Blogs Search Engine Adopting RSS Syndication Using Fuzzy Logic

The rapid development of Internet increases the writers of blog sites. Sometimes these blog sites focused on solving some important problems. To find specific blogs are hard problem for the users because a lot of these blogs contain unuseful information such as online advertisements, notice and noise which minimize the rank of blog site. Furthermore to retrieve more relevant blogs is another problem which lowering the search performance. This study proposes blogs search engine adopting RSS syndication using Fuzzy logic. The blogs search engine consists of three main phases which are crawling using RSS feeds algorithm, indexing weblogs algorithm and searching technique with Fuzzy logic. In RSS crawling process RSS feeds need to be gathered to extract useful information such as title, links, publish time and description. Indexing weblogs use the links to retrieve the blogs sites for text processing and construct indexing database. In order to retrieve such information needed by any user, there is user interface to search for keyword with importance degree and compute the density of keyword from the indexing database. The rank of the pages is computed based on fuzzy weighted average value. A prototype is built using visual basic 2008 to validate the proposed blogs search engine. It is a windows application with http connection protocol. In system evaluation used two measurement performances which are precision and mean average precision. The parameters of precision determine based on respondents whom determine the total retrieved links and the total relevant links for the keyword search result. The number of keywords that used in testing system is five pairs keywords. The experimental results show that the mean average precision is 81.7% of the whole system performance. The percent of respondents is 80% who knows and uses the blogs and 20% don’t have knowledge. The execution time of the system based on respondents is 70% between 3-5 minute and 30% less than 3 minute. This percentage is good considering the rate of satisfaction for system is 80% satisfied and 20% strongly satisfied.