Design And Implementation Of Hotel Reviews Classification System Based On The Maxinum Entropy

With the rapid development of Internet technology,due to the sharp decline in the cost of production of computers,smart phones and other electronic products,3G and 4G wireless technologies become more mature,network speeds increase and the high coverage,using the network terminal users are increasingly and more, the Internet data generated into explosive growth of data and users which do be on the Internet are also produced in accordance with lazy getting stronger,the user is difficult to quickly obtain accurate information from such the large data. So for such a large data such as automatic classification,organization and management methods have become an important research topic. Automatic text classification referred text categorization, which is the direction of natural language processing(NLP) and pattern recognition are more tightly bound direction of research in this area has proven a better classification results Classifier maximum entropy classifier,support SVM (support vector machine) classifier,Naive Bayes classifiers.Internet traditional text classification is based on the content;Purpose of the study is to these texts into various types of entertainment,sports and politics. With the rapid development of the mobile Internet,The classification is far from meeting the needs of users,independent of individual users in the mobile Internet on the emotional color products more weight, which evolved into an Internet important research direction-sentiment analysis, which is the Recent research focus at home and abroad, the main task of sentiment analysis is the use of computer help users quickly organize and analyze information related to the evaluation of subjective emotional color of text with analysis and processing,induction and inference,The emotion Classification,views extract, opinions and ideas discussed and other questions and answers. The rapid development of mobile Internet has also led to growth in the customer network consumption,the natural formation of a large number of user comments on the consumption of products on the Internet,as these will increase in the number of short text,automatic text classification techniques emotional tendency more and more attention.