The ACRONYM Project: Discovering the textual thesaurus
نویسنده
چکیده
The ACRONYM Project began in April 1994, funded by the British government and industry for three years. It is a collaborative project, its partners being the University of Liverpool, FT Info (the archiving branch of the Financial Times), and the Press Association (a news agency supplying news to the regional press). The project has as its objective the development of an automated system for identifying thesaural items in electronic text. ACRONYM is an acronym, standing for 'The Automatic Collocational Retrieval Of Nyms'.
منابع مشابه
Discovering and visualizing narrative themes
This paper presents a framework for indexing and browsing databases of stories, in particular characterizing and visually exploring each narrative’s thematic content. We introduce a method for discovering thematic content in texts via lexical dissimilarity statistics. A maximumlikelihood algorithm clusters words into pools of similar meaning, using a thesaurus for rough estimates of word sense ...
متن کاملIdentifying diachronic change in semantic relations
test the hypothesis that words which are similar in meaning occur in similar contexts, this being a corollary of the initial hypothesis that the meaning of a word is defined by the collocational company which it keeps. This similarity is measured by creating 'collocational profiles'-summaries of the words typically occurring in the context of a given word-for each unique word form (type) in a l...
متن کاملFuzzy relational thesauri in information retrieval: automatic knowledge base expansion by means of classified textual data
In our ongoing project we develop a tool which provides domain engineers with a facility to create fuzzy relational thesauri (FRT) describing subject domains. The created fuzzy relational thesauri can be used as knowledge base for an intelligent information agent when answering user queries relevant to the described domains, or for textual searching on the web. However, the manual creation of (...
متن کاملFuzzy relational thesauri expansion using categorized textual data
In our ongoing project we develop a tool which provides domain engineers with a facility to create fuzzy relational thesauri (FRTi) describing subject domains. The created FRTi can be used as knowledge base for an intelligent information agent when answering user queries relevant to the described domains, or for textual searching on the web. However, the manual creation of thesauri is quite ted...
متن کاملAutomatic textual annotation of video news based on semantic visual object extraction
In this paper, we present our work for automatic generation of textual metadata based on visual content analysis of video news. We present two methods for semantic object detection and recognition from a cross modal image-text thesaurus. These thesaurus represent a supervised association between models and semantic labels. This paper is concerned with two semantic objects: faces and Tv logos. I...
متن کامل