Twitter'da veri analizi

With the development of technology, the internet has reached our pockets and innovations have come with it. Social media has become a very important media tool with the development of the internet. Since people from all over the world use this medium, it has become a huge data source. Twitter accounts for a large part of this data formation. With text mining, we can take the raw text from Twitter, preprocess it and get it structured. After taking the structured form of the text, it is passed to the sentiment analysis part. Sentiment analysis grades a text as positive, negative, or neutral based on the emotion given by the words. By looking at the text in general, we decide whether the positive aspect of this text is more weighted or the negative aspect of it. In this thesis, first text mining and then sentiment analysis of the data we pulled from Twitter were made. We can easily pull tweets under the name of Twitter's Covid19 and data mining hashtags by opening a developer account. The text mining of the data collected for about a month was performed, and then the sentiment analysis step was started and the sentiment scores of the data were examined. While performing text mining and sentiment analysis, Python software language and Jupyter Notebook were used. To conduct the research, lessons were taken from an online site and learned to a certain point. In the application part of the thesis, how we pull the data with Python coding, which date range is used, the codes we write for the preprocessing of the data are mentioned.