[Announcements]
[General Information]
[Syllabus]
- Deadline for Literature Review: March 30 at 11:59pm. Again one sumission per team.
- Notice slight change in schedule as discussed in class! The change in schedule of discussion presentation has led to some conflict for 2 presenters so those have been moved to solve the conflict.
- [9/29/2018] Unassigned papers can be presented for extra credit as mentioned in class. If you are interested please email the instructor (will be assigned based on request first come first served).
- [9/29/2018] Assignment of students to papers is done (based on responses we had at 9:00am today) . We manage to match the top choices of everyone. WHile majority of students filled to survey there are few who did not. If you did not fill up the survey you can email the instructor and select papers that are no currently assigned. If you do not email by end of class today, we will assign you to a paper.
- [9/25/2018] List of papers for discussion is up. You will have to select your top 10 choices by filling up this Survey
You will have until Sunday January 28 at 8pm to fill in the survey.
Note: there are some short papers (5 pages), we expect that for those papers we will only assign one student to present.
- [9/24/2018] List of papers for discussion will be up soon and an announcement will be posted here and on Coursework.
Once the announcement is sent you will have 2 days to pick papers (we will mention day and time). And a link to survey
to select 5 top choices will be posted.
- As discussed in class, to facilitate the discussions on sentiment starting on Wed 1/31 papers for that section
have been posted. If you want ot present one those papers you can email the instructor before even the survey is up.
your name will be added to the papers in first come first serve.
- Please fill in the survey about your background by Friday January 23 at 12pm: Survey about Background
- Welcome to Computational Models of Social Meaning!
[Top]
Description
Computational Models of Social Meaning is a seminar in Natural Language Processing, focusing on computational methods for extracting social
and interactional meaning, mainly from text (both traditional media and social media) and speech. Topics include detection of speaker's sentiments, emotions, opinions and beliefs,
sarcasm, deception, persuation, perspective, power and influence, politeness, and personality. Analysis of meaning-bearing characteristics of the
speaker and topic, including text, discourse, prosodic and other cues.
Prerequisites
- COMS 3133/4/7/9 (Data Structures) or equivalent programming ability in at least one systems or scripting language (C++, Java, Python)
- smara [who is at] columbia [dot] edu
- Office Hours: Wed 5:30pm-6:30pm
Office: Interchurch Center (61 Claremont Avenue --- SW corner of 120th Street and Claremont Avenue), Room 850
TA: Kevin Raji Cherian
- Email: kr2741@columbia.edu
- Office Hours: Wednesday 10:00-12:00pm
- Office: TA room Mudd Building
TA: Ryan Serrao
- Email: rs3774@columbia.edu
- Office Hours: Fridays 6:00-8:00pm
- Office: TA room Mudd Building
Lectures
- Monday, Wednesday 4:10-5:25, Hamilton 303. Link to Columbia University Campus Map
The classes consists of lectures and discussions of research papers led by students. To facilitate the discussion of research articles
on a particular topic each week, the topic is introduced the previous week. Usually Wednesdays will be lectures and Mondays discussion of research articles led by students
Lecture notes will be posted on CourseWorks.
Grade Breakdown
- 15% Data Analysis Assignments
There will be 3 data analysis assignments that will require relatively small amount of work to learn
about a data source. You will be asked to answer one or two questions and to notice some interesting
points about the data source. The grading for this will be Excellent/Good/Insufficient. The data
analysis assignments will be due before class, on Courseworks site. Late assignments will not be accepted, except emergency situations.
- 20% Discussion of Papers
Each student will do a critical discussion of ONE of the research articles proposed for reading.
Students will prepare a brief presentation of the paper (What is the goal of the paper? The research hypothesis? What data they use? What methods? How is evaluated?) followed by leading a critical
discussion on key positive and negative aspects, availability of code and datasets. You can consider as inspiration guidelines given to reviewers of conference papers Review Form ). But all ratings should be accompanied by a critical discussion that justifies your scores.
Due to class size you will present in pairs (two students per paper). Teaming will be based on your paper choices (you will be asked to send a ranked list of top 5 papers you'd like to discuss)
- 5% Class Participation
- 60% Final Project
Design and implement a software
product of appropriate scope and complexity given the time
constraints. Project should be related to one topic discussed in class or related issues.
Prepare a literature review (by mid semester), give a class presentation and a paper write-up
describing the methodology and results. The projects will be done in groups of 3-4 students.
