We propose a perception-based search methodology established on fuzzy semantic for improving the web search. Unlike most existing relevant work that focuses on filtering the returned search result, we study the search query semantic and expand the search to include information related to the linguistic and numerical fuzzy counterparts of the search query. The top 20 search results returned from this search methodology showed impressive improvement over the conventional one. Fifty percent average gain in search relevancy is obtained when our search methodology is applied to websites matching the chosen fuzzy semantic theme. We demonstrate the effectiveness of our methodology on the search domain of health and food.