CMU Information Retrieval Discussion Series

Le Zhao and Jonathan Elsas -- Wednesday October 13th

2010-10-07T06:44:00.000-07:00

Please join us for two upcoming IR series talks on Wednesday, Oct 13, 2010.

Lunch will be provided by Yahoo!.

Date/Time: Wednesday, Oct 13, 2010, noon
Place: GHC 4405

First Speaker: Le Zhao
Title: Term Necessity Prediction
Abstract:

The probability that a term appears in relevant documents (P(t|R)) is a fundamental quantity in several probabilistic retrieval models, however it is difficult to estimate without relevance judgments or a relevance model. We call this value term necessity because it measures the percentage of relevant documents retrieved by the term – how necessary a term’s occurrence is to document relevance. Prior research typically either set this probability to a constant, or estimated it based on the term's inverse document frequency, neither of which was very effective.

This paper identifies several factors that affect term necessity, for example, a term’s topic centrality, synonymy and abstractness. It develops term- and query-dependent features for each factor that enable supervised learning of a predictive model of term necessity from training data. Experiments with two popular retrieval models and 6 standard datasets demonstrate that using predicted term necessity estimates as user term weights for the original query terms leads to
significant improvements in retrieval accuracy.

The paper will appear in CIKM 2010.

----------------------------------------------------

Second Speaker: Jon Elsas
Title: Rank Learning for Factoid Question Answering with Linguistic and Semantic Constraints
Abstract:

This work presents a general rank-learning framework for leveraging deep linguistic and semantic features for passage ranking within Question Answering (QA) systems. The passage ranking framework enables query-time checking of these complex and long-distance constraints among question features such as keywords and named entities. These constraints can include keyword ordering, annotation type-checking, verb-argument attachment and arbitrary long-distance paths through an annotation graph. We show that a trained ranking model using this rich feature set achieves greater than a 20% improvement in Mean Average Precision over baseline keyword retrieval models. We also show that for questions expressing the most complex linguistic semantic constraints, further gains in MAP are realized, yielding a 40% improvement over the baseline.

The paper will appear in CIKM 2010.

Jaime Arguello -- Friday July 16th

2010-07-13T09:00:00.000-07:00

Time: Noon
Location: TBA

Speaker: Jaime Arguello, Language Technologies Institute, School of Computer Science, CMU

Title: Vertical Selection in the Presence of Unlabeled Verticals.

Abstract:

Vertical aggregation is the task of incorporating results from specialized search engines or verticals (e.g., images, video, news) into Web search results. Vertical selection is the subtask of deciding, given a query, which verticals, if any, are relevant. State of the art approaches use machine learned models to predict which verticals are relevant to a query. When trained using a large set of labeled data, a machine learned vertical selection model outperforms baselines which require no training data. Unfortunately, whenever a new vertical is introduced, a costly new set of editorial data must be gathered. In this paper, we propose methods for reusing training data from a set of existing (source) verticals to learn a predictive model for a new (target) vertical. We study methods for learning robust, portable, and adaptive cross-vertical models. Experiments show the need to focus on different types of features when maximizing portability (the ability for a single model to make accurate predictions across multiple verticals) than when maximizing adaptability (the ability for a single model to make accurate predictions for a specific vertical). We demonstrate the efficacy of our methods through extensive experimentation for 11 verticals.

This is joint work with Fernando Diaz and Jean-Francois Paiement from Yahoo! Labs and will be presented at SIGIR 2010.

Grace Yang -- Friday July 9th

2010-07-06T10:47:00.000-07:00

Time: Noon
Location: GHC 4405

Speaker: Grace Hui Yang, Language Technologies Institute, School of Computer Science, CMU

Title: Collecting High Quality Overlapping Labels at Low Cost.

