نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه مدیریت فناوری اطلاعات، واحد تهران مرکزی، دانشگاه آزاد اسلامی
2 گروه مدیریت صنعتی، واحد تهران مرکزی، دانشگاه آزاد اسلامی، تهران، ایران.(نویسنده مسئول)
3 گروه مدیریت، پژوهشگاه علوم انتظامی و مطالعات اجتماعی، تهران، ایران.
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
The age we live in is the age of information, and the most important issue for organizations is the mastery of this information. With the ever-increasing growth of news in the digital world and the Internet, the issue that becomes important is the classification of this information and our quick and cheap access to it. This importance cannot be achieved except by doing the methods referred to as text classification. The purpose of this research is to classify news texts into predefined categories, which is done using the automatic model tool, which is considered one of the subsets of text mining. Considering the importance of the subject and the work that has been done in this field for other languages of the world, the need to classify Persian texts is well felt. It is noteworthy that research has been developed and used for English texts, but since the Persian language has structural complexities compared to other languages and also less research has been done in this field, this research is of an applied type. It is a development that can be done using the experimental research method and the use of text mining tools, as it is done in a completely controlled environment with the ability to keep other variables constant. In the intelligence society, the classification of texts is done manually by elite people. It seems impossible to categorize texts with this volume manually, so we are forced to look for methods to automatically categorize texts. On the other hand, storing, processing and analyzing this amount of information has become a serious challenge. Due to the high volume of news, data, information, documents and the complexity of maintaining and maintaining them, it is necessary to use a system to manage receiving, maintaining and maintaining existing news. The complexity of organizations creates the need for centralization of news, documents, correct classification, correct circulation of news and ease of access to them. Document management provides the possibility for information organizations to correctly classify received or existing news and documents, preserve, maintain and retrieve them. By examining, analyzing and processing in this research, we come to the conclusion that the accuracy and results of the proposed method on online news texts show; The support vector machine model has 93.29 precision, 93.32 accuracy, 92.96 recall, and 6.71 error.
کلیدواژهها [English]
categorization.
4(7), 2101-2104.