Key-phrase Extraction for Classification

In this paper we consider the problem of extracting key-phrases from a bilingual texts collection and using them for text classification. A key-phrase could be defined as a sequence of words of a given size in a given partial order that occur within a sentence. We describe an algorithm for the discovery of key-phrases. Then, a framework of handling multilingual texts / documents is described which combines the use of the traditional vector space model with a new similarity measure which is based on the key-phrases. This framework is used for finding the most similar documents of a training set with any new document and selecting the classes of the similar documents as the most plausible ones for the new document. Some experimental results are also presented.