The TREC question answering track

Voorhees, Ellen M.

doi:10.1017/s1351324901002789

Cited by 881 publications

(1,145 citation statements)

References 14 publications

Supporting

Mentioning

1,129

Contrasting

Unclassified

Order By: Relevance

“…Factoid : This category contains factoid questions that have a definite answer (which have been the subject of study in many question answering systems, e.g. [14]). Such questions are not related to the agent and are considered out-of-domain.…”

Section: Classification Schemementioning

confidence: 99%

How Do People Talk with a Virtual Philosopher: Log Analysis of a Real-World Application

Wang

Nakatsu

2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Conversation with computers is an important form of human computer interaction. Inappropriately designed conversational agents can easily lead to unsatisfying user experience and even frustration, and this is especially true when the application is deployed in the real world. Currently, research on casual non-task oriented systems and our understanding in how people interact with such agents are still limited. To gain more insights on this issue, we carried out both quantitative and qualitative content analysis of conversation logs collected from a realworld application, featuring a non-task oriented conversational agent as a virtual philosopher. We construct a taxonomy of user utterances to the agent and discuss a few strategies that an agent might employ to provide a better user experience.

show abstract

Section: Classification Schemementioning

confidence: 99%

How Do People Talk with a Virtual Philosopher: Log Analysis of a Real-World Application

Wang

Nakatsu

2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Such results would prove useful for those users looking at the top 5 or 10 documents returned [27]. Moreover, these results would also prove interesting when locating online services [14] or when examining question-answering systems [29] where most of the time, the expected answer is a single URL or a short sentence of only a few words extracted from a document.…”

Section: Test Collectionsmentioning

confidence: 99%

Term Proximity Scoring for Keyword-Based Retrieval Systems

Rasolofo

Savoy

2003

Lecture Notes in Computer Science

119

111

View full text Add to dashboard Cite

Abstract. This paper suggests the use of proximity measurement in combination with the Okapi probabilistic model. First, using the Okapi system, our investigation was carried out in a distributed retrieval framework to calculate the same relevance score as that achieved by a single centralized index. Second, by applying a term-proximity scoring heuristic to the top documents returned by a keyword-based system, our aim is to enhance retrieval performance. Our experiments were conducted using the TREC8, TREC9 and TREC10 test collections, and show that the suggested approach is stable and generally tends to improve retrieval effectiveness especially at the top documents retrieved.

show abstract

“…TREC formalized definitional question answering as extracting and combining answers from multiple documents [1] for questions such as "What are fractals?" or "Who is Andrew Carnegie?"…”

Section: Introductionmentioning

confidence: 99%

Descriptive Question Answering with Answer Type Independent Features

Yoon

Lee

Kim

et al. 2012

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn this paper, we present a supervised learning method to seek out answers to the most frequently asked descriptive questions: reason, method, and definition questions. Most of the previous systems for question answering focus on factoids, lists or definitional questions. However, descriptive questions such as reason questions and method questions are also frequently asked by users. We propose a system for these types of questions. The system conducts an answer search as follows. First, we analyze the user's question and extract search keywords and the expected answer type. Second, information retrieval results are obtained from an existing search engine such as Yahoo or Google. Finally, we rank the results to find snippets containing answers to the questions based on a ranking SVM algorithm. We also propose features to identify snippets containing answers for descriptive questions. The features are adaptable and thus are not dependent on answer type. Experimental results show that the proposed method and features are clearly effective for the task. key words: descriptive question answering

show abstract

The TREC question answering track

Cited by 881 publications

References 14 publications

How Do People Talk with a Virtual Philosopher: Log Analysis of a Real-World Application

How Do People Talk with a Virtual Philosopher: Log Analysis of a Real-World Application

Term Proximity Scoring for Keyword-Based Retrieval Systems

Descriptive Question Answering with Answer Type Independent Features

Contact Info

Product

Resources

About