjusha | 40-78: Paheli | 79 onwards: Shounak

Assignment 3 (Hate speech detection): Problem statement
For any queries, contact TA Soham Poddar
Deadline: 14 November 2020, 23:59 pm IST
Evaluation by TAs: Sl. no. 01-39: Shounak | 40-78: Manjusha | 79 onwards: Paheli

[Serial numbers according to the list of students given above]

Topics

The broad topics to be covered in the course include:
  1. Social network analysis -- introduction, applications and challenges
  2. Structural properties of large networks (social networks, random networks, technological networks, etc.)
  3. Network centrality; identifying popular users/experts
  4. Community structure in social networks
  5. Social media text analysis -- introduction, applications and challenges
  6. Harmful users/content on social media -- hate speech, fake news, spammers in social networks, etc.
  7. Bias in social computing systems
  8. Different types of social platforms -- Anonymous social networks, E-commerce sites
  9. Social media in times of COVID


Text and Reference Literature

  1. Networks, Crowds and Markets - Easley and Kleinberg
  2. Social Network Data Analytics - Charu Aggarwal (ed.) - Springer, 2011
  3. Mining of Massive Datasets - Jure Leskovec, Anand Rajaraman, Jeff Ullman
  4. Research papers to be pointed out in class

Class slides and reference materials

Class dates Topic Slides Notes (Relevant papers discussed in class, additional references, etc.)
Sep 2, 3, 4 Introduction Slides Introduction to the course
Social networks in the offline and online worlds
Research challenges on social networks / social media
Sep 8, 9, 10, 15, 16 Structural properties of large networks Slides 1. The Structure and Function of Complex Networks - Newman
2. Assortative Mixing in Networks - Newman
3. Measurement and Analysis of Online Social Networks - Mislove et al.
4. The Anatomy of the Facebook social graph - Ugander et al.
5. What is Twitter, a Social Network or a News Media? - Kwak et al.
Sep 17, 23,
Sep 24, 29, 30,
Oct 01, 06, 08, 14, 15		 Network centrality Slides (part 1)
Slides (part 2)
Slides (part 3)
 1. Authoritative Sources in a Hyperlinked Environment - Kleinberg
2. The PageRank Citation Ranking: Bringing Order to the Web - Page et al.
3. Topic-sensitive PageRank - Haveliwala
4. Combating Web Spam with TrustRank - Gyongyi et al.
5. Measuring User Influence in Twitter: The Million Follower Fallacy - Cha et al.
6. Understanding and combating link farming in the twitter social network - Ghosh et al.
7. Cognos: Crowdsourcing Search for Topic Experts in Microblogs - Ghosh et al.
Oct 07, 13 Fairness in two-sided platforms Slides Guest lecture by Dr. Abhijnan Chakraborty

Additional References on Fairness of Machine Learning:
SIGKDD 2019 tutorial
Oct 20, 21 Subgraphs and Community Structure Slides 1. Community structure in social and biological networks - Newman, Girvan, PNAS 2002
2. Empirical Comparison of Algorithms for Network Community Detection - Leskovec et al, WWW 2010
3. Deep Twitter Diving: Exploring Topical Groups in Microblogs at Scale - Bhattacharya et al, CSCW 2014

Additional References:
Community detection in graphs - Fortunato, Physics Reports, 2010
Generalized Measures for the Evaluation of Community Detection Methods - Labatut
Oct 28 onward Research paper presentations by students Topics covered:

Fairness & bias
Hate Speech
Fake News / rumour / claim identification and verification
Social search and Recommendation
E-commerce sites
Echo chambers and filter bubbles
Epidemic / Convention Spreading
Sentiment Analysis
COVID and social media
Social media during disasters