Searching in Archaeological Texts: Problems and Solutions Using an Artificial Intelligence Approach
暂无分享,去创建一个
Searching in documents using full text indices is a powerful tool for retrieving relevant portions of text. However, performance is impeded by ambiguity in texts: similar words may have totally different meanings according to context. This also is true if the words are numbers, periods and place names, especially in archaeological and historical contexts. A new way of indexing texts allows for better and easier searching. This system has been developed in a collaboration between the RCE (The Dutch National Service for Cultural Heritage) 1 and the University of Tilburg. With Open Boek, 2 it is possible to search on chronological and geographical expressions, as well as regular keywords. In the newest version of Open Boek a number of additions to the system have been made to further improve the functionality.
[1] G. Lange,et al. Het zoeken en interpreteren van jaartallen en plaatsnamen in Nederlandstalige retrieval systemen , 2008 .
[2] Bruno Pouliquen,et al. Geocoding Multilingual Texts: Recognition, Disambiguation and Visualisation , 2006, LREC.
[3] Vladimir I. Levenshtein,et al. Binary codes capable of correcting deletions, insertions, and reversals , 1965 .
[4] Walter Daelemans,et al. TiMBL: Tilburg Memory-Based Learner, version 2.0, Reference guide , 1998 .