Abstract:
This paper studies quality of human labels used to train search engines’ rankers. Our specific focus is performance improvements obtained by using overlapping relevance labels, which is by collecting multiple human judgments for each training sample. The paper explores whether, when, and for which samples one should obtain overlapping training labels, as well as how many labels per sample are needed. The proposed selective labeling scheme collects additional labels only for a subset of training samples, specifically for those that are labeled relevant by a judge. Our experiments show that this labeling scheme improves the NDCG of two Web search rankers on several real-world test sets, with a low labeling overhead of around 1.4 labels per sample. This labeling scheme also outperforms several methods of using overlapping labels, such as simple k-overlap, majority vote, the highest labels, etc. Finally, the paper presents a study of how many overlapping labels are needed to get the best improvement in retrieval accuracy.

This paper is published in Proceedings of the 33th Annual ACM SIGIR Conference (SIGIR2010), Geneva, Switzerland, July 19-23, 2010.

Lee Jensen (Ancestry.com) -- Thursday, June 10

2010-05-28T07:53:00.000-07:00

Speaker: Lee Jensen (Ancestry.com)

Date/Time: Thursday, June 10, 2010, noon

Place: GHC 6115

Title: Exploiting Sequential Relationships for Familial Classification

Abstract:
Information hidden in sequences of related data provides a significant exploitable source for information extraction. In this work we demonstrate techniques for increasing the signal of classifications from sequences of dependent data. This includes using algorithms, as well as generating features, that take advantage of the sequence oriented nature of the data. The efficacy of these techniques is evaluated with the classification of familial relationships within United States census data. The census data proves an interesting corpus for this work because of the highly sequential nature of the instances, and the explicit classification relationship between one instance and an instance previously found in the sequence.

Vasudeva Varma -- Friday April 16, 2010, noon

2010-04-08T12:35:00.000-07:00

Speaker: Dr. Vasudeva Varma (Institute of Information Technology, Hyderabad, India)

Date/Time: Friday April 16, 2010, noon

Place: Newell-Simon Hall 3002

Title: Cross Language Information Access - An Indian Experience

In this talk, I briefly explain the cross language information retrieval research for Indian languages. There is a national wide mission mode project on cross language information access sponsored by Ministry of Communications and Information technology, Government of India involving ten Universities. I will talk about the major research issues and a possible roadmap for the CLIR research in Indian setting. I will also share our experiences with multi-lingual summarization.

Vasudeva Varma is a faculty member at International Institute of Information Technology, Hyderabad Since 2002. His research interests include search (information retrieval), information extraction, information access, knowledge management, cloud computing and software engineering. He is heading Search and Information Extraction Lab and Software Engineering Research Lab at IIIT Hyderabad. He is also the chair of Post Graduate Programs since 2009. He published a book on Software Architecture (Pearson Education) and over sixty technical papers in journals and conferences. In 2004, he obtained young scientist award and grant from Department of Science and Technology, Government of India, for his proposal on personalized search engines. In 2007, he was given Research Faculty Award by AOL Labs.

Jaime Arguello -- Thursday July 16th, 2009, 11:00 AM

2009-07-13T11:51:00.000-07:00

Speaker: Jaime Arguello (Language Technologies Institute, School of Computer Science, Carnegie Mellon University)

Time/Date: Thursday July 16th, 2009, 11:00 AM.

Place: Wean Hall 7220

Title: Sources of Evidence for Vertical Selection

Web search providers often include search services for domain-specific subcollections, called verticals, such as news, images, videos, job postings, company summaries, and artist profiles. We address the problem of vertical selection, predicting relevant verticals (if any) for queries issued to a search engine's main web search page. In contrast to prior collection selection tasks, vertical selection is associated with unique resources that can inform the classification decision. We focus on three sources of evidence: (1) the query string, from which features are derived independent of external resources, (2) logs of queries previously issued to the vertical directly by users, and (3) corpora representative of vertical content. These sources of evidence are integrated as features in a classification-based approach. We make use of and compare against prior work in federated search and retrieval effectiveness prediction. Our evaluation focuses on 18 different verticals, which differ in terms of semantics, media type, size, and level of query traffic. An in-depth error analysis reveals unique challenges across different verticals and provides insight into vertical selection for future work.

Based on work conducted at Yahoo! Labs Montreal to be presented at SIGIR 2009.

