نوع مقاله : مقاله پژوهشی
نویسنده
استادیار گروه مدیریت صنعتی، دانشکده مدیریت، دانشگاه آزاد اسلامی واحد الکترونیکی، تهران،
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسنده [English]
The advent of the web and its continuous growth has created a huge amount of information generated by users. In this information, valuable mental information can be easily found, especially in social networks and e-commerce platforms that contain important information about users. As a result, the field of belief mining has attracted considerable attention in recent years. Many new research papers are published every day in which various artificial intelligence techniques are applied to various tasks and programs related to ideology. In this work, a new approach based on support vector machine technique and gray wolf optimization algorithm was proposed to improve the belief mining process. In this system, the gray wolf algorithm is used to determine the effective features in the belief mining process and improves the performance of the support vector machine. The results of this research showed that the proposed system was able to help increase the accuracy and cover the error of the backup vector technique by selecting effective features. The proposed system was evaluated using three criteria of accuracy, recall and precision, which accuracy for the first and second class is 0.68 and 0.92%, respectively, recall for the first and second class is 0.94, respectively. and 0.63% and the accuracy was 0.77%. The results indicate that the proposed system of this research was able to achieve favorable results in both classes.
کلیدواژهها [English]
)