Evaluating multi-label classifiers and recommender systems in the financial service sector

Autor: Dirk Van den Poel, Matthias Bogaert, Justine Lootens, Michel Ballings
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Technology
Information Systems and Management
General Computer Science
Computer science
MODELS
0211 other engineering and technologies
Social Sciences
02 engineering and technology
Recommender system
Management Science and Operations Research
Machine learning
computer.software_genre
PRODUCT
Industrial and Manufacturing Engineering
CLASSIFICATION
Business and Economics
cross-sell
multi-label classifiers
Business & Economics
Modelling and Simulation
0502 economics and business
Collaborative filtering
Relevance (information retrieval)
AdaBoost
Financial services
CUSTOMER
050210 logistics & transportation
021103 operations research
Science & Technology
business.industry
Operations Research & Management Science
05 social sciences
ALGORITHMS
OR in marketing
Random forest
Management
Statistical classification
CRM
Modeling and Simulation
Artificial intelligence
Classifier chains
recommender systems
business
computer
predictive modeling
Zdroj: Bogaert, M, Lootens, J, Van den Poel, D & Ballings, M 2019, ' Evaluating multi-label classifiers and recommender systems in the financial service sector ', European Journal of Operational Research, vol. 279, no. 2, pp. 620-634 . https://doi.org/10.1016/j.ejor.2019.05.037
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
ISSN: 0377-2217
1872-6860
DOI: 10.1016/j.ejor.2019.05.037
Popis: The objective of this paper is to evaluate multi-label classification techniques and recommender systems for cross-sell purposes in the financial services sector. We carried out three analyses using data obtained from an international financial services provider. First, we tested four multi-label classification techniques, of which the two problem transformation methods were combined with several base classifiers. Second, we benchmarked the performance of five state-of-the-art recommender approaches. Third, we compared the best performing multi-label classification and recommender approaches with each other. The results identify user-based collaborative filtering as the top performing recommender system, with a cross-validated F1 measure of 42.20% and G-mean of 42.64%. Classifier chains binary relevance with adaboost and binary relevance with random forest are the top performing multi-label classification algorithms for respectively F1 measure and G-mean, yielding a cross-validated F1 measure of 53.33% and G-mean of 54.37%. The statistical comparison between the best performing approaches confirms the superiority of multi-label classification techniques. Our study provides important recommendations for financial services providers, who are interested in the most effective methods to determine cross-sell opportunities. In previous studies, multi-label classification techniques and recommender systems were always investigated independently of each other. To the best of our knowledge, our study is therefore the first to compare both techniques in the financial services sector.
Databáze: OpenAIRE
Externí odkaz: