Ranking Complex Relationships on the Semantic Web

Abstract

Abstract. The focus of contemporary Web information retrieval systems has been to provide efficient support for the querying and retrieval of relevant documents. More recently, information retrieval over semantic metadata extracted from the Web has received an increasing amount of interest in both industry and academia. In particular, discovering complex and meaningful relationships among this metadata is an interesting and challenging research topic. Just as ranking of documents is a critical component of today's search engines, the ranking of complex relationships will be an important component in tomorrow's Semantic Web analytics engines. Building upon our recent work on specifying and discovering complex relationships in RDF data, called Semantic Associations, we present a flexible ranking approach which can be used to identify more interesting and relevant relationships in the Semantic Web. Additionally, we demonstrate our ranking scheme's effectiveness through an empirical evaluation over a real-world dataset.

Effect of Heuristics on Serendipity in Path-Based Storytelling with Linked Data

“…When ranking semantic associations, approaches semantically reinterpret query results in relation to the query context by using semantic distance (or similarity) to the datasets or search graph. Alternatively, a ranking can vary from rare relationships discovery mode to common relationships in conventional mode [3]. Techniques that support context driven ranking take into account ontological relations of the result instances in respect to the query context [2].…”

“…As can be seen in Table 2, the optimized algorithm seemed to be able to improve the link estimation of the resulting paths. To evaluate the results we used three different similarity measures: W2V 4 , NGD [6], and SemRank [3,4]. We used an online available Wiki2VecCorpus using vectors with dimension 1000, no stemming and 10skipgrams 5 .…”

Querying the Semantic Web with Preferences

“…Aleman-Meza et al [20] present a ranking approach which is based on measuring complex relationships. Two entities are related (semantically associated) if there is at least one binding property.…”

A Pragmatic Approach to Model and Exploit the Semantics of Product Information

“…In [8] the authors list the difficulties of building, maintaining, and integrating product information, and propose that an ontological approach may be the answer. In [25] the authors propose to use cross industry standard classifications such as UNSPSC 1 and eCl@ss 2 as the upper ontology and industry specific classifications as lower ontology. An upper ontology is about concepts that are generic, abstract and, therefore, are general enough to address (at a high level) a broad range of domain areas, while lower ontology contains domain-specific knowledge [35].…”

“…However, they use ontology only for refining the meaning of search terms while the underlying data is not ontologically modeled but stored in plain HTML or XML format. An abundance of semantic relationships and their characteristics in the context of the Semantic Web are well documented in [1]. In that paper, the authors blended semantics and information-theoretic techniques for their general search model, in which users can vary their search modes to affect the ordering of results depending on whether they need conventional search or investigative search.…”