Analysis On The Subject And Emotion Of The Medical Forum Of Cerebrovascular Disease

With the development of society, people pay more attention to their own health. They are no longer satisfied with the passive treatment of the disease. Instead, they are turning to the autonomous Management of the health. While, the rise of the medical forum meets this demand to a great extent. In the medical forum, people can put forward their own interests, exchange ideas and discuss with other forum members. At the same time, the medical forum also provides a communication platform for patients who seek medical help and emotional comfort. The in-depth analysis of the medical forum, on the one hand, could grasp the real needs of the forum members and provide a reference to the disease diagnosis and the treatment process. On the other hand, it can optimize the interactive interface and improve the experience of users, which could improve the enthusiasm of the user to participate in and promote the development of medical forum to a great extent.A wide range of health concerns has also increased the enthusiasm in the medical field, and the research point of view is also focused. Domestic medical researches are mostly focused on the improvement of medical system or specific treatment. The researches on the medical forum are more used to adopt the method of investigation questionnaire and content analysis, which could not meet the demand of the growing mass text. The post texts of medical forum are mostly subjective, and each post has two attributes, which is the hot topic and the emotion of people’s attention, so the research can be based on them. Based on this, the article is devoted to use more scientific and effective method, and put the medical forum for cerebrovascular disease as the breakthrough point, to analyze the theme and text sentiment of the medical forum. The main research contents are as follows:Firstly, the theme identification method based on the improved LDA theme model is proposed. By researching the LDA theme model and combining the characteristics of the post text of the medical forum, the improved LDA theme model is proposed. Based on the model, the overall framework of the subject identification is proposed.Secondly, a method based on depth neural network for medical forum text sentiment analysis is introduced. Through the research on the depth of the neural network, we select the specific model and analyze the emotional polarity of the medical forum post text. Based on the model, we propose the whole frame of the text sentiment analysis.Finally, an empirical study is applied to the subject identification and text sentiment analysis. With the help of the post text of the forum in the three nine health network, we use an empirical research on the topic recognition and text sentiment analysis and evaluate the results. In the end, we analyze the results of the subject identification, the emotional polarity and the combination of the two, after which we present the suggestions for the further application of the results.Through the research on the subject identification and text sentiment analysis of the medical forum, the following conclusions are drawn:Firstly, by compared the experimental results with the LDA topic model and the weighted LDA topic model, the improved LDA model is proved suitable for the subject identification field of brain vascular disease medical forum;Secondly, by compared with the experimental results of Naive Bayesian and support vector machine, the application of multilayer perceptron based on deep neural network is proved suitable for the emotion analysis field of the medical forum of cerebral vascular disease;Finally, through the empirical research, we recognize the 6 hot topics in cerebrovascular disease forum of three nine health network which users attention, the sentiment polarity of each post, and the polarity distribution of each subject, which can be applied in the diagnosis and treatment process of the doctor mentioned and the subject and polarity label processing for the optimization of Medical Forum. They also can be used to make an reference for the establishment of the medical forum in medical cloud platform.