Tommaso Di Noia

A system for principled matchmaking in an electronic marketplace

More and more resources are becoming available on the Web, and there is a growing need for infrastructures that, based on advertised descriptions, are able to semantically match demands with supplies.We formalize general properties a matchmaker should have, then we present a matchmaking facilitator, compliant with desired properties.The system embeds a NeoClassic reasoner, whose structural subsumption algorithm has been modified to allow match categorization into potential and partial, and ranking of matches within categories. Experiments carried out show the good correspondence between users and system rankings.

Linked open data to support content-based recommender systems

The World Wide Web is moving from a Web of hyper-linked Documents to a Web of linked Data. Thanks to the Semantic Web spread and to the more recent Linked Open Data (LOD) initiative, a vast amount of RDF data have been published in freely accessible datasets. These datasets are connected with each other to form the so called Linked Open Data cloud. As of today, there are tons of RDF data available in the Web of Data, but only few applications really exploit their potential power. In this paper we show how these data can successfully be used to develop a recommender system (RS) that relies exclusively on the information encoded in the Web of Data. We implemented a content-based RS that leverages the data available within Linked Open Data datasets (in particular DBpedia, Freebase and LinkedMDB) in order to recommend movies to the end users. We extensively evaluated the approach and validated the effectiveness of the algorithms by experimentally measuring their accuracy with precision and recall metrics.

Concept abduction and contraction for semantic-based discovery of matches and negotiation spaces in an e-marketplace

In this paper, we present a Description Logic approach - fully compliant with the Semantic web vision and technologies - to extended matchmaking between demands and supplies in a semantic-enabled Electronic Marketplace, which allows the semantic-based treatment of negotiable and strict requirements in the demand/supply descriptions. To this aim, we exploit two novel non-standard Description Logic inference services, Concept Contraction - which extends satisfiability - and Concept Abduction - which extends subsumption. Based on these services, we devise algorithms, which allow to find negotiation spaces and to determine the quality of a possible match, also in the presence of a distinction between strictly required and optional elements. Both the algorithms and the semantic-based approach are novel, and enable a mechanism to boost logic-based discovery and negotiation stages within an e-marketplace. A set of simple experiments confirm the validity of the approach.

Semantic matchmaking as non-monotonic reasoning: a description logic approach

Matchmaking arises when supply and demand meet in an electronic marketplace, or when agents search for a web service to perform some task, or even when recruiting agencies match curricula and job profiles. In such open environments, the objective of a matchmaking process is to discover best available offers to a given request. We address the problem of matchmaking from a knowledge representation perspective, with a formalization based on Description Logics. We devise Concept Abduction and Concept Contraction as non-monotonic inferences in Description Logics suitable for modeling matchmaking in a logical framework, and prove some related complexity results. We also present reasonable algorithms for semantic matchmaking based on the devised inferences, and prove that they obey to some commonsense properties. Finally, we report on the implementation of the proposed matchmaking framework, which has been used both as a mediator in e-marketplaces and for semantic web services discovery.

Top-N recommendations from implicit feedback leveraging linked open data

The advent of the Linked Open Data (LOD) initiative gave birth to a variety of open knowledge bases freely accessible on the Web. They provide a valuable source of information that can improve conventional recommender systems, if properly exploited. In this paper we present SPrank, a novel hybrid recommendation algorithm able to compute top-N item recommendations from implicit feedback exploiting the information available in the so called Web of Data. We leverage DBpedia, a well-known knowledge base in the LOD compass, to extract semantic path-based features and to eventually compute recommendations using a learning to rank algorithm. Experiments with datasets on two different domains show that the proposed approach outperforms in terms of prediction accuracy several state-of-the-art top-N recommendation algorithms for implicit feedback in situations affected by different degrees of data sparsity.

A System for Principled Matchmaking in an Electronic Marketplace

More and more resources are becoming available on the Web, and there is a growing need for infrastructures that, based on advertised descriptions, can match demands with supplies in an electronic marketplace. Several matchmakers rely on simple keyword matching, but significant matching results can only be obtained by exploiting the semantics inherent in structured descriptions. To this end, a novel knowledge representation approach is proposed, based on description logics, that is superior to both keyword- and basic subsumption-based matchmaking. The properties a matchmaker should have are formalized, and a matchmaking facilitator, compliant with the desired properties, is presented. The system embeds a NeoClassic reasoner whose structural subsumption algorithm has been modified to allow match categorization into potential and partial, and ranking of matches within categories. Experiments show good correspondence between users and system rankings.

A Nonmonotonic Approach to Semantic Matchmaking and Request Refinement in E-Marketplaces

Matchmaking in e-marketplaces consists of finding and retrieving promising counterparts for a given request from the set of available advertisements. This paper proposes the use of nonmonotonic inferences (concept contraction and concept abduction) in a semantic-matchmaking process for ranking resource descriptions. Concept contraction can be used to amend requests incompatible with the resource descriptions. The more amendments needed, the less is the degree of match. If a request is compatible with an advertisement but does not subsume it, concept abduction can be used to hypothesize extra features in the advertisement. The more it is necessary to hypothesize, the less is the degree of match. These basic ideas are utilized to compute a meaningful matchmaking ranking. Using logical explanations on matchmaking results, an approach and algorithms are proposed for the progressive refinement and revision of requests, up to an almost exact match. The related issue of user interaction is also tackled, and a user-friendly tool is presented that allows full utilization of the semantic-based query/revision/refinement process while completely hiding logical technicalities.

Exploiting the web of data in model-based recommender systems

The availability of a huge amount of interconnected data in the so called Web of Data (WoD) paves the way to a new generation of applications able to exploit the information encoded in it. In this paper we present a model-based recommender system leveraging the datasets publicly available in the Linked Open Data (LOD) cloud as DBpedia and LinkedMDB. The proposed approach adapts support vector machine (SVM) to deal with RDF triples. We tested our system and showed its effectiveness by a comparison with different recommender systems techniques -- both content-based and collaborative filtering ones.

Ranking the linked data: the case of DBpedia

The recent proliferation of crowd computing initiatives on the web calls for smarter methodologies and tools to annotate, query and explore repositories. There is the need for scalable techniques able to return also approximate results with respect to a given query as a ranked set of promising alternatives. In this paper we concentrate on annotation and retrieval of software components, exploiting semantic tagging relying on Linked Open Data. We focus on DBpedia and propose a new hybrid methodology to rank resources exploiting: (i) the graphbased nature of the underlying RDF structure, (ii) context independent semantic relations in the graph and (iii) external information sources such as classical search engine results and social tagging systems. We compare our approach with other RDF similarity measures, proving the validity of our algorithm with an extensive evaluation involving real users.

Semantic matchmaking in a P-2-P electronic marketplace

Matchmaking is the problem of matching offers and requests, such as supply and demand in a marketplace, services and customers in a service agency, etc., where both partners are peers in the transaction. Peer-to-Peer (P-2-P) e-commerce calls for an infrastructure treating in a uniform way supply and demand, which should base the match on a common ontology for describing both supply and demand. Knowledge representation --- in particular description logics --- can deal with this uniform treatment of knowledge from vendors and customers, by modelling both as generic concepts to be matched. We propose a logical approach to supply-demand matching in P-2-P e-commerce, which allows us to clearly distinguish between exact, potential and partial match, and to define a ranking within the categories. The approach is deployed in a prototype system implemented for a particular case study (but easily generalizable) and is based on Classic, a well-known knowledge representation system.