The Fuzzy-Go search engine develops a fuzzy ontology to capture the similarities of terms in the ontology for accomplishing the semantic search of keywords, a web crawler to gather and classify web pages, and a fuzzy search mechanism to aggregate all fuzzy factors based on their degrees of importance and degrees of satisfaction. In this paper, we apply the genetic algorithm to propose a self-adaptation approach to Fuzzy-Go search engine. For each search, the fuzzy search engine records the difference between the ordering of search results and user's real behavior on clicking web pages. The feedbacks are gathered and analyzed to adjust the fuzzy similarities between terms in the fuzzy ontology, the domain classification of web pages, and the importance degrees of fuzzy factors. The ordering of search results can thus be improved gradually by continuous learning and adaptation.