Previous researches have indicated that only 20% to 45% of the common search results on the meaning of a particular word are relevant. The search becomes harder when the keyword used is a homograph. Homographic word is a word that is spelled the same as another word but has a different meaning and sometimes a different pronunciation but differs in meaning, derivation, or pronunciations. In a conventional search, the search engine will rank the result using general condition without considering the different context in which the word can be used. This would result in the search matching the wrong intended meaning as not conveyed in the sentence. This research attempts to design an initial dominant user context (DUC) filtering framework for a more personalized search. Previous literatures are reviewed to identify types of data commonly used for personalization search results. This research will develop the search result cluster and dominant user contexts by carrying out empirical studies and testing the framework. To achieve this objective, the research performs two phases of survey. The first survey is for exploratory study. The research develops a search result cluster and dominant user contexts based on the exploratory data and measure the success of DUC Filtering Framework implementation. The second survey is for validation of the DUC Filtering Framework developed from the result of the exploratory study. Result introduces new idea to improve the search result and contribute to the body of knowledge in Distributed and Parallel Information Retrieval area. The results enable the enhancement of personalization search result by matching the user behaviour, interest and ontology of metadata with the search keyword. This is called humanizing the search result, in which it will reduce the gap of knowledge and understanding between humans and computer. The research also would produce a new conceptual understanding of types of data commonly used in developing the personalized search result. The DUC Filtering Framework provides a foundation in developing the algorithm for searching tools. This algorithm will largely benefit search engine as well as website search function in producing personalized search result.
Anitawati Mohd Lokman
[email protected]
Aishah Ahmad [email protected]
UiTM Shah Alam
