Time-Sensitive Language Modelling for Online Term Recurrence Prediction

Zhang, Dell; Lu, Jinsong; Mao, Robert; Nie, Jian‐Yun

doi:10.1007/978-3-642-04417-5_12

Cited by 6 publications

(3 citation statements)

References 21 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The exponential decay model (that is a special case of our model) has been extensively used in many branches of natural science (fluid dynamics, radioactivity), social science (finance), and computer science (routing protocol). Indeed, in information retrieval, Zhang et al [2009] use exponential decay kernel to predict what term is going to recur next in an online stream, given the term occurrence history. They propose a time sensitive language modeling approach that uses the term frequency and the term recency information.…”

Section: Model Derivationmentioning

confidence: 99%

Parameterized Decay Model for Information Retrieval

Paik

2016

ACM Trans. Intell. Syst. Technol.

View full text Add to dashboard Cite

This article proposes a term weighting scheme for measuring query-document similarity that attempts to explicitly model the dependency between separate occurrences of a term in a document. The assumption is that, if a term appears once in a document, it is more likely to appear again in the same document. Thus, as the term appears again and again, the information content of the subsequent occurrences decreases gradually, since they are more predictable. We introduce a parameterized decay function to model this assumption, where the initial contribution of the term can be determined using any reasonable term discrimination factor. The effectiveness of the proposed model is evaluated on a number of recent web test collections of varying nature. The experimental results show that the proposed model significantly outperforms a number of well known retrieval models including a recently proposed strong Term Frequency and Inverse Document Frequency (TF-IDF) model.

show abstract

Section: Model Derivationmentioning

confidence: 99%

Parameterized Decay Model for Information Retrieval

Paik

2016

ACM Trans. Intell. Syst. Technol.

View full text Add to dashboard Cite

show abstract

“…It is possible to take this into account by learning a real number of semantic thumb-ups for each graded rating from the user clickthrough data etc. Furthermore, our approach can also be easily extended to take the ageing of user-ratings into account without affecting the computational efficiency through Time-Sensitive Language Modelling [10] techniques.…”

Section: Generalisationsmentioning

confidence: 99%

How to Count Thumb-Ups and Thumb-Downs: User-Rating Based Ranking of Items from an Axiomatic Perspective

Zhang

Mao

et al. 2011

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. It is a common practice among Web 2.0 services to allow users to rate items on their sites. In this paper, we first point out the flaws of the popular methods for user-rating based ranking of items, and then argue that two well-known Information Retrieval (IR) techniques, namely the Probability Ranking Principle and Statistical Language Modelling, provide simple but effective solutions to this problem. Furthermore, we examine the existing and proposed methods in an axiomatic framework, and prove that only the score functions given by the Dirichlet Prior smoothing method as well as its special cases can satisfy both of the two axioms borrowed from economics.

show abstract

“…For example, a 3star rating in the 5-star scale system can be regarded as 3 thumb-ups and 5-3=2 thumb-downs. It can also be easily extended to take the ageing of user-ratings into account through Time-Sensitive Language Modelling techniques [5].…”

Section: Proposed Approachmentioning

confidence: 99%

How to count thumb-ups and thumb-downs?

Zhang

Mao

et al. 2011

Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

View full text Add to dashboard Cite

It is a common practice among Web 2.0 services to allow users to rate items on their sites. In this paper, we first point out the flaws of the popular methods for user-rating based ranking of items, and then argue that two well-known Information Retrieval (IR) techniques, namely the Probability Ranking Principle and Statistical Language Modelling, provide a simple but effective solution to this problem.

show abstract

Time-Sensitive Language Modelling for Online Term Recurrence Prediction

Cited by 6 publications

References 21 publications

Parameterized Decay Model for Information Retrieval

Parameterized Decay Model for Information Retrieval

How to Count Thumb-Ups and Thumb-Downs: User-Rating Based Ranking of Items from an Axiomatic Perspective

How to count thumb-ups and thumb-downs?

Contact Info

Product

Resources

About