Multi-objective Grammatical Evolution of Decision Trees for Mobile Marketing user conversion prediction. (15th April 2021)

• Record Type: Journal Article
• Authors: Pereira, Pedro José; Cortez, Paulo; Mendes, Rui
• Is Part Of: Expert systems with applications. Volume 168(2021)
• Publication Date: 2021-04-15
• Subjects: Conversion Rate (CVR) prediction -- Decision Trees -- Explainable Artificial Intelligence (XAI) -- Grammatical Evolution -- Lamarckian Evolution; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2020.114287
• Languages: English
• ISSNs: 0957-4174

Abstract

Abstract: The worldwide adoption of mobile devices is raising the value of Mobile Performance Marketing, which is supported by Demand-Side Platforms (DSP) that match mobile users to advertisements. In these markets, monetary compensation only occurs when there is a user conversion. Thus, a key DSP issue is the design of a data-driven model to predict user conversion. To handle this nontrivial task, we propose a novel Multi-objective Optimization (MO) approach to evolve Decision Trees (DT) using a Grammatical Evolution (GE), under two main variants: a pure GE method (MGEDT) and a GE with Lamarckian Evolution (MGEDTL). Both variants evolve variable-length DTs and perform a simultaneous optimization of the predictive performance and model complexity. To handle big data, the GE methods include a training sampling and parallelism evaluation mechanism. The algorithms were applied to a recent database with around 6 million records from a real-world DSP. Using a realistic Rolling Window (RW) validation, the two GE variants were compared with a standard DT algorithm (CART), a Random Forest and a state-of-the-art Deep Learning (DL) model. Competitive results were obtained by the GE methods, which present affordable training times and very fast predictive response times. Highlights: We propose two novel methods (MGEDT and MGEDTL) to evolve decision trees. The methods optimize the classification performance and model interpretability. The methods were designed for the Mobile Performance Marketing domain. A realistic experimentation was conducted using big data (6 million records). Competitive results were obtained by the proposed MGEDT and MGEDTL methods.