Time Period Identification of Events in Text

Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the ACL, pages 1153–1160, Sydney, July 2006. c⃝2006 Association for Computational Linguistics Time Period Identification of Events in Text Taichi Noro† Takashi Inui†† Hiroya Takamura‡ Manabu Okumura‡ †Interdisciplinary Graduate School of Science and Engineering Tokyo Institute of Technology 4259 Nagatsuta-cho, Midori-ku, Yokohama, Kanagawa, Japan ††Japan Society for the Promotion of Science ‡Precision and Intelligence Laboratory, Tokyo Institute of Technology {norot, tinui}@lr.pi.titech.ac.jp,{takamura, oku}@pi.titech.ac.jp Abstract This study aims at identifying when an event written in text occurs. In particular, we classify a sentence for an event into four time-slots; morning, daytime, eve- ning, and night. To realize our goal, we focus on expressions associated with time-slot (time-associated words). How- ever, listing up all the time-associated words is impractical, because there are numerous time-associated expressions. We therefore use a semi-supervised learning method, the Naïve Bayes classi- fier backed up with the Expectation Maximization algorithm, in order to it- eratively extract time-associated words while improving the classifier. We also propose to use Support Vector Machines to filter out noisy instances that indicates no specific time period. As a result of ex- periments, the proposed method achieved 0.864 of accuracy and outperformed other methods. 1 Introduction In recent years, the spread of the internet has ac- celerated. The documents on the internet have increased their importance as targets of business marketing. Such circumstances have evoked many studies on information extraction from text especially on the internet, such as sentiment analysis and extraction of location information. In this paper, we focus on the extraction of tem- poral information. Many authors of documents on the web often write about events in their daily life. Identifying when the events occur provides us valuable information. For example, we can use temporal information as a new axis in the information retrieval. From time-annotated text, companies can figure out when customers use their products. We can explore activities of users for marketing researches, such as “What do people eat in the morning?”, “What do people spend money for in daytime?” Most of previous work on temporal processing of events in text dealt with only newswire text. In those researches, it is assumed that temporal ex- pressions indicating the time-period of events are often explicitly written in text. Some examples of explicit temporal expressions are as follows: “on March 23”, “at 7 p.m.”. However, other types of text including web diaries and blogs contain few explicit temporal expressions. Therefore one cannot acquire suffi- cient temporal information using existing meth- ods. Although dealing with such text as web dia- ries and blogs is a hard problem, those types of text are excellent information sources due to their overwhelmingly huge amount. In this paper, we propose a method for estimat- ing occurrence time of events expressed in in- formal text. In particular, we classify sentences in text into one of four time-slots; morning, day- time, evening, and night. To realize our goal, we focus on expressions associated with time-slot (hereafter, …

Content truncated to 500 words