Exploring Social Annotation Tags to Enhance Information Retrieval Performance

Zheng, Ye; Huang, Xiangji; Jin, Song; Lin, Hongfei

doi:10.1007/978-3-642-15470-6_27

Cited by 4 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These two hyperparameters control to what extent the LDA model is smoothed. For the settings of hyperparameters α and β, we use a method detailed in the section Parameter Training to train them on the AP88‐90 Ye, Huang, Jin, and Lin (2010) collection with topics 51‐100. In particular, we sweep both α and β over (0.001, 0.005, 0.01, 0.05, 0.1, 0.2, …, 0.9) in order to find optimal settings.…”

Section: A Topic‐based Feedback Modelmentioning

confidence: 99%

See 1 more Smart Citation

Finding a good query‐related topic for boosting pseudo‐relevance feedback

Zheng

Huang

Lin

2011

J. Am. Soc. Inf. Sci.

Self Cite

View full text Add to dashboard Cite

Pseudo-relevance feedback (PRF) via query expansion (QE) assumes that the top-ranked documents from the first-pass retrieval are relevant. The most informative terms in the pseudo-relevant feedback documents are then used to update the original query representation in order to boost the retrieval performance. Most current PRF approaches estimate the importance of the candidate expansion terms based on their statistics on document level. However, a document for PRF may consist of different topics, which may not be all related to the query even if the document is judged relevant. The main argument of this article is the proposal to conduct PRF on a granularity smaller than on the document level. In this article, we propose a topic-based feedback model with three different strategies for finding a good query-related topic based on the Latent Dirichlet Allocation model. The experimental results on four representative TREC collections show that QE based on the derived topic achieves statistically significant improvements over a strong feedback model in the language modeling framework, which updates the query representation based on the top-ranked documents.

show abstract

Section: A Topic‐based Feedback Modelmentioning

confidence: 99%

“…In our experiments, we train this parameter on a small collection, namely the AP88‐90 (Ye et al, 2010) collection with topics 51‐100. In particular, we sweep over the threshold (τ∈0.05,0.1,0.15,0.2,…, 1$) in order to find optimal setting.…”

Section: A Topic‐based Feedback Modelmentioning

confidence: 99%

Finding a good query‐related topic for boosting pseudo‐relevance feedback

Zheng

Huang

Lin

2011

J. Am. Soc. Inf. Sci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Also, these tagging-based searching process and systems have been applied to various online contents, e.g. video [14] and music [15].…”

Section: Tag-based Irmentioning

confidence: 99%

Discovering Community of Lingual Practice for Matching Multilingual Tags from Folksonomies

Jung

2011

The Computer Journal

View full text Add to dashboard Cite

Many existing studies have investigated to discover a variety of co-occurrence patterns between entities (e.g. users, tags and resources) from a folksonomy system. The common purposes among them are (i) to understand collective behaviors between online users and (ii) to provide online services (e.g. tag recommendation and information searching) to users. However, most of the existing studies assume that all tags in the folksonomy should be written in an identical language. In this paper, we focus on analyzing a multilingual folksonomy generated by various lingual practices of online users, and discovering meaningful relationships between multilingual tags (e.g. between 'Seoul' in English and 'Corée'in French) co-occurred in the folksonomy. Thereby, we propose novel methods for (i) identifying lingual practices from user tagging patterns to build a community of lingual practice and (ii) exploiting the tag matchings to extend simple term-based queries. Thus, additional resources tagged by other languages can be retrieved. To evaluate the proposed multilingual tag matching method, we have collected real tagging datasets from several well-known social tagging websites (e.g. Del.icio.us), and applied to translating queries to other languages without any external dictionaries.

show abstract