This paper describes an inductive logic programming learning method designed toacquire from a corpus specific Noun-Verb (N-V) pairs---relevant in informationretrieval applications to perform index expansion---inorder to build up semantic lexicons based on Pustejovsky's generative lexicon(GL) principles (Pustejovsky, 1995). In one of the components of this lexical model,called the qualia structure, words are described in terms of semanticroles. For example, the telic role indicates the purpose or function ofan item (cut for knife), the agentive role its creation mode(build for house), etc. The qualia structure of a noun ismainly made up of verbal associations, encoding relational information. Thelearning method enables us toautomatically extract, from a morpho-syntactically and semantically taggedcorpus, N-V pairs whose elements are linked by one of the semantic relationsdefined in the qualia structure in GL. It also infers rules explaining what inthe surrounding context distinguishes such pairs from others alsofound in sentences of the corpus but which are not relevant. Stress is put here on thelearning efficiency that is required to be able to deal with all the availablecontextual information, and to produce linguistically meaningful rules. color="gray">
展开▼
机译：本文介绍了一种归纳逻辑编程学习方法，该方法旨在从语料库特定的名词-动词（NV）对中获取-与信息检索应用程序相关以执行索引扩展--以基于Pustejovsky的生成词典（GL）建立语义词典原则（Pustejovsky，1995）。在这种词汇模型的一个组成部分中，称为 qualia结构，单词是根据语义角色来描述的。例如， telic 角色指示某项的目的或功能（刀为 的 cut ），其创建模式为 （用于 house ）。等等。名词的质素结构主要由语言联想组成，对关联信息进行编码。这种学习方法使我们能够自动从形态-句法和语义标记语料库中提取N-V对，其元素通过GL的质素结构中定义的语义关系之一链接。它还推论出一些规则，这些规则解释了在周围环境中哪些可以将此类对与在语料库句子中也可以找到但并不相关的其他对区分开。这里强调的是学习效率，它必须能够处理所有可用的上下文信息，并产生有意义的语言规则。 color =“ gray”>
展开▼
