Cross-system user modeling and personalization on the Social Web

Abel, Fabian; Herder, Eelco; Houben, Geert‐Jan; Henze, Nicola; Krause, Daniel

doi:10.1007/s11257-012-9131-2

Cited by 172 publications

(122 citation statements)

References 24 publications

Supporting

Mentioning

121

Contrasting

Order By: Relevance

“…First of all, we should better correlate algorithm performance to the characteristics of the data sets (sparsity, distribution of tags, overlap of tags between domains). Secondly, the performance of our proposed models on other datasets should be assessed and a comparison with other cross-domain recommenders is in order [11,12,5]. Moreover, we are interested in better understanding the conditions when the tag-based models can be exploited, e.g., in context-aware recommender systems, and if these techniques could be used to generate more diverse recommendations.…”

Section: Discussionmentioning

confidence: 99%

Cold-Start Management with Cross-Domain Collaborative Filtering and Tags

Enrich

Braunhofer

Ricci

2013

Lecture Notes in Business Information Processing

View full text Add to dashboard Cite

Abstract. Recommender systems suffer from the new user problem, i.e., the difficulty to make accurate predictions for users that have rated only few items. Moreover, they usually compute recommendations for items just in one domain, such as movies, music, or books. In this paper we deal with such a cold-start situation exploiting cross-domain recommendation techniques, i.e., we suggest items to a user in one target domain by using ratings of other users in a, completely disjoint, auxiliary domain. We present three rating prediction models that make use of information about how users tag items in an auxiliary domain, and how these tags correlate with the ratings to improve the rating prediction task in a different target domain. We show that the proposed techniques can effectively deal with the considered cold-start situation, given that the tags used in the two domains overlap.

show abstract

Section: Discussionmentioning

confidence: 99%

Cold-Start Management with Cross-Domain Collaborative Filtering and Tags

Enrich

Braunhofer

Ricci

2013

Lecture Notes in Business Information Processing

View full text Add to dashboard Cite

show abstract

“…For example, [27] obtains Facebook data for recommendation processes; [28] retrieves user interactions from Facebook to produce behavioural patterns; [29] proposes a model to capture social relationships and status from different social media contexts; or [30] describes how to aggregate different user profiles from several social networks. The study presented in [31] includes an in depth comparison of the user modelling strategies in a multi-social context.…”

Section: Related Workmentioning

confidence: 99%

Creating and Modelling Personal Socio-Economic Networks in On-Line Banking

Miguel

Álamo

Yelmo

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The banking industry is observing how new competitors threaten its millennial business model by targeting unbanked people, offering new financial services to their customer base, and even enabling new channels for existing services and customers. The knowledge on users, their behaviour, and expectations become a key asset in this new context. Well aware of this situation, the Center for Open Middleware, a joint technology center created by Santander Bank and Universidad Politécnica de Madrid, has launched a set of initiatives to allow the experimental analysis and management of socio-economic information. PosdataP2P service is one of them, which seeks to model the economic ties between the holders of university smart cards, leveraging on the social networks the holders are subscribed to. In this paper we describe the design principles guiding the development of the system, its architecture and some implementation details.

show abstract

“…The performance of several cross-system user modeling strategies in the context of recommender systems is developed and evaluated to solve the cold-start problem and improve recommendation quality [13].…”

Section: Related Workmentioning

confidence: 99%

“…By definition, cross-domain recommendation is providing recommendations of items in one (source) domain using the preferences expressed on items in a second (target) domain [7]- [5]. Another task for cross-domain recommendation is making joint recommendations for items belonging to different domains [13].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cross-domain Recommendations for Personalized Semantic Services

Moe¹,

Aung²

2012

IJCATR

View full text Add to dashboard Cite

An increasing amount of work has been published in various areas related to the Recommender System. Among them, cross-domain recommendation is an emerging research topic and in this field, it is important to investigate how to manage personalization and how to consider customer's contextual features to keep more user satisfaction and accuracy. This paper tends to provide cross-domain recommendations for personalized semantic services using Taxonomic CCBR, directed acyclic graph by FordFulkerson algorithm and TOPSIS method. Taxonomic CCBR helps the system get the accurate problem by engaging a user in a series of questions and answers from the user's partial definition of the problem. Semantic concepts between different domains are considered by using weighted directed acyclic graph to find meaningful solutions. Then TOPSIS method is used to get the results more precisely considering contextual features such as season, place, etc. which have not been addressed in the current cross-domain recommender systems.

show abstract

Cross-system user modeling and personalization on the Social Web

Cited by 172 publications

References 24 publications

Cold-Start Management with Cross-Domain Collaborative Filtering and Tags

Cold-Start Management with Cross-Domain Collaborative Filtering and Tags

Creating and Modelling Personal Socio-Economic Networks in On-Line Banking

Cross-domain Recommendations for Personalized Semantic Services

Contact Info

Product

Resources

About