NAGA: Searching and Ranking Knowledge

Abstract

The Web has the potential to become the world’s largest knowledge base. In order to unleash this potential, the wealth of information available on the Web needs to be extracted and organized. There is a need for new querying techniques that are simple and yet more expressive than those provided by standard keyword-based search engines. Searching for knowledge rather than Web pages needs to consider inherent semantic structures like entities (person, organization, etc.) and relationships (isA, locatedIn, etc.). In this paper, we propose NAGA, a new semantic search engine. NAGA builds on a knowledge base, which is organized as a graph with typed edges, and consists of millions of entities and relationships extracted from Web-based corpora. A graph-based query language enables the formulation of queries with additional semantic information. We introduce a novel scoring model, based on the principles of generative language models, which formalizes several notions like confidence, informativeness and compactness and uses them to rank query results. We demonstrate NAGA’s superior result quality over state-of-the-art search engines and question answering systems

Searching web data: An entity retrieval and high-performance indexing model

“…Other works [45,46,37,25,47] have extended simple keyword-based search with structured queries capabilities. In [25,47], they propose a partial solution to the lack of expressiveness of the keyword-based approach by allowing search using conditions on attributes and values.…”

“…In [25,47], they propose a partial solution to the lack of expressiveness of the keyword-based approach by allowing search using conditions on attributes and values. In [45,46,37], they present more powerful query language by adopting a graph-based model. However, the increase of query expressiveness is tied with the processing complexity, and the graph-based models [45,46,37] are not applicable on a very large scale.…”

Semantic Web Search Based on Ontological Conjunctive Queries

“…We now discuss related work on Semantic Web search (see especially [28] for a recent survey), which can roughly be divided into (1) approaches that are based on structured query languages, such as [17,30,35,38,45,46,50], and (2) approaches for naive users, requiring no familiarity with structured query languages. In this category, we distinguish keyword-based approaches, such as [13,33,34,40,51,52,55], where queries consist of lists of keywords, and natural-language-based approaches, such as [16,23,29,32,42,43], where users can express queries in natural language.…”

“…We first focus on some more general approaches [17,35,38] closest in spirit to ours in that they aim at providing general semantic search facilities. We then discuss some proposals [30,50,45,46] that address some specific aspects of semantic search or that are targeted at specific domains, so that they cannot be strictly viewed as semantic search engines.…”

Mining and Explaining Relationships in Wikipedia

“…Lehmann et al [14] then developed a tool for exploring relationships extracted by DBpedia. NAGA [15] is also a semantic search engine which searches semantic relationships, using a semantic knowledge base YAGO [16] extracted from Wikipedia and WordNet. Several methods [17], [18] were proposed for extracting paths between two concepts on an RDF graph whose edges represent explicit relationships with semantics.…”