Personalization is being applied to great extend in many systems. This paper presents a multi-dimensional user data model and its application in web search. Online and Offline activities of the user are tracked for creating the user model. The main phases are identification of relevant documents and the representation of relevance and similarity of the documents. The concepts Keywords, Topics, URLs and clusters are used in the implementation. The algorithms for profiling, grading and clustering the concepts in the user model and algorithm for determining the personalized search results by re-ranking the results in a search bank are presented in this paper. Simple experiments for evaluation of the model and their results are described.