Utilizing Temporal Information for Taxonomy Construction

Utilizing Temporal Information for Taxonomy Construction Luu Anh Tuan ∗1, Siu Cheung Hui †2, See Kiong Ng #3 ∗Institute for Infocomm Research, Singapore 1at.luu@i2r.a-star.edu.sg †School of Computer Engineering, Nanyang Technological University, Singapore 2asschui@ntu.edu.sg #Institute of Data Science, National University of Singapore 3seekiong@nus.edu.sg Abstract Taxonomies play an important role in many applications by organizing domain knowledge into a hierarchy of ‘is-a’ relations between terms. Previous work on automatic construc- tion of taxonomies from text documents ei- ther ignored temporal information or used fixed time periods to discretize the time se- ries of documents. In this paper, we pro- pose a time-aware method to automatically construct and effectively maintain a taxon- omy from a given series of documents pre- clustered for a domain of interest. The method extracts temporal information from the docu- ments and uses a timestamp contribution func- tion to score the temporal relevance of the evidence from source texts when identifying the taxonomic relations for constructing the taxonomy. Experimental results show that our proposed method outperforms the state- of-the-art methods by increasing F-measure up to 7%-20%. Furthermore, the proposed method can incrementally update the taxon- omy by adding fresh relations from new data and removing outdated relations using an in- formation decay function. It thus avoids re- building the whole taxonomy from scratch for every update and keeps the taxonomy effec- tively up-to-date in order to track the latest in- formation trends in the rapidly evolving do- main. 1 Introduction The explosion in the amount of unstructured text data gives us the opportunity to explore knowledge in depth, but there are also challenges to recog- nize useful information for our interests. To pro- vide access to information effectively, it is impor- tant to organize unstructured data in a structured and meaningful manner. Taxonomies, which serve as backbones for structured knowledge, are use- ful for many NLP applications such as question answering (Harabagiu et al., 2003) and document clustering (Fodeh et al., 2011). However, hand- crafted, well-structured taxonomies such as Word- Net (Miller, 1995), OpenCyc (Matuszek et al., 2006) and Freebase (Bollacker et al., 2008), which are pub- licly available, can be incomplete for new or special- ized domains. As it is time-consuming and cum- bersome to create a new one manually, methods for automatic domain-specific taxonomy construc- tion from text corpora are highly desirable. Previous work on automatic construction of domain-specific taxonomies from text documents assumed that the data sets (that is, the document sets) and the underlying taxonomic relations are static. However, the data sets for certain domains may evolve over time, as new documents are added while older documents are deleted or modified. As such, the taxonomic relations for these potentially fast-changing domains may not remain static but become dynamic over time as new domain terms emerge while some older ones disappear. For ex- ample, in World Health Organization reports about disease outbreak, the term ‘smallpox’ used to be a hyponym of ‘dangerous disease’, but it has fallen off since 1980. On the other hand, since 2014 …

Content truncated to 500 words