Publication

Abstract : With the rapid growth of the internet the availability of online information has also increased. As a result, it has become increasingly important to provide improved techniques for effectively and efficiently finding and representing textual content. Obtaining a representation of a document which can contain the semantic aspect of the document is always a hot topic of discussion. There are many methods to represent a document in vector form. But in many of these methods, the minimal representation of the document is not discussed. Dictionary learning is a method which can be used to obtain a minimal representation of documents. This paper presents a model that can obtain the minimal representation of documents using a sparse model and dictionaries of embedding. Two suitable techniques that we employ are SVD (Singular Value Decomposition) for dictionary learning and OMP (Orthogonal Matching Pursuit) for sparse coding.