Abstract: - A large number of people use social networking sites such as Twitter to express themselves. As a message of their human feelings, a tweet is considered. The sentiment analysis of user tweets was the subject of our research. Many studies have been conducted on this sentiment analysis, which only uses text found in the user's Tweets and produces good results from the small number of words in Twitter messages. A few studies have been done on this, showing that the emotions expressed in tweets are used to determine the personality of users and the polarities of tweets. In our research, we combined text message information from user tweets with sentiment distribution models to produce a more precise sentimental analysis from user tweets. We have used the miracle of mood change and mood-changing patterns to investigate the spread of emotions. We suggested random forest machine reading to predict the magnitude of the feelings and emotions that the user conveys in their tweets, using both textual information from the user’s tweets and emotional and emotional distribution patterns. This is the first study to use emotional transmission methods to improve Twitter's emotional analysis, to our knowledge. Numerous studies in real-time dataset have shown that, compared with high-quality text analysis algorithms.Keywords: - Machine learning, Sentiment analysis, sentiment diffusion, Text Mining, Twitter