Category Development Using Complete Semantic Networks Category Development Using Complete Semantic Networks 3 Analysis of Text Sample Using Lexical Cohesion 4 Content Analysis Category System
نویسندگان
چکیده
1 Abstract In developing lexicons to support a scoring system for open-ended test questions and a content analysis system, we found it necessary to extend the conventional notion of a semantic network as consisting of concept nodes and (perhaps several kinds of) labeled links by adding nodes for pseudoentries capturing semantic features (as in typed feature structures) and by viewing links as nodes with selectional restrictions on their arguments. For the test scoring system, we incorporated a module for maintaining lexical cohesion, based on the extended semantic network, into a general discourse analysis algorithm. The module accomplishes this objective by requiring discourse entities to be instantiations of lexical relations licensed by the network. For the content analysis dictionary, which assigns categories to entries, we found that category definition required use of lexical semantic information beyond the ISA (and other standard relational) backbone of a semantic network. We conclude that these more "complete" semantic networks are necessary and desirable for various NLP tasks. We found that several principles for category development and extension readily emerged, including principles for characterizing the "category" of a text. These principles have some correspondence with techniques for identifying the topic of a text. We describe some summary procedures for category development and naming. 2 Introduction Two specific tasks involving lexicon development and analysis have led to the need for the broadened conceptualization of the knowledge contained in semantic networks of lexicons. The first task was to build a knowledge representation of texts (5-6 sentences of 30-100 words) for the Formulating-Hypotheses test item developed at the Educational Testing Service (ETS); this task required the use of lexical cohesion principles to build the discourse structure of the text. The second task involved the analysis of a content-analysis dictionary of 11,000 words. We describe these tasks, both of which involved use of WordNet (Miller, et al., 1990) and consideration of other semantic networks. We identified shortcomings of these networks in completing the tasks. We present extensions of the nodes and relations in semantic networks, principally by converting labels on links into lexicon entries and by adding pseudoentries corresponding to pieces of lexical semantic information, to enable completion of the tasks. In both tasks, we found that sublexicons can be characterized in a way resembling categorization schemes used in information retrieval (Hearst & Schütze, 1996). We compare these extended networks with previous conceptions of semantic networks and their relation to …
منابع مشابه
Dynamic Categorization of Semantics of Fashion Language: A Memetic Approach
Categories are not invariant. This paper attempts to explore the dynamic nature of semantic category, in particular, that of fashion language, based on the cognitive theory of Dawkins’ memetics, a new theory of cultural evolution. Semantic attributes of linguistic memes decrease or proliferate in replication and spreading, which involves a dynamic development of semantic category. More specific...
متن کاملNeural Networks in Chinese Lexical Classification
Lexical attributes, like syntactic (part-of-speech) and semantic (semantic category) attributes, are in most cases, ambiguous in every languages. Automatic resolution of ambiguity of these attributes can be achieved using different techniques; rule-based, statistical, NN-based and their hybrids. Moreover, one linguistic feature also has influence over the resolution of ambiguity of another feat...
متن کاملAnalysis of the Wikipedia Category Graph for NLP Applications
In this paper, we discuss two graphs in Wikipedia (i) the article graph, and (ii) the category graph. We perform a graphtheoretic analysis of the category graph, and show that it is a scale-free, small world graph like other well-known lexical semantic networks. We substantiate our findings by transferring semantic relatedness algorithms defined on WordNet to the Wikipedia category graph. To as...
متن کاملHuman Resource Development through Mentoring: Case of Iran Electricity Grid Management Company
The main purpose of this study is to design a mentoring model in Iran Grid Management Company. This research has been done with a qualitative approach. Participants in this study are 18 managers of Iran Grid Management Company who are active in the field of technology, engineering and human resources and participated in the research through purposeful sampling and theoretical saturation rule. D...
متن کاملDesign and implementation of Persian spelling detection and correction system based on Semantic
Persian Language has a special feature (grapheme, homophone, and multi-shape clinging characters) in electronic devices. Furthermore, design and implementation of NLP tools for Persian are more challenging than other languages (e.g. English or German). Spelling tools are used widely for editing user texts like emails and text in editors. Also developing Persian tools will provide Persian progr...
متن کامل