Split up of grades will be 10% for literature review, 5% class presentations, 45% for project and final paper.
Contribution of each participant will need to be clearly stated at each stage of the project deliverables.
[Top]
Schedule is tentative and subject to change.
Date |
Topics |
Readings |
Due dates |
1/17 |
Class Introduction |
|
|
1/22 |
Lecture: Sentiment Analysis (Word Level) |
|
|
1/24 |
Lecture: Sentiment Analysis (Sentence Level; Target-based Sentiment; multilinguality) |
|
|
1/29 |
Lecture: Deep Learning for Sentiment Analysis (word, sentence, and multilinguality) |
|
|
1/31 |
Discussion: Sentiment Analysis |
- [S1] Velikovich, Leonid, Sasha Blair-Goldensohn, Kerry Hannan, and Ryan McDonald. 2010. The viability of web-derived polarity lexicons. Proceedings of NAACL 2010.
- [S2] Brendan O'Connor, Ramnath Balasubramanyan, Bryan R. Routledge, and Noah A. Smith. 2010. From Tweets to Polls: Linking Text Sentiment to Public Opinion Time Series. In Proceedings of the International AAAI Conference on Weblogs and Social Media, Washington, DC, May 2010 [Carlo Provinciali]
|
|
2/5 |
Discussion: Sentiment Analysis |
|
|
2/7 |
Lecture: Emotion/Mood |
|
|
2/12 |
Discussion: Emotion/Mood |
- [EM1] Rada Mihalcea and Carlo Strapparava. 2012. Lyrics, music, and emotions. In Proceedings of Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL '12). [David Lee; Yunwe Cai]
- [EM2] Akari Asai, Sara Evensen, Behzad Golshan, Alon Halevy, Vivian Li, Andrei Lopatenko, Daniela Stepanov, Yoshihiko Suhara, Wang-Chiew Tan and Yinzhan Xu (2018). HappyDB: A Corpus of 100,000 Crowdsourced Happy Moments . Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018). [ Mukund Yelahanka Raghuprasad]
- [EM3] Muhammad Abdul-Mageed, Lyle H. Ungar (2017). EmoNet: Fine-Grained Emotion Detection with Gated Recurrent Neural Networks. Proceedings of ACL 2017 [Kilol Gupta; Haozheng Ni ]
|
|
2/14 |
Lecture: Irony/Sarcasm |
|
|
2/19 |
Discussion: Irony and Sarcasm Detection |
- [IS1] Abhijit Mishra, Diptesh Kanojia, Kuntal Dey, Seema Nagar and Pushpak Bhattacharyya (2016). Harnessing Cognitive Features for Sarcasm Detection . Proceedings of ACL 2016. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, pages 1095?1104. (extra credit:Surabhi Bhargava; Yogeshwar Mutneja)
- [IS2] Aniruddha Ghosh, Tony Veale (2017). Magnets for Sarcasm: Making Sarcasm Detection Timely, Contextual and Very Personal. Proceedings of EMNLP 2017 [Animesh Sharma; Tuhin Chakrabarty]
- [IS3] Bjarke Felbo, Alan Mislove, Anders S�gaard, Iyad Rahwan, Sune Lehmann (201). Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm. Proceedings of EMNLP 2017. [Josh Feldman; Han Xu]
|
|
2/21 |
Lecture: Argument Mining
|
|
|
2/26 |
Discussion: Sentiment (cont) + Argument Mining |
- [AM1] Vanessa Wei Feng and Graeme Hirst (2011). Classifying arguments by scheme . In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1 (HLT '11), [Somya Singhal]
- [AM2] Johannes Daxenberger, Steffen Eger, Ivan Habernal, Christian Stab and Iryna Gurevych (2017) What is the Essence of a Claim? Cross-Domain Claim Identification . Proceedings of EMNLP 2017. [Tariq Alhindi]
- [S5] Mitchell, M., and Aguilar, J., and Wilson, T., and Van Durme, B. (2013). Open Domain Targeted Sentiment . Proceedings of EMNLP 2013 [Zihan Ye]
|
|
2/28 |
Lecture: Perspective/Framing/Ideology |
|
|
3/5 |
Lecture: Persuasion Guest Lecturer: Chris Hidey |
|
|
3/7 |
Canceled due to snow |
- [PF1] moved to 3/21
- [PF2] moved to 3/19
- [PF3] moved to 3/28
|
|
3/12 |
No Class (Spring Break) |
3/14 |
No Class (Spring Break) |
3/19 |
Discussion : Perspective/Framing/Ideology(cont)+ Persuasion |
|
|
3/21 |
Canceled due to Snow |
|
|
3/26 |
Discussion: Framing and Perception/Ideology |
|
|
3/28 |
Lecture: Deception Detection Guest Lecture: Sarah Ita Levitan |
|
|
4/2 |
Discussion: Deception Detection |
[D1] Jiwei Li, Myle Ott, Claire Cardie and Eduard Hovy. (2014). Towards a General Rule for Identifying Deceptive Opinion Spam.
