Genetic Algorithms for Topical Web Search: A Study of Different Mutation Rates

CECCHINI, R. L.; LORENZETTI, C. M.; MAGUITMAN, A. G.; BRIGNOLE, N. B.

Corrientes y Resistencia, Argentina

Congreso; VIII WASI: VIII Workshop de Agentes y Sistemas Inteligentes - CACIC 2007: XIII Congreso Argentino de Ciencias de la Computación; 2007

Systems for searching the Web based on thematic contexts can be built on top of aconventional search engine and benefit from the huge amount of content as well asfrom the functionality available through the search engine interface. The quality ofthe material collected by such systems is highly dependant on the vocabulary used togenerate the search queries. In this scenario, selecting good query terms can be seenas an optimization problem where the objective function to be optimized is basedon the effectiveness of a query to retrieve relevant material. Some characteristics ofthis optimization problem are (1) the high-dimensionality of the search space, wherecandidate solutions are queries and each term corresponds to a different dimension,(2) the existence of acceptable suboptimal solutions, (3) the possibility of findingmultiple solutions, and in many cases (4) the quest for novelty. This article describesoptimization techniques based on Genetic Algorithms to evolve “good query terms”in the context of a given topic. The proposed techniques place emphasis on searchingfor novel material that is related to the search context. We discuss the use of amutation pool to allow the generation of queries with new terms, study the effectof different mutation rates on the exploration of query-space, and discuss the useof a especially developed fitness function that favors the construction of queriescontaining novel but related terms.