Automatic Personalized Marathi Content Generation

The purpose of the present work is to create a system to retrieve personalized documents in Marathi Language. The system mainly focuses on providing personalized documents to the end user by analyzing the browsing history and user profile of the user in Marathi language. The system also provides manual bookmark facility to the end user as per the user interest. This paper provides personalization of Marathi text documents by using Label Induction Grouping [LINGO] Algorithm based on Vector Space Model [VSM]. This paper presents the automatic personalization of Marathi documents and literature survey of the related work done in automatic categorization of Marathi text documents. Several learning techniques exist for the classification of text documents like Decision Trees, Support Vector Machine, Naïve Bayes, etc. Several clustering techniques are available for text categorization namely K-means, Suffix Tree Clustering, Label Induction Grouping Algorithm, etc. With the help of literature survey, it is found that Vector Space Model [VSM] gives better accuracy than other models.