Time Expression Analysis and Recognition Using Syntactic Token Types and General Heuristic Rules

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 420–429 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1039 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 420–429 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1039 Time Expression Analysis and Recognition Using Syntactic Token Types and General Heuristic Rules Xiaoshi Zhong, Aixin Sun, and Erik Cambria School of Computer Science and Engineering Nanyang Technological University, Singapore {xszhong,axsun,cambria}@ntu.edu.sg Abstract Extracting time expressions from free text is a fundamental task for many applica- tions. We analyze time expressions from four different datasets and find that only a small group of words are used to express time information and that the words in time expressions demonstrate similar syn- tactic behaviour. Based on the findings, we propose a type-based approach named SynTime1 for time expression recognition. Specifically, we define three main syntac- tic token types, namely time token, mod- ifier, and numeral, to group time-related token regular expressions. On the types we design general heuristic rules to rec- ognize time expressions. In recognition, SynTime first identifies time tokens from raw text, then searches their surroundings for modifiers and numerals to form time segments, and finally merges the time seg- ments to time expressions. As a light- weight rule-based tagger, SynTime runs in real time, and can be easily expanded by simply adding keywords for the text from different domains and different text types. Experiments on benchmark datasets and tweets data show that SynTime out- performs state-of-the-art methods. 1 Introduction Time expression plays an important role in infor- mation retrieval and many applications in natural language processing (Alonso et al., 2011; Campos et al., 2014). Recognizing time expressions from free text has attracted considerable attention since last decade (Verhagen et al., 2007, 2010; UzZa- man et al., 2013). 1Source: https://github.com/zhongxiaoshi/syntime We analyze time expressions in four datasets: TimeBank (Pustejovsky et al., 2003b), Giga- word (Parker et al., 2011), WikiWars (Mazur and Dale, 2010), and Tweets. From the analysis we make four findings about time expressions. First, most time expressions are very short, with 80% of time expressions containing no more than three tokens. Second, at least 91.8% of time expres- sions contain at least one time token. Third, the vocabulary used to express time information is very small, with a small group of keywords. Fi- nally, words in time expressions demonstrate sim- ilar syntactic behaviour. All the findings relate to the principle of least effort (Zipf, 1949). That is, people tend to act under the least effort in order to minimize the cost of energy at both individual level and collective level to language usage (Zipf, 1949). Time expression is part of language and acts as an interface of communication. Short ex- pressions, occurrence, small vocabulary, and sim- ilar syntactic behaviour all reduce the cost of en- ergy required to communicate. According to the findings we propose a type- based approach named SynTime (‘Syn’ stands …

Content truncated to 500 words