An application of artificial intelligence for detecting emotions in neuromarketing

Abstract

The subject of this paper is the application of artificial intelligence for detecting emotions in neuromarketing. The goal is to enable the identification of user emotions through a webcam, using convolutional neural networks. The first part of the paper describes the neural networks, the basic types, and their differences. The greatest attention has been given to the description and application of convolutional neural networks. A Convolutional Neural Network, also known as CNN, is specialized in processing data that has a grid-like topology, such as an image. User emotion recognition is enabled using the face-api.js library. It implements the following models: SSD Mobilenet V1, Tiny Face Detector and MTCNN. Tiny Face Detector, used in the application, is a model for real-time face detection with small size, speed, and moderate resource consumption. The model is compatible with the web and mobile platforms. In the second part of the paper, an application was developed, which uses the face-api.js library to detect emotions. It has been developed as a tool to support neuromarketing research. It allows the marketer to create research to analyze advertising material. Its basic functionality is to display advertising content and collect data while watching. Data is stored and graphically displayed to the marketer. This section describes in detail how the detection process works. In the third part of the paper, evaluation was made. Evaluation of the developed solution was performed by experiment. The results show that the emotions of the user can be recognized by the developed system, with a satisfactory level of precision. The advertising content has previously entered parameters, which represent the desired results. By comparing these parameters and the obtained results, the marketer decides whether the advertisement is successful.