engleski

Capturing clustering structure of the data in cross-language representation of textual documents

In the paper is proposed a new approach for representation of textual documents in lowerdimensional space. Approach aims at capturing clustering structure of the data in the goal of further application for cross-language classification of textual documents. The proposed approach is based on factorization of a term- document matrix by an iterative method of Reduced k-means clustering. Method of Reduced k- means aims at simultaneous reduction of objects (documents) and variables (index terms). Proposed method is compared to standard machine learning techniques of cross- language representation of textual documents by usage of latent semantic indexing and canonical correlation analysis.

clustering, textual documents, cross-language representation

nije evidentirano