A Semantic Composition Method for Deriving Sense Representations of Determinative-Measure Compounds in E-HowNet
1482Authors
nan
Abstract
In this paper, we take Determinative-Measure Compounds as an example to demonstrate how the E-HowNet semantic composition mechanism works in deriving the sense representations for all determinative-measure (DM) compounds which is an open set. We define the sense of a closed set of each individual determinative and measure word in E-HowNet representation exhaustively. We then make semantic composition rules to produce candidate sense representations for any newly coined DM. Then we review development set to design sense disambiguation rules. We use these heuristic disambiguation rules to determine the correct context-dependent sense of a DM and its E-HowNet representation. The experiment shows that the current model reaches 88% accuracy in DM identification and sense derivation. 關鍵詞:語意合成,定量複合詞,語意表達,廣義知網,知網 Keywords: Semantic Composition, Determinative-Measure Compounds, Sense Representations, Extended How Net, How Net 1.
Keywords
Full Content
A Semantic Composition Method for Deriving Sense Representations of Determinative-Measure Compounds in E-HowNet Chia-hung Tai, Shu-Ling Huang, Keh-Jiann Chen Institute of Information Science, Academia Sinica glaxy, josieh, kchen @iis.sinica.edu.tw 摘要 本篇論文利用定量複合詞為例,示範如何利用廣義知網的語意合成機制來推導複合詞的 語意及其表達式。首先我們定義了所有但有限數量的定詞跟量詞的廣義知網表達式,接 著我們利用語意合成的規則針對任何新的定量詞去產生候選的語意表達式。然後我們在 從調整語料集合去設計語意解歧規則,利用啟發式語意解歧規則跟參考上下文的詞來解 決定量詞的廣義知網表達式的歧異,實驗顯示在語意推導跟解歧之後有 88%的正確率。 Abstract In this paper, we take Determinative-Measure Compounds as an example to demonstrate how the E-HowNet semantic composition mechanism works in deriving the sense representations for all determinative-measure (DM) compounds which is an open set. We define the sense of a closed set of each individual determinative and measure word in E-HowNet representation exhaustively. We then make semantic composition rules to produce candidate sense representations for any newly coined DM. Then we review development set to design sense disambiguation rules. We use these heuristic disambiguation rules to determine the correct context-dependent sense of a DM and its E-HowNet representation. The experiment shows that the current model reaches 88% accuracy in DM identification and sense derivation. 關鍵詞:語意合成,定量複合詞,語意表達,廣義知網,知網 Keywords: Semantic Composition, Determinative-Measure Compounds, Sense Representations, Extended How Net, How Net 1. Introduction Building knowledge base is a time consuming work. The CKIP Chinese Lexical Knowledge Base has about 80 thousand lexical entries and their senses are defined in terms of the E-HowNet format. E-HowNet is a lexical knowledge and common sense knowledge representation system. It was extended from HowNet [1] to encode concepts. Based on the framework of E-HowNet, we intend to establish an automatic semantic composition mechanism to derive sense of compounds and phrases from lexical senses [2][3]. Determinative-Measure compounds (abbreviated as DM) are most common compounds in Chinese. Because a determinative and a measure normally coin a compound with unlimited versatility, the CKIP group does not define the E-HowNet representations for all DM compounds. Although the demonstrative, numerals, and measures may be listed exhaustively, their combination is inexhaustible. However their constructions are regular [4]. Therefore, an automatic identification schema in regular expression [4] and a semantic composition method under the framework of E-HowNet for DM compounds were developed. In this paper, we take DMs as an example to demonstrate how the E-HowNet semantic composition mechanism works in deriving the sense representations for all DM compounds. The remainder of this paper is organized as follows. The section 2 presents the background knowledge of DM compounds and sense representation in E-HowNet. We’ll describe our method in the section 3 and discuss the experiment result in the section 4 before we make conclusion in the section 5. 2. Background There are numerous studies on determinatives as well as measures, especially on the types of measures.1 Tai [5] asserts that in the literature on general grammar as well as Chinese grammar, classifiers and measures words are often treated together under one single framework of analysis. Chao [6] treats classifiers as one kind of measures. In his definition, a measure is a bound morpheme which forms a DM compound with the determinatives enumerated below. He also divides determinatives word into four subclasses: i. Demonstrative determinatives, e.g. 這” this”, that”那”… ii. Specifying determinatives, e.g. 每”every”, 各” each”… iii. Numeral determinatives, …
Content truncated to 500 words
Labeled Text
In this paper, we take Determinative-Measure Compounds as an example to demonstrate how the E-HowNet semantic composition mechanism works in deriving the sense representations for all determinative-measure (DM) compounds which is an open set. We define the sense of a closed set of each individual determinative and measure word in E-HowNet representation exhaustively. We then make semantic composition rules to produce candidate sense representations for any newly coined DM. Then we review development set to design sense disambiguation rules. We use these heuristic disambiguation rules to determine the correct context-dependent sense of a DM and its E-HowNet representation. The experiment shows that the current model reaches 88% accuracy in DM identification and sense derivation. 關鍵詞:語意合成,定量複合詞,語意表達,廣義知網,知網 Keywords: Semantic Composition, Determinative-Measure Compounds, Sense Representations, Extended How Net, How Net 1.
Publication Details
- Paper ID:
- 1482
- Year:
- 2008
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan