How Web Applications Complement Search Engines?

Abstract

Search engines use content and link cues when searching for relevant pages that meet with the user query. They depend on clustering hypothesis and clustering link hypothesis as a strategy for search. There is a serious need for designing models that could reorganize the content of the web dynamically to meet users' needs. As search engines don't satisfy the user's need for complete and recently updated information, the researchers must design models depending on new concepts and paradigms to improve the coverage and recency of search engines. This work will focus on models used new concepts and paradigms for search, like the emergence concept as a stigmergy mechanism, the multi-agents system and Web Ants. Using the synthetic pheromone in Web Ants, the documents in Content Usage Ants model are labeled using the keywords not only in the content of documents in the content space and usage space, but also in the header and links of the web pages. The documents will have more weight if the words are found in the title and the link of the pages. This weight will be used to calculate the content similarity measurements in both spaces and compare the results to complement search engines. We will also make an overview of existing literature regarding intelligent semantic search engines. An overview of how the whole system of a search engine works is provided. and an overview of some models that been used to improve the way in which search engines work like: Content Usage Ants , Web comb models and other semantic web models.