The World Wide Web has grown exponentially and has become the most pervasive source of information in the last few years. Since most of this information is scattered and heterogeneous in nature, efficient and accurate retrieval of literature on the web is a challenging area of research. Most of the time the users are not aware of what they are looking for or they have an abstract view of what they want. Modern search engines help them narrow their search but are not sophisticated enough to understand their needs. The Semantic Web is a vision of next generation World Wide Web where the information is defined with an explicit meaning, which machines can understand, process and integrate without human intervention. SemPub centers on semantic retrieval and user customization. In semantic search, we search for terms in a concept space (a graph of terms occurring in documents linked to each other by the frequency and relationships between them and with which they occur together). Hence by reducing the high dimensional and complex object space into a more manageable automatically-generated, meaningful concept space, customized retrieval and information routing is made easy.