Kai-min Kevin Chang -- Friday May 29, 2009, noon

2009-05-26T08:35:00.000-07:00

Speaker: Kai-min Kevin Chang (Language Technologies Institute, School of Computer Science, Carnegie Mellon University)

Time/Date: Friday May 29, 2009, noon.

Place: NSH 1507 (Note the room is not the usual NSH 3002)

Title: Quantitative modeling of the neural representation of adjective-noun phrases to account for fMRI activation

Abstract: Recent advances in functional Magnetic Resonance Imaging (fMRI) offer a significant new approach to studying semantic representations in humans by making it possible to directly observe brain activity while people comprehend words and sentences. In this study, we investigate how humans comprehend adjective-noun phrases (e.g. strong dog) while their neural activity is recorded. Classification analysis shows that the distributed pattern of neural activity contains sufficient signal to decode differences among phrases. Furthermore, vector-based semantic models can explain a significant portion of systematic variance in the observed neural activity. Multiplicative composition models of the two-word phrase outperform additive models, consistent with the assumption that people use adjectives to modify the meaning of the noun, rather than conjoining the meaning of the adjective and noun.

This talk is based on the author's ACL 2009 paper.

Hua Ai -- Friday May 15, 2009, noon

2009-05-11T11:06:00.001-07:00

Speaker: Hua Ai (Intelligent Systems Program at University of Pittsburgh)

Date/Time: Friday May 15, 2009, noon
Location: 3002 Newell-Simon Hall (NSH)

User Simulation for Spoken Dialog System Development

Abstract:
In this talk, I will present my thesis study on investigating how to
evaluate and how to build user simulations to help dialog system
development. When evaluating user simulations, I use both human judges and
automatic evaluation measures to assess the simulation model qualities.
When building user simulations, I examine three factors that impact
simulation models in the tasks of dialog strategy learning and dialog
system development.

The talk is based on the author's ACL 2009 paper.

Rosie Jones - Thursday October 9th, 2008

2008-10-03T07:00:00.000-07:00

3002 Newell-Simon Hall
Thursday, October 9, 2008
11:00am-12pm

Speaker : Rosie Jones

Title: Web Search Sessions

Abstract: Traditionally, information retrieval examines the search query in isolation: a query is used to retrieve documents, and the relevance of the documents returned are evaluated in relation to that query. However, users typically conduct web and other types of searches in sessions, issuing a query, examining results, and the re-issuing a modified query to improve the results. We decribe the properties of real web search sessions, and show that users conduct searches for both broad and finer grained tasks, which can be both interleaved and nested. We show that user search reformulations can be mined to identify related terms, and that we can identify the boundaries between tasks with greater accuracy than previous methods.

Bio:
Rosie Jones is a Senior Research Scientist at Yahoo!. Her research interests include web search, geographic information retrieval, and natural language processing. She received her PhD from the Language Technologies Institute at Carnegie Mellon University under the supervision of Tom Mitchell, where her doctoral thesis was titled Learning to Extract Entities from Labeled and Unlabeled Text. She is co-organizing the WSDM 2009 Workshop on Web Search Click Data (WSCD09). She served on the Senior PC for SIGIR in 2007 and 2008, and is a Senior Member of the ACM.

Le Zhao - Thursday 2nd Oct 2008

2008-10-02T07:19:00.000-07:00

We are going to have Le Zhao to give our first IR talk in this
semester. Reception will provided by Yahoo!. Here is the talk information:

Date: Thursday 2nd Oct 2008
Time: 2pm
Place: Wean Hall 7220

Speaker: Le Zhao
Title: A Generative Retrieval Model for Structured Documents

Abstract
Structured documents contain elements defined by the author(s) and annotations assigned by other people or processes. Structured documents pose challenges for probabilistic retrieval models when there are mismatches between the structured query and the actual structure in a relevant document or erroneous structure introduced by an annotator. This paper makes three contributions. First, a new generative retrieval model is proposed to deal with the mismatch problem. This new model extends the basic keyword language model by treating structure as hidden variable during the generation process. Second, variations of the model are compared. Third, term-level and structure-level smoothing strategies are studied. Evaluation was conducted with INEX XML retrieval and question-answering retrieval tasks. Experimental results indicate that the optimal structured retrieval model is task dependent, two-level Dirichlet smoothing significantly outperforms two-level Jelinek-Mercer smoothing, and with accurate structured queries, the proposed structured retrieval model outperforms keyword retrieval significantly, on both QA and INEX datasets.

