CIS 630 (Spring 2012: Seminar in Natural Language Processing)
Class: MW 3-4:30 pm, Towne 307
Office Hours: By appointment
The course will focus on topics in discourse, semantics and pragamtics. Many of the readings will be related to the use of the Penn Discourse Treebank (PDTB) in recent research papers. The specific selection of papers will be adjusted to take into account students' interests. A comprehensive list of papers is available here.
30% project 1: defined by instructor. The final project report is due on Feb 27.
40% project 2: on a topic proposed by the student, related to readings and topics covered in class. Final project reports are due on May 1.
10% weekly summary of readings and questions. Students should upload to coursekit a short summary and evaluation of the assigned papers. These summaries are due at midnight before the day of the class (midnight Sunday and Tuesday respectively). Readings will be made avaiable two weeks before class.
20% lecture notes: Each week two students will compile lecture notes which summarize the main discussion in class, key ideas, approaches and formulas that we have covered reflecting the assigned papers.
- Jan 11, Intro to class. Project 1 definition and discussion.
- Jan 18, Intro to discourse
Webber et al, "Discourse structure and language technology"
- Jan 23, Learning sentence importance
Louis et al, Discourse indicators for content selection in summarization
Leskovec et al, Impact of linguistic analysis on the semantic graph coverage and learning of document extracts
- Jan 25, Ranking, regression and co-training
Xie and Liu, Improving supervised learning for meeting summarization using sampling and regression
Wong et al, Extractive summarization using supervised and semi-supervised learning
Xie et al, Semi-supervised extractive speech summarization via co-training algorithm
Lin et al, Leveraging evaluation metric-related training criteria for speech summarization
- Jan 30, Lexical features for implicit discourse relations
Sporleder and Lascarides, Using automatically labelled examples to classify rhetorical relations"
Blair-Goldensohn et al, Building and refining rhetorical-semantic relation models
- Feb 1, Semantic and syntactic features for implicit discourse relations
Pitler et al, Automatic sense prediction for implicit discourse relations in text
Lin et al, " Recognizing implicit discourse relations in the PDTB
- Feb 6, PDTB annotations and explicit relations
Miltsakaki et al, Sense annotation in the PDTB
Miltsakaki et al, Experiments on sense annotation and sense disambiguation of discourse connectives"
- Feb 8, Other techniques for connective sense disambiguation
Meyer et al, Multilingual annotation and disambiguation of discourse connectives for machine translation
Schwartz and Gomez, Acquiring knowledge from the web to be used as selectors for noun sense disambiguation
- Feb 13, Textual entailment
Glickman et al, A lexical alignment model for probabilistic textual entailment
LoBue and Yates, Types of common-sense knowledge needed for recognizing textual entailment
- Feb 15, Causal relations
Riaz and Girju, Another look at causality: discovering scenario-specific contingency relationships with no supervision
Do et al, Minimally supervised event causality identification