A generic semantic-based framework for cross-domain recommendation

Abstract

In this paper, we present an ongoing research work on the design and development of a generic knowledge-based description framework built upon semantic networks. It aims at integrating and exploiting knowledge on several domains to provide crossdomain item recommendations. More specifically, we propose an approach that automatically extracts information about two different domains, such as architecture and music, which are available in Linked Data repositories. This enables to link concepts in the two domains by means of a weighted directed acyclic graph, and to perform weight spreading on such graph to identify items in the target domain (music artists) that are related to items of the source domain (places of interest).

Mobile Movie Recommendations with Linked Data

“…The Web Of Data paves the way to the usage of new and rich semantic datasets to compute recommendations. In [13] the authors present a generic knowledge-based description framework built upon semantic networks. The aim of the framework is to integrate and to exploit some knowledge on several domains in order to compute cross-domain recommendations.…”

Mining Semantic Data, User Generated Contents, and Contextual Information for Cross-Domain Recommendation

“…In [1] and [3] we present an approach that exploits the multi-domain DBpedia ontology (http://dbpedia.org) to build semantic networks linking items from different domains. The approach consists of two main components: a class-level network of DBpedia classes and relations belonging to certain domains of interest, and instance-level networks that are automatically generated instantiations of the class-level network for particular input items, by extracting and filtering information from the above ontology.…”

Cross-domain Recommendations for Personalized Semantic Services

“…Through the Internet, different products, services and customers can now easily interact with each other because the advance of Internet and Web technologies has continuously boosted the prosperity of e-commerce [4]- [7]. On the other hand, the more continuous development of electronic commerce, the more difficult it is for customers to single out products or services and find the most suitable ones with them.…”

“…Such systems are now popular both commercially and in the research community and even overwhelm human processing capabilities in a wide array of information seeking for decision making. So many Web sites such as Amazon, Netflix, Last.fm and many online retailers have proved that recommendation models are successful [2]- [7]. However, ample room and need for further improvements remain in the effective human decision support in a wide variety of applications.…”

“…In fact, joint recommendations in multiple domains are sometime needed for a customer. For instance, a system suggests not only a particular movie but also music CDs and books that are somehow related to that movie [2]- [7]- [5]- [12]. Such types are cross-domain recommendations.…”