Optimizing Semi-Automatic Semantic Matched Searching Concept

Abstract

The Semantic web presents an opportunity to convey the meanings of web documents in a format understandable by machines. However, the majority of web content remains in a format intended for human consumption, and it's anticipated that creators and developers will continue to prefer this format due to its simplicity. To bridge this gap and realize the vision of the Semantic Web, two main approaches have emerged: annotating information sources with machine-accessible semantics or developing programs to extract semantics from web sources. This proposed research aims to identify documents retrieved from web servers based on knowledge extraction using a Semi-automated semantic matching concept. This matching concept aids users in selecting the appropriate document categorized into factual, procedural, and conceptual based on Bloom's taxonomy. The analysis involves iterating through grammatical rules to apply those relevant and determining if a valid stem is found. The SAS algorithm entails complex grammatical rules, such as removing multiple suffixes and prefixes, which can lead to variations in results. One factor influencing the outcome is whether the algorithm requires the output word to be a real word in the given language. Some approaches don't mandate the word's existence in a lexicon database, while others maintain a database of known word roots that are actual words. Optimization is provided by the SAS algorithm through its efficient memory utilization and swift execution, enhancing overall performance.