An Improved Wrapper Generation Using Self Organizing Maps and Meta Heuristic Technique for Web Based Biomedical Data Mining

摘要

With databases becoming increasingly browser accessible searching and extracting data for data mining becomes an important area of research. The task involves extracting important attributes and assigning a label to it. This becomes increasingly useful with the advent on online shopping where prices are compared from many websites by individuals before purchasing. Other applications include comparing generic drugs online, comparison of clinical data by experts to name a few. In this work, an automatic annotation technique based on data alignment is proposed to identify the label associated with it. An improved wrapper generation using Self-Organizing Maps (SOM) and BAT algorithm is proposed. The proposed SOM technique is compared with Visual information aNd Tag structure based wrapper generator (VINT). Experimental results show the effectiveness of the proposed approach in improving the precision by an average of 1.48% when compared with proposed technique with Document Object Model (DOM) alignment.