ACL 2014. [Surabhi Bhargava; Weiqi Tong]
- [D2] Veronica Perez-Rosas, Mohamed Abouelenien, Rada Mihalcea, Yao Xiao, CJ Linton and Mihai Burzo (2015). Verbal and Nonverbal Clues for Real-life Deception Detection . Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2015), Lisbon, Portugal, September 2015.
[Xiaochun Ma; Yogeshwar Mutneja]
- [D3] Hannah Rashkin, Eunsol Choi, Jin Yea Jang, Svitlana Volkova & Yejin Choi. 2017. Truth of Varying Shades: Analyzing Language in Fake News and Political Fact-Checking. Proceedings EMNLP 2017 (short paper) [Chuqi Yang]
|
|
4/4 |
Lecture: Personality and Interpersonal Stanceand Tentiative: Social Impact and Ethics |
|
|
4/9 |
Discussion: Personality and Interpersonal Stance Tentative aslos Social Impact and Ethics/TD>
|
- [PI1] Daniel Jurafsky, Rajesh Ranganath, Daniel A. McFarland (2009) Extracting Social Meaning: Identifying Interactional Style in Spoken Conversation. HLT-NAACL 2009: 638-646 [Kailin Lu ]
- [PI2] T. Maheshwari, A. N. Reganti, S. Gupta, A. Jamatia, U. Kumar, B. Gamb�ck, and A. Das.(2017) A Societal Sentiment Analysis: Predicting the Values and Ethics of Individuals by Analysing Social Media Content. in the 15th European Chapter of the Association for Computational Linguistics (EACL 2017), Valencia, Spain (extra credit: Somya Singhal; Tuhin Chakrabarty) )
- Tentative also the papers on Ethics
- [Eh1]Nitin Madnani, Anastassia Loukina, Alina von Davier, Jill Burstein, Aoife Cahill (2017). Building Better Open-Source Tools to Support Fairness in Automated Scoring . Proceedings of the Workshop of Ethics in NLP at EACL 2017 (short paper) (extra credit: Ruiqi Zhong)
- [Eh2] Rachel Rudinger, Chandler May, and Benjamin Van Durme. 2017. Social Bias in Elicited Natural Language Inferences. In The 15th Conference of the European Chapter of the Association for Computational Linguistics (EACL): Workshop on Ethics in NLP. (short paper) (extra credit: Serina Chang; Navie Narula)
|
|
4/11 |
Lecture: Extracting Social Networks |
|
|
4/16 |
Discussion: Extracting Social Networks /TD>
|
|
|
4/18 |
Lecture: Social Power/Influence |
|
|
4/23 |
Discussion: Social Power/Influence; Social Impact of NLP |
- [SP1] Danescu-Niculescu-Mizil, Cristian, Moritz Sudhof, Dan Jurafsky, Jure Leskovec, and Christopher Potts. 2013. A computational approach to politeness with application to social factors . Proceedings of ACL. [Navie Narula]
- [SP2] Prabhakaran, Vinodkumar; Reid, Emily; Rambow, Owen (2014). Gender and Power: How Gender and Gender Environment Affect Manifestations of Power . In Proceedings of the conference on Empirical Methods for Natural Language Processing (EMNLP). October, 2014. Doha, Qatar. (extra credit: Sidhi Adkoli; Mukund Raghuprasad)
- [SP3] Maarten Sap, Marcella Cindy Prasetio, Ariel Holtzman, Hannah Rashkin & Yejin Choi (2017). Connotation Frames of Power and Agency in Modern Films. Proceedings of EMNLP (short paper) [Xingyu Chen]
|
|
4/25 |
Final Project Presentations |
|
4/30 |
Final Project Presentations |
Final Project Due May 7th |
smara [who is at] columbia [dot] edu
|