The Research Of The Semantic Feature Extraction Based On Genetic Algorithm

The information on the Internet is like a vast ocean so that people can not effectively select and use it. At present, text-based and keyword search engines have become the main way for people to obtain information from the Internet, but this kind of search results are always less relevant to the expectation, by listing a lot of irrelevant information. The evaluation of a search engine has two standards, the accuracy of the search results and the sorting of such search results. The accuracy of the search results refers to the quantity of the information which the user demanded among all the search results.The larger number we get, the greater the accuracy is, and vise versa. Search Results sorting rate refers to the positive correlation between the search results and the demand of the users. In order to optimize the sorting results, this paper presents a semantic feature extraction based on genetic algorithm approach.Based on the introduction of semantic search technology, this research discusses the shortcoming of traditional search technology. Starting from purpose of researching and designing, this paper develops a search method based on genetic algorithm to realize semantic search for the genetic algorithm. This paper first uses the binary code as the initial characteristics of the code generation population values; then uses dimensionality reduction as a preprocessing method of semantic feature classification, so as to calculate the fitness function of each chromosome in the population. The fitness function takes the classification accuracy rate to evaluate the value of each individual; Finally, take the individual of the initial population to make selection operator, crossover operator, mutation operator to get an optimized population, termination condition, output the best fitness value of optimal chromosome among the population as the satisfaction solution or optimal solution. Through the research of the user’s behavior patterns of semantic search, based on semantic features extraction method using GA to optimizing the SVM parameters, constructs the search user interaction model of semantic based on genetic algorithm, to examine the results by constructing a theoretical model of experimental method.