Towards a semantic tagger for analysing contents of Chinese corporate reports

Abstract

In this paper, we report on an experiment in which we explore the feasibility of applying a semantic tagger to analyze the textual contents of Chinese corporate reports andfocuse on the contents of corporate strategy. In recent years, Natural Language Processing (NLP) research has paid increasing attentions to the automatic analysis of the textual contents of corporate reports by using NLP approach on a large scale. We test the assumption that the semantic annotation tools can be useful for such a purpose and study the feasibility by testing a Chinese semantic tagger developed by UCREL, Lancaster University for extracting core Chinese terms and semantic concepts from Chinese corporate annual disclosures, focusing on three main USAS semantic categories, Money and Commerce, Architecture and Buildings, and Science and Technology, which we assume are closely related to the corporate strategy description, and use these categories and tags to extract core strategy terms. Our study shows that, by carefully applying the selected semantic categories, our semantic annotation tool is capable of extracting core Chinese terms which can further facilitate the content analysis of Chinese corporate reports.