Analysis of brand image effect on advertising awareness using a neuro-fuzzy and a neural network prediction models

Almost all the worldwide and nationwide companies utilize advertising to increase their sales volume and profit. These companies pay millions of dollars to reach consumers and announce their products or services. This forces companies to evaluate advertising effects and check whether ads meet companys strategies. They need to evaluate the ads not only after announcement, but also before advertising, i.e. they can be one step ahead by predicting the future advertising awareness through artificial intelligence tools such as fuzzy systems and neural networks. In this study, we propose to use adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN) to analyze advertising decision making. ANFIS creates fuzzy rules and trains the neural network using given input data. This training ability of ANFIS and ANN leads to predicting the advertising awareness outputs. Here, we investigate three advertising awareness outputs, namely, top of mind, share of voice, and spontaneous awareness. In order to achieve the valid predictions, data are randomly divided into training data with 70 percent, validation data with 15 percent, and testing data with remained 15 percent of data. The correlation between actual data and predictions are calculated to check the accuracy of the predicted outputs.