Probabilistic models for topic detection and tracking

Walls, Frederick; Jin, Hubert; Sista, Sreenivasa; Schwartz, Rich

doi:10.1109/icassp.1999.758177

Cited by 12 publications

(6 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our goal is to recognize stories that are "hot", in the sense that more similar stories follow them than is typically the case. Although similar to the "on-line new event prediction" [4,34] or "first story detection" [2] problem within the Topic Detection and Tracking (TDT) initiative [1,30,31,33,32], the problem we are addressing differs in two important ways. First, we do not require a story to be the very first on a topic, but rather that there are more than a normal amount of subsequent stories that are more similar to the story than is typical.…”

Section: Case Study I: Hot Story Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Intelligent information triage

Macskassy

Provost

2001

Proceedings of the 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

In many applications, large volumes of time-sensitive textual information require triage: rapid, approximate prioritization for subsequent action. In this paper, we explore the use of prospective indications of the importance of a time-sensitive document, for the purpose of producing better document filtering or ranking. By prospective, we mean importance that could be assessed by actions that occur in the future. For example, a news story may be assessed (retrospectively) as being important, based on events that occurred after the story appeared, such as a stock price plummeting or the issuance of many follow-up stories. If a system could anticipate (prospectively) such occurrences, it could provide a timely indication of importance. Clearly, perfect prescience is impossible. However, sometimes there is sufficient correlation between the content of an information item and the events that occur subsequently. We describe a process for creating and evaluating approximate information-triage procedures that are based on prospective indications. Unlike many informationretrieval applications for which document labeling is a laborious, manual process, for many prospective criteria it is possible to build very large, labeled, training corpora automatically. Such corpora can be used to train text classification procedures that will predict the (prospective) importance of each document. This paper illustrates the process with two case studies, demonstrating the ability to predict whether a news story will be followed by many, very similar news stories, and also whether the stock price of one or more companies associated with a news story will move significantly following the appearance of that story. We conclude by discussing how the comprehensibility of the learned classifiers can be critical to success.

show abstract

Section: Case Study I: Hot Story Detectionmentioning

confidence: 99%

“…DET curves[19,3] are used in a similar fashion in the TDT initiative[1,2,3,4,17,30,31,33,32], and are isomorphic to ROC curves, differing primarily on rescaling the axes 5. We used the versions of these learners found in the publicly available Rainbow package[20].…”

mentioning

confidence: 99%

Intelligent information triage

Macskassy

Provost

2001

Proceedings of the 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

show abstract

“…2) Probabilistic Model [9]: Topic detection based on probabilistic models which has been used for news and media.…”

Section: Fig 1: Text Mining Processmentioning

confidence: 99%

Users’ Topic Detection from Tweets based on Keyword Extraction

Kumar¹,

Talebi²,

Manoj³

2017

IJCA

View full text Add to dashboard Cite

In this paper, a different approach for detecting users' topic of interest in twitter based on keyword extraction methods and neural network has been shown. An approach to Text Mining is proposed by extracting the topics relevant to some keywords and further used in predicting topics from users' tweets (Twitter posts). The TF-IDF method has been used to extract keywords in this work. The proposed method, which uses neural network, has been shown to be efficient for topic detection and further comparison. Back propagation method is used to train and to learn the neural network.

show abstract

“…Document summarization has been widely explored in natural language processing and information retrieval. Extraction-based methods and abstraction-based methods [5] [14]are the main methods in this area. Recently, the graph-ranking based methods, TextRank [6] as an example, have been proposed for document summarization.…”

Section: Related Workmentioning

confidence: 99%

Describing Web Topics Meticulously through Word Graph Analysis

Sun

Shi

Kong

et al. 2009

2009 Ninth IEEE International Conference on Computer and Information Technology

View full text Add to dashboard Cite

Topic description is as important as topic detection. In this paper, we propose a novel method to describe Web topics with topic words. Under the assumption that representative words exist in important sentences and have high probability of occurrence with other representative words, two graphs are built, one of which represents the relationship for sentences, the other for words. Considering a topic cluster contains a set of different Web pages, sentence clusters are also introduced. Experimental results on a real data set show that our method achieves excellent performance in both high precision and efficiency, especially when real Web data contain mass of noises.

show abstract

Probabilistic models for topic detection and tracking

Cited by 12 publications

References 2 publications

Intelligent information triage

Intelligent information triage

Users’ Topic Detection from Tweets based on Keyword Extraction

Describing Web Topics Meticulously through Word Graph Analysis

Contact Info

Product

Resources

About