Predicting Effectiveness of Marketing through Analyzing Emotional Context in Advertisement using Deep Learning

In this modern age, marketing strategy is becoming a new challenge for both the local and worldwide companies. Basically, both the local and global brands are trying to increase their product selling rate and grab the attention of buyers. Therefore, they are promoting advertisements on every media platform. However, they are not aware of the utilization of emotional states in audio or video advertisements. Hence, the effectiveness of marketing is decreasing day by day. Therefore, we lead this study to recognize a successful advertisement and identify the rate of the emotional states which make a good impact in people's mind to purchase the product. To find out the emotional states, we have implemented deep learning and supervised machine learning algorithms as well as feature extraction methods such as LSTM-RNN, XGBOOST, Naive Bayes, Multiple Linear Regression, MFCC, Zero-Crossing Rate, Power Spectral Density and Short Time Energy. After that, we have evaluated the rate of the emotional states to Figure out the liking and purchase intent which makes an advertisement successful.