Evaluating similarity measures for emergent semantics of social tagging

Markines, Benjamin; Cattuto, Ciro; Menczer, Filippo; Benz, Dominik C.; Hotho, Andreas; Stumme, Gerd

doi:10.1145/1526709.1526796

Cited by 219 publications

(191 citation statements)

References 30 publications

Supporting

Mentioning

189

Contrasting

Order By: Relevance

“…In this work, we have not addressed this issue when categorising social tags based on their intention. We plan to study disambiguation strategies that take into account the "context" of a social tag within a user or item profile [21][31]. For example, let us assume that we retrieve the tag "java" from a user/item profile, and we have to decide whether it refers to the well known programming language or to the Indonesian island.…”

Section: Discussionmentioning

confidence: 99%

Categorising social tags to improve folksonomy-based recommendations

Cantador

Konstas

Jose

2011

Journal of Web Semantics

102

View full text Add to dashboard Cite

This is the author’s version of a work that was accepted for publication in Web Semantics: Science, Services and Agents on the World Wide Web. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Web Semantics: Science, Services and Agents on the World Wide Web, 9, 1, (2011) DOI: 10.1016/j.websem.2010.10.001In social tagging systems, users have different purposes when they annotate items. Tags not only depict the content of the annotated items, for example by listing the objects that appear in a photo, or express contextual information about the items, for example by providing the location or the time in which a photo was taken, but also describe subjective qualities and opinions about the items, or can be related to organisational aspects, such as self-references and personal tasks. Current folksonomy-based search and recommendation models exploit the social tag space as a whole to retrieve those items relevant to a tag-based query or user profile, and do not take into consideration the purposes of tags. We hypothesise that a significant percentage of tags are noisy for content retrieval, and believe that the distinction of the personal intentions underlying the tags may be beneficial to improve the accuracy of search and recommendation processes. We present a mechanism to automatically filter and classify raw tags in a set of purpose-oriented categories. Our approach finds the underlying meanings (concepts) of the tags, mapping them to semantic entities belonging to external knowledge bases, namely WordNet and Wikipedia, through the exploitation of ontologies created within the W3C Linking Open Data initiative. The obtained concepts are then transformed into semantic classes that can be uniquely assigned to content- and context-based categories. The identification of subjective and organisational tags is based on natural language processing heuristics. We collected a representative dataset from Flickr social tagging system, and conducted an empirical study to categorise real tagging data, and evaluate whether the resultant tags categories really benefit a recommendation model using the Random Walk with Restarts method. The results show that content- and context-based tags are considered superior to subjective and organisational tags, achieving equivalent performance to using the whole tag space.This research was supported by the European Commission (SALERO, FP6-027122), by the Spanish Ministry of Science and Education (RIM3, TIN2008-06566-C04-02), and by the Spanish Ministry of Industry (i3media, CENIT-2007-1012)

show abstract

Section: Discussionmentioning

confidence: 99%

Categorising social tags to improve folksonomy-based recommendations

Cantador

Konstas

Jose

2011

Journal of Web Semantics

102

View full text Add to dashboard Cite

show abstract

“…The studies from [Markines et al, 2009] and [Cattuto et al, 2008] propose an analysis of the different types of similarity measures and the semantic relations they each tend to convey. The simplest approach consists in counting the co-occurrence of tags in different contexts (users or resources).…”

Section: Extracting the Emergent Semanticsmentioning

confidence: 99%

A Complete Life-Cycle for the Semantic Enrichment of Folksonomies

Limpens

Gandon

Buffa

2013

Advances in Knowledge Discovery and Management

View full text Add to dashboard Cite

Abstract. Tags freely provided by users of social tagging services are not explicitly semantically linked, and this significantly hinders the possibilities for browsing and exploring these data. On the other hand, folksonomies provide great opportunities to bootstrap the construction of thesauri. We propose an approach to semantic enrichment of folksonomies that integrates both automatic processing and user input, while formally supporting multiple points of view. We take into account the social structure of our target communities to integrate the folksonomy enrichment process into everyday tasks. Our system allows individual users to navigate more efficiently within folksonomies, and also to maintain their own structure of tags while benefiting from others contributions. Our approach brings also solutions to the bottleneck problem of knowledge acquisition by helping communities to build thesauri by integrating the manifold contributions of all their members, thus providing for a truly socio-semantic solution to folksonomy enrichment and thesauri construction.

show abstract

“…Markines et al [10] evaluated the performance of some similarity metrics using classical IR evaluation measures, when computing the similarity between tagged resources. However, this study was conducted on a single folksonomy data set (BibSonomy.org -a social bookmarking system), with the task being to predict URL-to-URL similarity.…”

Section: Similarity Metricsmentioning

confidence: 99%

“…However, to the best of our knowledge the usage of social tags for matching heterogeneous objects has not been investigated so far. We have used the similarity measures presented in [10] as a reference point, but we could not rely on their evaluation results since our study deals with different types of resources and a different ground truth. However, in [10], Matching, Overlap, Dice and Jaccard metrics performed slightly better than Cosine metric -a result that was also observed in our experiment.…”

Section: Related Workmentioning

confidence: 99%

“…-A set of similarity metrics, which were previously tested in a different setting [10], is evaluated first offline, comparing them in term of the generated ranked music recommendations for a POI, and then with a live user study where users expressed their subjective evaluations for the proposed match between music and POI. -The matching of music to POIs is carefully evaluated for each considered POI as opposed to the previous study where the user feedback was collected for itineraries, i.e., a collection of three POIs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Location-Adapted Music Recommendation Using Tags

Kaminskas

Ricci

2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Context-aware music recommender systems are capable to suggest music items taking into consideration contextual conditions, such as the user mood or location, that may influence the user preferences at a particular moment. In this paper we consider a particular kind of context aware recommendation task -selecting music content that fits a place of interest (POI). To address this problem we have used emotional tags attached by a users' population to both music and POIs. Moreover, we have considered a set of similarity metrics for tagged resources to establish a match between music tracks and POIs. In order to test our hypothesis, i.e., that the users will reckon that a music track suits a POI when this track is selected by our approach, we have designed a live user experiment where subjects are repeatedly presented with POIs and a selection of music tracks, some of them matching the presented POI and some not. The results of the experiment show that there is a strong overlap between the users' selections and the best matching music that is recommended by the system for a POI.

show abstract

Evaluating similarity measures for emergent semantics of social tagging

Cited by 219 publications

References 30 publications

Categorising social tags to improve folksonomy-based recommendations

Categorising social tags to improve folksonomy-based recommendations

A Complete Life-Cycle for the Semantic Enrichment of Folksonomies

Location-Adapted Music Recommendation Using Tags

Contact Info

Product

Resources

About