Facial Expression Recognition in The Deep-Learning-Based Business Intelligence
محورهای موضوعی : International Journal of Finance, Accounting and Economics Studies
kamelya dehghani kohnehshahri
1
,
Mohammad Ali Afshar Kazemi
2
,
alireza Puorebrahimi
3
1 - Ph.D. Student, Department of Information Technology Management, Science and Research Branch, Islamic Azad University, Tehran, Iran
2 - Associate Professor, Management Group, Faculty of Management ,Tehran North Branch, Islamic Azad University, Tehran, Iran
3 - Assistant Prof, Department of Industrial Management, Faculty of Management and Accounting, Islamic Azad University, Karaj, Alborz, Iran
کلید واژه: Businesse Intelligence, Facial Expression Recognition, Sentiment Analysis, Image Processing, Deep learning.,
چکیده مقاله :
purpose: This research purpose to analyse data to predicting customer behavior using facial expression detection in intelligent businesses based on deep learning. currently among researchers, presenting methods that increase accuracy and efficiency in predicting customer behavior in intelligent businesses is important and has tremendous effects on e-commerce profitability, marketing, sales, economy, stock prediction and etc. Methodology: The methodology of this research are descriptive analytical and practical in terms of research objectives. The research was conducted using image processing and machine vision techniques to predict customer behavior and detect facial expressions to improve the performance of intelligent businesses in organizations and social networks. The information gathering tool was a library as well as using TensorFlow library in Google Colab environment and Python programming to examine research topics and subjects using qualitative and quantitative content analysis. Findings: By comparing the outputs obtained from image processing, it can be said that among the seven facial expressions, happiness has a more effective role with 67.3% in utilizing business intelligence and profitability, while disgust with 9.2% had the least impact on predicting customer behavior. In addition to ranking the facial expressions in order of priority, the research aimed to improve accuracy and reduce errors in detecting customer behavior and analysing emotions on social networks using a combined algorithm designed in the conceptual model of this study. The accuracy of analysing user emotions in intelligent businesses on Instagram increased from 77% to 96%, surpassing previous research studies.
purpose: This research purpose to analyse data to predicting customer behavior using facial expression detection in intelligent businesses based on deep learning. currently among researchers, presenting methods that increase accuracy and efficiency in predicting customer behavior in intelligent businesses is important and has tremendous effects on e-commerce profitability, marketing, sales, economy, stock prediction and etc. Methodology: The methodology of this research are descriptive analytical and practical in terms of research objectives. The research was conducted using image processing and machine vision techniques to predict customer behavior and detect facial expressions to improve the performance of intelligent businesses in organizations and social networks. The information gathering tool was a library as well as using TensorFlow library in Google Colab environment and Python programming to examine research topics and subjects using qualitative and quantitative content analysis. Findings: By comparing the outputs obtained from image processing, it can be said that among the seven facial expressions, happiness has a more effective role with 67.3% in utilizing business intelligence and profitability, while disgust with 9.2% had the least impact on predicting customer behavior. In addition to ranking the facial expressions in order of priority, the research aimed to improve accuracy and reduce errors in detecting customer behavior and analysing emotions on social networks using a combined algorithm designed in the conceptual model of this study. The accuracy of analysing user emotions in intelligent businesses on Instagram increased from 77% to 96%, surpassing previous research studies.
