RSS based news recommendation system with a weighted coverage density algorithm approach

Today the amount of non-structural data, especially textual data, significantly increased due to become widespread of internet. News portals are the most important examples of textual data. Huge amount of news documents are stored in databases, internet and intranet. This information is kept in a document, or text documents. Our primary problem is to extract significant information and obtain the documents that we need. The main aim is to learn the people?s preferences and habits of reading news documents.In this thesis, weighted coverage density, which is a timesaving text categorization technique, enabling access to data in text form, algorithm was analyzed. Accordingly an rss-based smart system with dynamic content, a system that dynamically aggregates the similar news read by the users by searching the news included in different categories of the news sites with different contents that determines news habits of the user and recommends the news that might be read in accordance with the related groups, was developed.