| Due to the rapid development technology of network information, the major employment websites can provide job hunters with lots of information, while people don't need to worry about the lacking information any more. However, when facing with the increasing large number of network about employment and personnel information on the Internet, artificial intelligence technologies is urgently needed. In order to improve the timeless and success rate of job-hunting, it is necessary to deeply analysize the massive data information and quickly sort and extract the useful knowledge which is helpful to those graduates.In this paper, we use the method of text mining to efficiently deal with the massive recruitment information from the filed of electronic employment. Main contributions of the dissertation include:(1)This paper presents an QI-DW information extraction algorithm(QIDWDS):according to the information research technology, we sort the interface which has the Deep Web based on the Deep Web information sorting algorithm, resulting in sorting the webpage information hidden in the background database.(2) A DI-DOM information extraction algorithm (DIDOME) is presented:according to the dynamic feature about web information, an regular expression algorithm based on the DOM Tress is proposed to precompute the web-text data.(3) A system about artificial text mining is developed:Design and implemented a typical three-layer C/S structure of the text mining system, using the support vector machine classification method to achieve an application system oriented to the field of employment for the electronic employment on the text mining.At the same time, in order to satisfy practical applications,we created a special corpus on the professional field of employment and improved some places about the Tianjin segmentation participle system,at the same time give a special pretreatment to the feature selection and computation about the text information. |
PDF Full Text Request