logo

Social Media Computing


In this course, we study social networks by analyzing social relation between users, contents they share, and ways contents and events are propagated through such networks. This helps us to better understand the structure of social networks and uncover the live and emerging social phenomena that will affect communities. You will dig up new innovative ideas in this context and evaluate them on real world datasets. At the end of this course, you will hopefully have a good understanding about the above issues and will have hands-on experience on a range of tasks from identifying important nodes to detecting communities and trends in social networks.


GENERAL INFO & ANNOUNCEMENTS

Time: Mon / Wed 2:00pm - 3:15pm (Spring 2015)
Class: CSI 1121
Piazza: CMSC 498J

Lecturer: Hadi Amiri, Office: AVW 3161, Office hours: Mons 3:30-4:30

Final Exam: Monday, May 18, 1:30-3:30pm
More info at https://ntst.umd.edu/soc/201501/CMSC/CMSC498



05-15: Fianl exam! Mon 18th, 1:30-3:30 pm CSI 1121

05-11: Recap session! Mon 11th, 2:00-3:15 pm

04-30: Watch Timelines at Scale.

04-30: HW4 is out.

04-27: Project Reports due date: 05/11/2015

04-17: Guest lecture by Yulu Wang on Mon 04/20 class.

04-15: Watch Documentary on Six Degrees of Separation

04-13: HW3 is out.

04-10: Course survey on week 06-11

04-06: 3-min TED talk! on information cascade!

04-03: Project ideas and groups

03-30: HW2 is out.

03-03: Watch Predicting Positive and Negative Links in Online Social Networks

03-03: Watch WTF!

02-27: Course survey on week 01-05

02-19: Proposal due date: 03/13/2015.

02-17: Guest lecture by Svitlana Volkova on Wed 02/18 class.

02-15: HW1 is out.

02-02: Piazza: CMSC 498J

01-27: No Class on Wed 01-28, watch this instead!

01-26: Welcome to CMSC 498J!

01-10: Homepage uploaded!


TEXTBOOKS

The textbooks are freely available:

[NCM] Networks, Crowds, and Markets: Reasoning About a Highly Connected World
David Easley and Jon Kleinberg

[MMD] Mining of Massive Datasets
Jure Leskovec, Anand Rajaraman, Jeffrey D. Ullman


SYLLABUS

Slides will be uploaded before classes, often Mon mornings.

W01: Jan 26 - Jan 28 Introduction: course overview and logistics, social media history, and Twitter [Slides-0] [-1]
Ch.01 Overview [NCM]
Ch.10.1 Social Networks as Graphs [MMD]
TED talk: From Gaga to Water
What is Twitter, a social network or a news media? Kwak, Lee, Park, Moon. WWW 2010.
W02: Feb 02 - Feb 04 Network Basics: connectivity, bfs, shortest path, bipartite graphs [Slides-1] [-2]
Ch.02 Graphs [NCM]
Ch.22 Elementary Graph Algorithms [Introduction to Algorithms (CLRS), Cormen, Leiserson, Rivest, Stein. 2009].
Ch.24 Single-Source Shortest Paths [Introduction to Algorithms (CLRS), Cormen, Leiserson, Rivest, Stein. 2009].
W03: Feb 09 - Feb 11 Strong and Weak Ties: centrality, betweenness, graph clustering [Slides-1] [-2]
Ch.03 Strong and Weak Ties [NCM]
Ch.07 Clustering [MMD]
Ch.10.2 Clustering of Social-Network Graphs [MMD]
Community Detection in Graphs. Fortunato. 2009
W04: -- - Feb 18 Guest Lecture: User Profiling [Slides-1]
Inferring Latent User Properties from Texts Published in Social Media. Volkova, Bachrach, Armstrong, Sharma. AAAI 2015.
W05: Feb 23 - Feb 25 Fun topics: ideas, tools and datasets, vw tutorial. [Slides-1] [-2]
Recipe recommendation using ingredient networks. Teng, Lin, Adamic. Web Science 2012
Target-dependent Churn Classification in Microblogs. Amiri, Daume III. AAAI 2015
vw tutorial
W06: Mar 02 - Mar 04 Node Analysis: node similarity, homophily, and link prediction [Slides-1] [-2]
Ch.04 Networks in Their Surrounding Context [NCM]
The Link Prediction Problem for Social Networks . Liben-Nowell and Kleinberg. CIKM 2003.
Link Mining: a survey. Getoor, Lise, and Christopher Diehl. SIGKDD 2005.
W07: Mar 09 - Mar 11 Balanced and unbalanced Networks: positive and negative relationships and structural balance [Slides-1] [-2]
Ch.05 Positive and Negative Relationships [NCM]
Predicting Positive and Negative Links in Online Social Networks. Leskovec, Huttenlocher, Kleinberg. WWW, 2010.
W08: Mar 16 - Mar 18 ENJOY the SPRING BREAK :)
No class
W09: Mar 23 - Mar 25 Web Graph and Network Popularity: connected components, bow-tie structure, power Law, heavy tail [Slides-1] [-2]
Ch.13 The Structure of the Web [NCM]
Finding Strongly Connected Components
Ch.18 Power Laws and Rich-Get-Richer Phenomena [NCM]
W10: Mar 30 - Apr 01 Link Analysis: ranking and retrieval, hits and page rank [Slides-1] [-2]
Ch.14 Link Analysis and Web search [NCM]
Ch.05 Link Analysis [MMD]
W11: Apr 06 - Apr 08 Information Cascading: information diffusion and basic cascade model [Slides-1] [-2]
Ch.16 Information Cascades [NCM]
Ch.19 Cascading Behavior in Networks [NCM]
TED talk: how to start a movement!
W12: Apr 13 - Apr 15 Small world Phenomenon: six degree separation and decentralized search [Slides-1] [-2]
Ch.20 The Small-World Phenomenon [NCM]
Four Degrees of Separation . Backstrom et. al. Web Science Conference. 2012.
The Small World Problem . Milgram. Psychology Today 1967.
W13: Apr 20 - Apr 22 Guest Lecture: Social Search; Mining Data Streams [Slides-1] [-2]
Overview of the TREC-2014 Microblog Track. Lin, Efron, Wang, and Sherman. TREC 2014.
#TwitterSearch: a comparison of microblog search and web search. Teevan, Ramage, and R. Morris. WSDM 2011.
Finding High-quality Content in Social Media. Agichtein, Castillo, Donato, Gionis, Mishne. WSDM 2008.
Ch.04 Mining Data Streams [MMD]
W14: Apr 27 - Apr 29 Trend detection and tracking, locality-sensitive hashing, and distance measures [Slides-1] [-2]
Making Sense of Micro-posts for organizations and brands. Amiri. 2013.
Ch.03 Finding Similar Items [MMD]
W15: May 04 - May 06 Project Presentations! [Slides-1] [-2]
In class presentation.
W16: May 11 Recap. [Slides-1]


