Wrapper Semi To Structured Database From Multi Web Site Based On Natural Language Processing

Abstract

The number of data source on internet has increased in volume and type since the last decade, causing problems to query the data or information because of the diversity, dynamic and heterogeneity of the data source or information. Therefore, to simplify the task of obtaining information, several tools have been created for extracting the data from multiple web sources, including Wrapper. Wrapper facilitates the access to Web-Based information sources by providing a uniform querying and data extraction capability. It consists of a set of extraction rules and the code required to apply the rules in order to make the wrapper extracts the right and specified information. The research focuses on how to query the data of rooms and rates hotels in Indonesia by proposed a single wrapper which will change the semi data to structured database based on Natural Language Processing.