Real-Time Customer Behavior and Satisfaction Insight System in Shopping Malls

The research this research was tasked to perform on the different technologies; Pose Network and Move-Net, to monitor the customers in shopping malls, with other studies pointing out the weaknesses of the systems in exactly identifying and interpreting customer interests is the study. Basically, these systems--mainly, pose detection systems--are used to set customer attributes and ignore other multiplicity of behaviours of the customer, which may be potentially influencing those attributes it is Concurrently, the other option that we are providing to this problem which is the integration an advanced person detection framework known as YOLO with Deep SORT algorithms that can tell the customer movement direction precisely is offering improvement. The technical aspect of our model we have the ability to increase the accuracy of the proposed model as well as its flexibility and adaptability as compared to the methods which are in existence. Also, we provide fast and anonymous PC feedback collection through QR codes that lets us to determine the final results of the survey chatter of customer interactions. The report of our attempt to clean code will enable the operators of shopping malls to know customer preferences, annoyances, and satisfaction levels with the use of computer vision and machine learning. A well-grounded decision will be the starting point of data-driven approach that will profit the customers through operational efficiency of the mall.