Based on work accepted at CIKM'08.

Justin Betteridge - Friday May 23rd

2008-05-16T13:30:00.000-07:00

Please join us for an upcoming talk.

Lunch will be provided by Yahoo!

Title:
Linguistic Pattern Learning for Web Information Extraction

Who: Justin Betteridge
When: Friday, May 23rd, 12:00pm
Where: NSH 3002

Abstract:
Most approaches to automatically extracting structured information from the web
rely on surface text patterns. However, the manner in which such patterns are
defined, learned, and employed in the larger system varies with each case. In
this talk, I will outline the spectrum of previous work in this area and argue
for a linguistically-motivated definition, a hybrid heuristic/classifier-based
assessment, and a multi-purpose employment of textual patterns in the context of
Web Information Extraction (WIE). I will also give preliminary results from
adopting such an approach in our WIE system.

Grace, Hui Yang - Friday May 16th

2008-05-07T10:20:00.001-07:00

Please join us for an upcoming talk.

Lunch will be provided by Yahoo!

Title:
Ontology Learning by Supervised Hierarchical Clustering

Who: Grace, Hui Yang
When: Friday, May 16th, 12:00pm
Where: NSH 3002

Abstract:
This work makes novel use of supervised clustering as the basic
framework to construct concept ontology interactively or
automatically. Supervised hierarchical clustering is used to
organize ontology fragments, which are identified by techniques in
natural language processing and information retrieval, into
hierarchies. At each clustering iteration, a distance metric is
learned from the clustering given by either pseudo or real
feedback. K-medoids clustering with sampling is then used to group
the concepts at the higher level. A web-based cluster naming
algorithm is also presented. By conducting a user evaluation, the
system is shown to be effective to save human efforts in the
interactive runs. Both automatic and interactive runs of the
experiments show that the approach is effective.

Nico Schlaefer - Friday, April 4, 12:00pm, NSH 3002

2008-03-28T19:24:00.000-07:00

Please join us for an upcoming talk from Nico Schlaefer.

Lunch will be provided!

Title:
The Ephyra Question Answering System: Recent Results and Current Directions

Who: Nico Schlaefer
When: Friday, April 4, 12:00pm
Where: NSH 3002

Abstract:
This talk gives an overview of recent work on English question answering (QA) at CMU and our participation in last year’s TREC evaluation. QA is the task of retrieving accurate answers to natural language questions from a knowledge source such as the Web. The presentation includes a brief introduction to QA and the TREC competition, thus prior knowledge on QA is not required though helpful.

The talk focuses on the challenges that an end-to-end QA system needs to address, and the architectural and algorithmic solutions implemented in Ephyra, our English QA system. Ephyra is a modular and extensible framework that facilitates the integration of different QA techniques. The system is organized as a pipeline of reusable standard components for question analysis, query generation, search, answer extraction, and answer selection. The most recent setup combines a syntactic pattern learning and matching approach with answer-type based extraction techniques and a semantic answer extractor that is based on semantic role labeling.

Recently we have placed the Ephyra QA system into open source, making most of our code available to the research community. I will discuss why we took this step, and how you may benefit from our open source system - OpenEphyra - for your own research.

Upcoming IR Talk at CMU: John Tait

2008-02-20T18:04:00.000-08:00

The upcoming IR-related LTI Seminar talk, John Tait on Patent Retrieval.

Jan Wiebe -- Subjectivity Analysis -- Friday, Feburary 22nd 2008, 12:00 pm (noon)

2008-02-18T07:10:00.000-08:00

Please join us for our first IR Series talk this spring!

Lunch will be provided by Yahoo!

