Design A Enterprise Information Search Engine And Research Its Key Technology

Vertical search engine technology has gradually used in user life as worthless status, users requirements of specific information search gradually be thinning, but the vertical search engine about the enterprise information are not improved. Through analyzing its requirements, designing and researching its key technologies of the enterprise information search engine, established an enterprise information search engines that can provide users with the enterprise information and its relevant information.1. A systematic study on hidden markov model (HMM) Web information extraction. Set up enterprise information hidden markov models, by using the main algorithms of hidden markov model solve training and decoding problems to extract the web information. Meanwhile optimize the model's parameters and extract again. Experimental results show the optimization hidden markov model extraction effect is better.2. For enterprise information search engine Chinese word segmentation technology is studied. Based on the positive maximal matching words segmentation algorithm, optimize the dictionary mechanism, design a double-word double layers and independent long word list in the dictionary mechanism. Evaluate this mechanism by threes aspects: occupy space, inquires the time and diction maintain. Establish the enterprise information professional dictionary according to this mechanism, and used in the enterprise information search engines.3. On the basis of the the open source architecture and application two key technologies: information extraction and the Chinese word segmentation set up enterprise information search engines. Experiment shows the improved enterprise information search engines can easily meet user's requirement for enterprise information search service.