M.Sc. Tezi Görüntüleme

      Especially since the beginning of the 21st century, with the increasing speed of information technology and integration into almost every phase of everyday life, a large amount of data has been collected in many areas. Database management systems are used to store such data in order to utilize them systematically, manage them quickly, and analyze them easily. The vast majority of data in the information world is non-systematic, such as pictures and audio files, text files in pdf, word and txt. formats, e-mails, and log files kept on web pages. Methods such as data mining, text mining, sentiment analysis, image and sound processing are used to extract significant information from these data, which cannot be analyzed by traditional statistical methods. For this study text mining methods have been used since the data analyzed in this study are in text format. The main target of text mining is to separate the texts according to their subjects in order to summarize them, to add their titles and to determine their authors. In this study, a system has been developed to determine the authors of 25 poetry works belonging to Divan Literature. In this system, which is based on analyzing the words by using the text classification algorithms of text mining, 20 different models have been established for the possible values of each parameter. As a result of our individual comparisons of each model, 91.26% accuracy and 90.23% f-value ratings were achieved. Such this study is thought to be able to support estimates of the identification of authors of unknown works in the long term.