Figure 1.1 illustrates the basics of an information retrieval process. The analysis is based on simplified representation of documents created during the indexing process and on the representation of the targeted user profile. The analysis algorithm calculates the similarity based on these representations and determines how well each of the documents satisfies the user information requirements (how similar it is to the user profile). As a simplified representation can be less precise and more ambiguous that the original document (or profile), the search results can be less accurate than if a full original document had been compared with full profile, however the computational and storage requirements for such comparison would be higher.