Speaker: Jan Wiebe
Professor, Department of Computer Science
Director, Intelligent Systems Program
University of Pittsburgh

Date/Time: Friday, 22nd, 12:00 pm (noon)

Location: 3002 Newell-Simon Hall (NSH)

Title: Subjectivity Analysis

Abstract: A growing area of research, "subjectivity analysis", is the computational study of affect, opinions, and sentiments expressed in text. Blogs, editorials, reviews (of products, movies, books, etc.), and even "objective" newspaper articles (which include many opinions and sentiments) are just some of the genres for which accurate identification and interpretation of opinions is critical for full text understanding. Subjectivity analysis will support developing tools for information analysts in governmental, commercial, and political domains who want to automatically track attitudes and feelings in the news and on-line forums. How do people feel about the latest iPod? Is there a change in the support for the new Medicare bill? A system able to automatically identify and extract opinions and sentiments from text would be an enormous help to someone sifting through the vast amounts of news and web data, trying to answer these kinds of questions. In this talk, I will first give an overview of our work in subjectivity analysis, and then will focus on experiments exploring interactions between subjectivity and word sense, showing that subjectivity is a property that can be associated with word meanings and that subjectivity classification can be beneficial for word sense disambiguation.

Bio: My research areas are artificial intelligence and natural language processing (NLP). My work with students and colleagues has been in discourse processing, pragmatics, word-sense disambiguation, and probabilistic classification in NLP. Our most recent work investigates automatically recognizing and interpretating expressions of opinions and sentiments in text, to support NLP applications such as question answering, information extraction, text categorization, and summarization.

New Home

2008-02-18T06:41:00.001-08:00

Welcome to the new home of the CMU Information Retrieval Discussion Series. We're in the process of moving the old site here, so please be patient.

Past IR-Series Presentations

2008-01-01T12:00:00.000-08:00

Friday, November 2, 2007 - 12:00-1:00 pm, Newell-Simon Hall (NSH) 3002
Title: CMU at TREC 2007
Speakers: Jonathan Elsas, Le Zhao and Yangbo Zhu (CMU)

Friday, October 5, 2007 - 12:00-1:00 pm, Newell-Simon Hall (NSH) 3002
Title: Estimating and Exploiting Uncertainty in Pseudo-Relevance Feedback
Speakers: Kevyn Collins-Thompson (CMU)

Friday, July 13, 2007 - 12:00-1:00 pm, Newell-Simon Hall (NSH) 3002
Title: Utility-based Information Distillation Over Temporally Sequenced Documents
Speakers: Yiming Yang (CMU)

Friday, May 18, 2007 - 12:00-1:00 pm, Newell-Simon Hall (NSH) 3002
Title: Collaborative Web Search - Exploiting User Activity for User Benefit
Speaker: Jill Freyne (University College Dublin)
Details

Friday, January 19, 2007 - 12:00 NSH 3002
Title: Using Graphs and Random Walks to Discover Latent Similarities in Text
Speaker: Gunes Erkan
Details

Friday, November 10, 2006, 2007 - 12:00 NSH 3002
Title: Personal Metasearch
Speaker: Paul Thomas
Details

Friday, May 19, 2006 - 12:00 NSH 3002
Title: Collaborative Adaptive User Profile with Implicit and Explicit User Feedback
Speaker: Yi Zhang
Details

Wednesday, April 19, 2006 - 12:00, NSH 3002
Title: Deriving Marketing Intelligence from Online Discussion
Speaker: Matthew Hurst and Natalie Glance
Details

Wednesday, April 5, 2006 - 12:00, NSH 3002
Title: A Graphical Framework for Contextual Search and Name Disambiguation in Email
Speaker: Einat Minkov
Details

Wednesday, March 8, 2006 - 12:00, NSH 3002
Title: Structured and Dynamic Topic Models
Speaker: John Lafferty
Details

Wednesday, February 22, 2006 - 12:00, NSH 3002
Title: Automatically Labeling Hierarchical Clusters
Speaker: Pucktada (Puck) Treeratpituk
Details

Friday, June 3, 2005 - 3:30, WeH 5409
Title: PageRank without Hyperlinks: Structural Re-ranking using Links Induced by Language Models
Speaker: Oren Kurland
Details

