Personalized search by tag-based user profile and resource profile in collaborative tagging systems

Cai, Yi; Li, Qing

doi:10.1145/1871437.1871561

Cited by 67 publications

(76 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These scores are the result of an indexing process, which invoke a query and a collection of resources. The use of these methods has shown their utility and robustness in the information retrieval [19].…”

Section: Interest Detection From Resourcesmentioning

confidence: 99%

Producing relevant interests from social networks by mining users' tagging behaviour: A first step towards adapting social information

Mezghani

Pninou

Zayani

et al. 2017

Data & Knowledge Engineering

View full text Add to dashboard Cite

Section: Interest Detection From Resourcesmentioning

confidence: 99%

Producing relevant interests from social networks by mining users' tagging behaviour: A first step towards adapting social information

Mezghani

Pninou

Zayani

et al. 2017

Data & Knowledge Engineering

View full text Add to dashboard Cite

“…Bouadjenek et al [4] presented an enhanced document representation based on user relationships to re-rank documents on a social platform. Cai et al [17] treated the relevance as fuzzy satisfaction between users and queries. However, if the relevant documents cannot be returned in the retrieval list, the strategy has no way to fetch more relevant results.…”

Section: Related Workmentioning

confidence: 99%

An Augmented User Model for Personalized Search in Collaborative Social Tagging Systems

Zhao¹,

Zhou²,

Wu³

et al. 2017

EAI Endorsed Transactions on Collaborative Computing

View full text Add to dashboard Cite

Alongside the enormous volume of user-generated content posted to World Wide Web, there exists a thriving demand for search personalization services. To provide personalized services, a user model is usually required. We address the setting adopted by the majority of previous work, where a user model consists solely of the user's past information. We construct an augmented user model from a number of tags and documents. These resources are further processed according to the user's past information by exploring external knowledge base. A novel generative model is proposed for user model generation. This model utilizes recent advances in neural language models such as Word Embeddings with latent semantic models such as Latent Dirichlet Allocation. We further present a new query expansion method to facilitate the desired personalized retrieval. Experiments conducted on two real-world collaborative social tagging datasets show that our proposed methods outperform state-of-the-art methods.

show abstract

“…The first step of the IR personalization aims at modeling the user's profile and social context [1], [3], [18], [20], [17]. Indeed, several studies have proven that user's ISC can be effectively harvested from the social bookmarking systems [1], [17].…”

Section: Related Workmentioning

confidence: 99%

“…Indeed, several studies have proven that user's ISC can be effectively harvested from the social bookmarking systems [1], [17]. These works assume that the documents and the tags posted by users depend highly on their interests and provide rich information for building user profiles [3], [18], [20].…”

Section: Related Workmentioning

confidence: 99%

“…Authors in [3], [18], [20], [11], propose to personalize the user's search by ranking the resource based on, a matching between the user's interests and the documents' topics [20], or between the user's profile and the resource's profile [18], [3]. Rather than considering the resource's content, authors in [3], [18], [11] propose to build a resource's profile through the resource's annotations and compute a matching function between this resource's profile and the query terms, on the one hand and between the resource's profile and the user's profile, on the other hand.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

User-Centered Social Information Retrieval Model Exploiting Annotations and Social Relationships

Bouhini

Géry

Largeron

2013

Information Retrieval Technology

View full text Add to dashboard Cite

Abstract. Social Information Retrieval (SIR) has extended the classical information retrieval models and systems to take into account social information of the user within his social networks. We assume that a SIR system can exploit the informational social context (ISC) of the user in order to refine his retrieval, since different users may express different information needs as the same query. Hence, we present a SIR model that takes into account the user's social data, such as his annotations and his social relationships through social networks. We propose to integrate the user's ISC into the documents indexing process, allowing the SIR system to personalize the list of documents returned to the user. Our approach has shown interesting results on a test collection built from the social collaborative bookmarking network Delicious.

show abstract

Personalized search by tag-based user profile and resource profile in collaborative tagging systems

Cited by 67 publications

References 17 publications

Producing relevant interests from social networks by mining users' tagging behaviour: A first step towards adapting social information

Producing relevant interests from social networks by mining users' tagging behaviour: A first step towards adapting social information

An Augmented User Model for Personalized Search in Collaborative Social Tagging Systems

User-Centered Social Information Retrieval Model Exploiting Annotations and Social Relationships

Contact Info

Product

Resources

About