Identifying Temporal Components in a Chinese Temporal Information System

Abstract

Temporal structures are important semantic carriers in information extraction. Over the past few years, temporal information processing and temporal database management have gradually become popular research topics. Unfortunately, few researchers have investigated these areas in Chinese. This impels us to put efforts on exploiting Chinese temporal expressions for concept-based reasoning. The lack of regular morphological tense markers makes Chinese temporal expressions complicated. Therefore, the study of Chinese temporal information (or concept) extraction has profound significance. In this paper, an algorithm for Chinese temporal information extraction is presented. We argue that the type of Chinese concept, which is embedded in a sentence, must first be identified by verb properties. Thereafter, the overall temporal concept can be determined when the other temporal and non-temporal components in the sentence are considered. Based on the investigation of Chinese temporal expressions, we then propose a heuristic algorithm to identify temporal components from the running text. The performance of the algorithm in terms of recall and precision is reported. The algorithm was proved successful in the test. It is worth noting that the work presented in this paper is only a part of an ongoing project, which aims at extracting and evaluating temporal relations over Hong Kong financial news reports.