GRADING and WORKLOAD

Grading for this course will comprise of the following assessments:
Workload for this class is expected to be in range of (2.5-4.0-2.5) that is translated as:
If you are spending significantly more (or less!) time than the above suggestion, you should talk to us to see if we need to do any adjustment.


POLICIES

Attendance: is NOT mandatory but highly recommended. Your participation will make the class more fun and interactive and help us to have a better estimation about your understanding of materials. If you plan to attend the classes, please be on time. Also, giving constructive feedback is highly recommended and appreciated as it helps us to improve the quality of the class.

Collaboration: students are encouraged to work together, and to teach and help each other. Teaching others will extend and deepen your understanding of the topic. However, cheating is a very serious offense and will be revealed and reported to the dean office with no exceptions. Please do not do it and always follow the UMD's honor code. You must always write name(s) of your collaborators on your assignments.

The Facebook Rule (CREDIT: Min Yen Kan): This rule says that you are free to meet with fellow students and discuss assignments with them. Writing on a board or shared piece of paper is acceptable during the meeting; however, you may not take any written (electronic or otherwise) record away from the meeting. This applies when the assignment is supposed to be an individual effort. After the meeting, engage in a half hour of mind-numbing activity (like catching up with your friends and family's activities on Facebook), before starting to work on the assignment. This will assure that you are able to reconstruct what you learned from the meeting, by yourself, using your own brain.

Homeworks: Homework assignments should be done individually and are due to by 3:15pm on Monday classes. NO emails please (except where otherwise stated). The following penalties will apply for late submissions:

We do our best to return all homework assignments to you within three weeks of their due date. Students have a right to question the grading of their work within 3 days of the return of the preliminary grades by email.

Final Project: Final project is a substantial part of this class. You should form a team of four students for the final project (this is required). All projects will be presented in the class on the designated week. Note that each student should clearly determine his duties in the project, actively contribute in the project, and try to help other team members. Project report and proposal should be returned in the provided templates (see the Final Project Section below).

Final Exam: The final exam will be closed-book, closed-note, and closed-*.

Academic Accommodation: Students who are eligible for academic accommodations due to a disability should provide a letter of accommodation from the Office of Disability Support Services (DSS) within the first two weeks of the semester. You may reach them at 301-314-7682 or by visiting Shoemaker Building.


HOMEWORKS

There are 4 homework assignments that cover 30% of your final grade. We only consider top 3 of your homework grades, each 10%.

Homework Due Date
HW1 03/02/15, 3:15 PM
HW2 04/08/15, 3:15 PM
HW3 04/22/15, 3:15 PM
HW4 05/11/15, 3:15 PM


FINAL PROJECT

Project is a substantial part of this class and covers 40% of your final grade. The project should be an innovative idea that is evaluated on real or synthetic datasets. You need to write a 2-page proposal and a 7-page final report (including all references, figures and tables, etc):

If you need any help or advice about your project, visit us at office hours.


RESOURCES

These are some useful resources that you might use in your projects:

Tools Datasets Tutorials