engleski

Collecting Big Data in Cinemas to Improve Recommendation Systems - A Model with Three Types of Motion Sensors

With the advent of video-on-demand services on the Internet, research on recommendation systems has shifted to these services. In classic cinemas, recommendation systems are also used, but they are also most often associated with devices connected to the Internet (computers, smartphones). As a rule, these systems collect different data types that users consciously generate, and data that users unconsciously generate are rarely used. The category of unconsciously generated data includes ECG, EEG, GSR, pulse rate, blinking, unconscious body movements etc. The development of the Internet of Things device has enabled mass monitoring of some unconsciously generated data by users intending to improve the recommendation system's accuracy. This paper defines a model for monitoring cinema spectators unconscious body movements with the help of three different non- invasive sensors and based on a simple neural network. The above data can be used for better user profiling, provided that we know where the user is sitting in the cinema. Since various loyalty systems (cards, apps, etc.) are available in cinemas today, seating location information is often known for registered users. The aim of the paper is to define a simple model for monitoring cinema viewers’ unconscious body movements to increase the accuracy of recommendation systems

Body movement ; Cinema ; Recommendation systems

nije evidentirano