Extended HowNet 2.0 – An Entity-Relation Common-Sense Representation Model

Extended HowNet 2.0 – An Entity-Relation Common-Sense Representation Model Wei-Yun Ma, Yueh-Yin Shih Institute of Information Science, Academia Sinica 128 Academia Road, Section 2, Nankang, Taipei, Taiwan 115 {ma, yuehyin}@iis.sinica.edu.tw Abstract In this paper, we propose Extended HowNet 2.0 – an entity-relation common-sense representation model. Comparing to HowNet and Extended HowNet, E-HowNet 2.0 has the following improvements: (a) Reorganizing the hierarchical structure of primitives and basic concepts; (b) Rich lexical information: In addition to sense definition, each entry of lexical sense may also include operational expressions as well as semantic functions which facilitate future semantic composition processes. (c) Improvement of sense definitions and sense definitions for basic concepts. (d) Developing a new automatic ontology reconstruction system. (e) Developing a query system called E-HowNet Relation Database for flexibly clustering concepts.We hope Extended HowNet 2.0 can bring significant benefits to the community of lexical semantics and natural language understanding. Keywords: E-HowNet, entity-relation, lexical semantics, ontology 1. Introduction The purpose of designing the lexical semantic representation model E-HowNet is for natural language understanding. E-HowNet, an evolution and extension of HowNet (Dong and Dong, 2006), is a frame-based entity- relation representation model to define lexical senses and to achieve compositional semantics. The current E-HowNet 2.0 1 shows the following improvements: (a) Reorganizing the hierarchical structure of primitives and basic concepts; (b) Rich lexical information: In addition to sense definition, each entry of lexical sense may also include operational expressions as well as semantic functions which facilitate future semantic composition processes. Event frames are also provided. (c) Improvement of sense definitions and sense definitions for basic concepts. (d) Developing a new automatic ontology reconstruction system: In case of revisions of lexical sense expressions or nodes of conceptual hierarchy, the ontology reconstruction system may re-attach each lexical entry to appropriated ontological nodes and results a new ontology. (e) Developing a query system called E- HowNet Relation Database for flexibly clustering concepts. The remainder of this paper is organized as follows. We describe the background of developing E-HowNet in Section 2 and elaborate the feature improvements of the current E-HowNet version 2.0 in Section 3. The online systems based on E-HowNet are introduced in Section 4. Section 5 is conclusions and future work. 2. Background HowNet is an on-line common-sense knowledge base unveiling the inter-conceptual relations and inter-attribute relations of concepts conveyed by Chinese words and their English equivalents (Dong & Dong, 2006). Compared with WordNet, HowNet’s architecture provides richer information apart from hyponymy relations. It also 1 http://ehownet.iis.sinica.edu.tw/index.php enriches relational links between words via encoded feature relations. The advantages of HowNet are (a) inherent properties of concepts are derived from encoded feature relations in addition to hypernymous concepts, and (b) information regarding conceptual differences between different concepts and information regarding morpho- semantic structure are encoded. HowNet’s advantages make it an effective electronic dictionary for NLP applications. In recent years, HowNet has been applied to the researches of word similarity calculation (Liu & Li, 2002 ), machine translation (Dong 1999), and Information Retrieval (Dorr, Levow and Lin, …

Content truncated to 500 words