Wednesday, April 27, 2005 - 4:30, WeH 4601
Title: Dynamic Construction of Content-Based Topologies in Hierarchical Peer-to-Peer Networks
Speaker: Jie Lu
Details

Wednesday, March 16, 2005 - 4:30, WeH 4601
Title: Modeling Search Engine Effectiveness for Federated Search
Speaker: Luo Si
Details

Wednesday, March 2, 2005 - 4:30, WeH 4623
Title: What is the matter? Explorations in text categorization
Speaker: Lillian Lee
Details

Wednesday, January 19th, 2005 - 4:30, WeH 4601
Title: Detecting Action-Items in E-mail
Speaker: Paul N. Bennett
Details

Wednesday, December 1, 2004 - 3:00, WeH 4625
Title: Probabilistic Models of Text and Images
Speaker: David Blei
Details

Wednesday, November 17, 2004 - 3:00, WeH 4625
Title: Merging Rank Lists from Multiple Sources in Video Classification
Speaker: Wei-Hao Lin
Details

Wednesday, November 10, 2004 - 3:00, WeH 4625
Title: Associating Names with Persons in Broadcast News Video
Speaker: Jun Yang
Details

Wednesday, October 20, 2004 - 3:00, WeH 4625
Title: Graph Mining
Speaker: Christos Faloutsos
Details

Wednesday, October 6, 2004 - 3:00, WeH 4625
Topic: Review of the SIGIR 2004 Best Paper, “ A Formal Study of Information Retrieval Heuristics” by Hui Fang, Tao Tao, and ChengXiang Zhai
Speaker: Kevyn Collins-Thompson
Details

Friday, October 1, 2004 - 1:30, NSH 4513
Title: Combining Language Modeling Approach with String-matching in Near-Duplicate Detection in E-Rulemaking
Speaker: Puck Treeratpituk
Details

Wednesday, September 22, 2004 - 2:30, NSH 4632
Learning to Summarize Interviews for Project Reports
Nikesh Garera
Details

Thursday, August 26, 2004 - 3:30, WeH 4625
Analyzing Time Series Gene Expression Data
Jason Ernst
Details

Tuesday, August 17, 2004 - 2:00, WeH 4625
Learning Table Extraction from Examples
Ashwin Tengli
Details

Thursday, August 12, 2004 - 3:30, WeH 4625
Learning to Classify Email into "Speech Acts"
Vitor Carvalho
Details

Thursday, July 8, 2004 - 3:30, WeH 4625
Resource Selection for Domain-Specific Cross-Lingual IR
Monica Rogati
Details

Tuesday, January 22, 2004 - 12:00, NSH 4513
Dynamic Recommender System on User Taste Tendency Model
Soojung Lee
Details

Thursday, December 4, 2003 - 12:00, NSH 4513
The Robustness of Content-Based Search in Hierarchical Peer to Peer Networks
M. Elena Renda
Details

Thursday, October 30, 2003 - 12:00, NSH 4632
Boosting Support Vector Machines for Text Classification through Parameter-free Threshold Relaxation
Dr. James G. Shanahan
Details

Thursday, October 23, 2003 - 12:00, NSH 4632
Content-Based Retrieval in Hybrid Peer-to-Peer Networks
Jie Lu
Details

Thursday, October 16, 2003 - 12:00, NSH 4632
The Utility of Question Analysis in an Open-Domain Question Answering System
Yifen Huang
Details

Thursday, August 28, 2003 - 3:30, NSH 4632
Searching Peer-to-Peer Networks
Dr. Bin Yu
Details

Thursday, August 14, 2003 - 3:30, NSH 3001
Flexible Mixture Model for Collaborative Filtering
Luo Si

Modified Logistic Regression: An Approximation to SVM and its Applications in Large-Scale Text Categorization
Jian Zhang
Details

Thursday, June 19, 2003 - 3:30, NSH 3001
Improving Text Classifier Probability Estimates
Paul Bennett
Details

