Artificial Intelligence Marketing: An application of a novel Lightly Trained Support Vector Data Description. Issue 5 (2nd September 2016)

• Record Type: Journal Article
• Authors: Rekha, A. G.; Abdulla, Mohammed Shahid; Asharaf, S.
• Is Part Of: Journal of information & optimization sciences. Volume 37:Issue 5(2016)
• Page Start: 681
• Page End: 691
• Publication Date: 2016-09-02
• Subjects: AIM -- Direct Marketing -- SVDD -- SVM -- Classification; Electronic data processing -- Periodicals; Information science -- Periodicals; Mathematical optimization -- Periodicals; 519.6
• DOI: 10.1080/02522667.2016.1191186
• Languages: English
• ISSNs: 0252-2667

Abstract

Abstract: Artificial Intelligence Marketingis a form of direct marketing which leverages the techniques of database marketing along with the concepts and models of AI such as machine learning. Direct marketing can increase marketing efficiency by directing the efforts specifically towards the right customer.Even though the popularity of direct marketing campaigns is increasing over time, the selection of contacts is becoming more rigorous, and it has become essential to reduce the number of contacts to do, due to economic factors as well as consumer privacy issues. Direct marketing analytics requires algorithms that are efficient and scalable. In this study we explore the use ofSupport Vector Data Description(SVDD) to facilitate the selection of contacts. Support Vector based classification methods have been proved to be successful in a variety of applications; but they tend to take more computing time when dealing with large scale data. Here we apply a novel low complexity SVDD method for direct marketing. Experiments on a real-world dataset related to a direct marketing campaign have demonstrated that SVDD based methods can provide promising accuracies when compared to other approaches. Also, the accuracy achieved by the proposed low complexity method is comparable with theclassical SVDD and it has huge advantage over it in terms of the execution time.