Abstract
Nowadays, the use of micro-blogging services such as Twitter seems to increase exponentially with the popularity of social media. Through these services, users share their opinions, complaints, requests and suggestions about the subjects or institutions and organizations they wish. In this study, it is aimed to develop a system which can detect general opinion s and determine what they are by classifying and summarizing the Turkish tweets sent by users via Twitter. At classification step of this system, SVM and Naive Bayes were used together. The Hybrid TF-IDF method was preferred for the purpose of summarizing classified tweets. According to the results obtained, this developed system has been found to be able to successfully determine the opinions of the users and to get a general idea of what is going on.