Thursday, June 5, 2003 - 3:30, NSH 3002
Radio Station Playlist Generation
Andrew P. Widdowson
Details

Thursday, May 22, 2003 - 3:30, NSH 3001
Discussion on Secondary Structure Prediction for Protein Sequences
Yan Liu
Details

Thursday, May 8, 2003 - 3:30, NSH 3001
Negative Pseudo Relevance Feedback for Multimedia Retrieval
Rong Yan
Details

Thursday, April 10, 2003 - 3:30, NSH 3001
Web Image Retrieval Re-Ranking with Relevance Model
Wei-Hao Lin
Details

Thursday, March 27, 2003 - 3:30, NSH 3001
Clustering Genes
Fan Li
Details

Thursday, March 13, 2003 - 3:30, NSH 3001
Exploration and Exploitation in Adaptive Filtering Based on Bayesian Active Learning
Yi Zhang
Details

Thursday, February 27, 2003 - 3:30, NSH 3001
Beyond Independent Topical Relevance: Evaluation Metrics and Methods for Aspect Retrieval
Dr. William Cohen
Details

Thursday, February 13, 2003 - 3:30, NSH 3001
Overview of Database Selection Methods
Luo Si
Details

Tuesday, January 14, 2003 - 11:00-12:30, Wean 4632
Topics and Techniques in (Structured) Document Retrieval
Paul Ogilvie
Details

Instructions for Presenters

2008-01-01T00:00:00.001-08:00

As the discussion series has progressed, our goals for the presentations have evolved. In addition to generating discussion among researchers, we have come to recognize the value of the series as a means to refine our presentation skills. In light of this, here are some guidelines for creating your presentation:

When preparing your presentation, view this as a normal conference talk and prepare accordingly.
Please prepare a short (20-30 minute) talk or a long (45 minute) talk according to the time slot the organizer has reserved.
You should assume that the audience is knowledgeable in IR and many of the techniques commonly used in the field. Unless the purpose of your talk is a general overview of a research problem, you should assume that the related research can be covered very briefly (one or two slides).
Focus on presenting your thoughts, issues, and contributions to the problem at hand.
If you have extra material that won't fit in the talk, prepare slides for them as it is very likely that we will be willing to hear more about the subject after the main talk is over.
Please don't be afraid to present work in progress. Even with the change of presentation format to conference talk style , we are still driven by our original goals of learning about current research and fostering collaboration on work in progress.

Thanks are due to Yiming Yang for some helpful suggestions.

About the Series

2008-01-01T00:00:00.000-08:00

The CMU-IR Discussion Series is a student-run initiative to encourage discussion and foster research between the multiple Information Retrieval research groups here at Carnegie Mellon. We intend the informal presentations to be a vehicle for:

learning about each others' research,
discussing the big (and little) problems of a research area, and
fostering collaboration across groups.

The presentations in this series are of an informal nature. Interruptions, questions, and discussions on tangents are encouraged and welcomed. We wish this to be a low-stress, relaxed opportunity for our presenters to discuss topics of interest with us. Our interpretation of the meaning of "information retrieval" is fairly broad; we welcome discussions of related fields (Machine Learning, Agents, Statistics, and so on). While most of the presentations are from students researching Information Retrieval here at CMU, we also welcome discussions from students and faculty in other programs and institutions as well as industry professionals.

One of the goals of the series is to strike a nice balance between area overview presentations and technical presentations on specific approaches. Because many of us work on quite different areas of Information Retrieval, we often find it beneficial to have discussions that focus on the important problems in our respective research areas and the techniques that have been found to be broadly useful (and occasionally the spectacular failures). In order to keep grounded, we also have some technical discussions on specific techniques and approaches.

In an effort to make this series a valuable resource for others, we plan on posting the authors' slides (with permission). We also ask that authors provide a short reading list of articles (preferably online) for people who want to learn about the topics in more depth.

For LTI students: a presentation in the IR Discussion Series can fulfill your annual LTI talk requirement. Let us know if you wish to do this more than a week in advance, so that we can advertise the talk according to policy. You will still be required to make sure two faculty are present and that they fill out the form after your presentation.