چکیده مقاله
The content produced in social networks may have different textual, visual, or audio structures Each of these structures can be used to classify generated content A significant number of produced contents have both textual and graphical features Some of them, such as the stories published on Instagram, have the usual text and graphical features In addition to text features, background color, text color, and font as graphical features can be used to improve the accuracy of the classification model In this research, our 3660 Persian data published in Instagram stories have been used for the dataset The data has been divided into 18 different classes by human supervision The 80% of the data has been used for training and 20% remaining for testing the learning model The approach of this research is to use transformer architecture and a multilingual model for text classification and a neural network for graphical features classification and then combine these two classification models in one model based on ensemble learning The obtained results of proposed method show about 10% improvement in accuracy and F1 score respected to text classification
کلیدواژهها
نویسندگان
Pooya Chavoshi Asl
MSc Student Electrical and Computer Engineering Department University of Tabriz Tabriz, Iran
Mohammad Asadpour
Assistant Professor Electrical and Computer Engineering Department University of Tabriz Tabriz, Iran
Pedram Salehpour
Assistant Professor Electrical and Computer Engineering Department University of Tabriz Tabriz, Iran
شیوه ارجاع
Chavoshi Asl, Pooya and Asadpour, Mohammad and Salehpour, Pedram,1401,Topic classification of social networks contents: Text and graphical features fusion using transformer-based architecture,1st International Conference and 6th National Conference on Computers, information technology and applications of artificial intelligence
ارائهشده در
مجموعه مقالات اولین کنفرانس بین المللی و ششمین کنفرانس ملی کامپیوتر، فناوری اطلاعات و کاربردهای هوش مصنوعی3 اسفند 1401 · اهواز