A semantic based terms relation research for classification in web information mining

Abstract

The rapid growth of web information and its services in different areas such as e-commerce, healthcare, digital marketing, online booking, etc. is a challenge in providing accurate information in the domain services related to the user's query. The current web information of services classifies the retrieval of the relevant service and assists the classification by supporting the knowledge and classifications of the specific service information. Because of these limitations and the complexity of automatic update mechanisms to see this service information, a large number of non-related service information for a requested query, and getting the required web information of services is a cumbersome problem. This paper proposes Semantic based Terms Relation Approach (STRA) for classifying information for effective classification of WIS on the web. The approach utilize Concept Terms Similarity (CTS) method for the most relevant terms in a service domain and construct a Related Terms Hierarchal Model (RTHM), which will be used for classification. A modified Naive Bayes classifier is used to perform the classification of the web information of services using RTHM, to categorize and present accurately. The experiment evaluation of the proposed approach shows an improvement in the classification of information and achieve a highly related matching results against different number of users queries.