Analysis And Design Of The Auxiliary Generation System Of Guide Contents Based On Context And Sentiment Orientation

With the rapid development of the Internet, And because of the features:fast spreading, widely audience, and difficult to control, etc., some problems in the real-world published on the Internet can easily lead to strong pressure of public opinion. This requires Internet organizers for timely and appropriate response and counseling. Current domestic and international aspects of content generation for guided primarily using traditional manual way, which is not only slow, but the cost is relatively high.To solve these problems, this paper proposed an auxiliary generation system of guide contents based on context and emotional orientation, and consolidated planned the overall architecture of the system. The system is divided into three modules:corpus import module, auxiliary artificial generation module and automatic generation module. The key technical aspects of each module are designed in detail. Thus, there forms the complete auxiliary generation system of guide content.The main research work and achievements are:(1) Based on the existing descriptive text, drawing on emotion classification and analysis of sentence context to mine which contains the writing mode, the auxiliary generation of guide content is achieved by the rewritten of sentences based on the machine translation. On this basis, the overall of the system is designed.(2) Support vector machine (SVM) classification algorithm is deeply studied, for the over-fitting problems existing in actual text classification due to documents unevenly distributed on the category level. The text categorization techniques based on adaptive hierarchical is proposed and applied in sentiment classification.(3) Clustering algorithms which are commonly used are comparatively analysis in detail. For the actual clustering demand in the system, the DBScan clustering algorithm is simplified and improved to apply it to the clustering of the sentences and words.(4) The SPG rewriting model is proposed, the rewriting of the sentence is refined to three steps:Sentence pretreatment, Plan of rewriting and Generation. And on this basis, this method is achieved.(5) The auxiliary generation system of guide contents based on context and emotional orientation is designed and implemented, which provides the user with auxiliary artificially generation and automatically generation to decision support.