Research Of Music Emotional Classification Oriented Chinese Lyrics

With the development of information technology, China's online music market develops more rapidly, a variety of music products spread through the Internet and mobile communication network widely, and gradually form a standardized patterns of digital music products to product, communicate and consume. It promotes the development of network cultural industry, and enriches people's entertainment and cultural life. At the same time, with the using of a large number of online music, the demand of how to organize and retrieve music become increasingly important. There are many ways to organize, classify and retrieval music, in which it is a way to solve the above problems that is to classify music according with the different emotions. And this has been related in existing research. Especially, in recent years, researchers paid more attention in the studies of music emotion. But for music emotions is a feeling which is complex and subjective, itself has a certain ambiguity, which bring some difficulties in the research of music emotional problems by using the computer to calculate and organize.Currently, the classification of music emotions is generally analyzed by audio.To research the emotions of the music by analying the audio characteristics which reflect human's emotions. In this paper, the research oriended Chinese lyrics of music, combined with knowledge of linguistics, psychology, music, computer and other areas, to study using computer text classification,statistical methods and semantics, in order to classify the song which have Chinese lyrics according to emotion.This paper describes the background knowledge of music emotion classification firstly. Describe the relationship between music and emotion, and classify the emotion expressed in music by combining with the knowledge of music, the latest sentiment classification results of the field of psychology, and considering the actual situation in this study. Then introduce the related technologies of Chinese text classification,which mainly include the steps of Chinese text classification, the knowledge of statistical techniques and semantic. Choose the appropriate existing results to assist this study by comparing the mature technical achievementsBased on the above, to make the emotion word-Category table by combining with the knowledge of music, psychology, linguistics. And to expand this table by using the technology of semantics.Then do emotion classification experiment oriented chinese lyrics for music During this experiment, the first step is to store chinese lyrics as. txt format; The second step is to use ICTCLAS 2011 which is a Word Segmentation System and developed by Chinese Academy of Sciences to segment the words; The third step is to get the emotional words by extracting the features according to the results of word segmentation; The fourth step is to receive the weight of characteristics words by using statistical methods to count the characteristics words; The fifth step is to classify the emotion words into their category, by analyzing and understanding the feature vocabulary; The sixth step is to calculate the weight of each emotion category, according to certain rules and the emotional vocabulary which is included in each categories; The seventh step is to get the final results of emotional categories in this reasearch, take the higher weight-based categories of subcategories as the final result according to certain rules.Finally, consider the performance of the classification designed in this paper, by analyzing the results of this experiment. Furthermore, to summarize this study, and make follow-up prospects for the inadequacies which appear in the research process.