This paper presents our implementation techniques for a personalized semantic search engine. The
system includes several components such as a crawler, a preprocessor, searcher and ranking module.
PSSE uses multi-crawlers to traverse web to gather resources. The preprocessor is used to identify
crawled page importance based on link analysis techniques, annotate resources using agents that also
mine document content and determine term importance. In this process, natural language processing
techniques (NLP) i.e. stop-word removing and word stemming are applied to the raw resources.
Searcher is responsible in turn for query completion activities making use on ontology as well as
maintaining a log from users’ search activities. Finally, the query engine delivers search results
ranked based on a final score calculated based on traditional link analysis, content analysis and a
weighted user profile. In this paper we evaluate an implementation of PSSE using traditional
information retrieval performance measures namely, precision, recall and F-measure. Results of this
implementation have shown that PSSE worked efficiently.