Harvesting Domain Specific Ontologies from Text

Abstract

Abstract-Ontologies are a vital component of most knowledgebased applications, including semantic web search, intelligent information integration, and natural language processing. In particular, we need effective tools for generating in-depth ontologies that achieve comprehensive converge of specific application domains of interest, while minimizing the time and cost of this process. Therefore we cannot rely on the manual or highly supervised approaches often used in the past, since they do not scale well. We instead propose a new approach that automatically generates domain-specific ontologies from a small corpus of documents using deep NLP-based text-mining. Starting from an initial small seed of domain concepts, our OntoHarvester system iteratively extracts ontological relations connecting existing concepts to other terms in the text, and adds strongly connected terms to the current ontology. As a result, OntoHarvester (i) remains focused on the application domain, (ii) is resistant to noise, and (iii) generates very comprehensive ontologies from modest-size document corpora. In fact, starting from a small seed, OntoHarvester produces ontologies that outperform both manually generated ontologies and ontologies generated by current techniques, even those that require very large well-focused data sets.

Text-Mining, Structured Queries, and Knowledge Management on Web Document Corpora

“…Extensive experiments were conducted to evaluate the effectiveness of our systems [6,16,18,19,21,9]. In this section, we present the results we obtained by applying IBminer onto the text of the entire English Wikipedia, which is a corpus containing 4.4 Million subjects each described by 18.2 sentences on average.…”

“…IBminer [19] and OntoMiner [20,21], described in this section, use a deep NLP-based knowledge extraction approach to improve the completeness, consistency, and accuracy of the KB. …”

“…IBminer [19] and OntoMiner [20,21] (Section 4), which respectively build structured summaries and ontologies from document corpora using text-mining, IKBStore and CS 3 [17] (Section 5) that integrate knowledge from different sources and realign subject and attribute names to construct a general KB having the quality and coverage needed for semantic applications, and IBE [18] (Section 6) that supports the knowledge editing and query functions needed for managing and upgrading the KB, while maintaining provenance information on each piece of knowledge entered into the system.…”

Mining Semantics Structures from Syntactic Structures in Web Document Corpora

“…Thus, we propose the Ontoharvester system [39,40] which performs NLP analysis on TextGraphs to generate ontologies from text. Using a small set of initial seed concepts, Ontoharvester iteratively extracts ontological relations connecting existing concepts to other terms in TextGraphs, and adds strongly connected terms to the current ontology.…”

“…Besides the components listed in Fig. 8, the tools and techniques developed by this framework will also bene¯t other TextGraph based applications: IBminer [37], OntoHarvester system [40], and CS 3 [36].…”