Accurately predicting consumer preferences is one of the key success factors in marketing strategies. This research investigated the effectiveness of different artificial intelligence models in predicting consumer preferences in the Tehran market. The required data was collected through a questionnaire including demographic characteristics and purchase preferences from 400 consumers in Tehran. After data preprocessing, random forest, support vector machine (SVM) and logistic regression models were implemented to predict consumer preferences. The results showed that the random forest model has the best performance among the investigated models with an accuracy of 87%. Also, the data analysis showed that the variables of age, income and education have the greatest influence in predicting purchase preferences. These findings can help local businesses to optimize their marketing strategies and achieve better results in attracting customers.
Read full abstract