Abstract
The development of information systems (IS) has increased in the e-commerce field. The need for continuous improvement of decision support systems implies the integration of multiple methodologies such as expert knowledge, data mining, big data, artificial intelligence, and multicriteria decision analysis (MCDA) methods. Artificial intelligence algorithms have proven their effectiveness as an engine for data-driven information systems. MCDA methods demonstrated usefulness in domains dealing with multiple dimensions. One of the most critical points of any MCDA procedure is criteria weighting using subjective or objective methods. However, both approaches have several limitations when there is a need to map the preferences of unavailable experts. EVO-SPOTIS library integrating a stochastic evolutionary algorithm with the MCDA method, introduced in this paper, attempts to address this problem. In this approach, the Differential Evolution (DE) algorithm is used to identify decision-makers’ preferences based on datasets evaluated by experts in the past. The Stable Preference Ordering Towards Ideal Solution (SPOTIS) method is used to compute the DE objective function’s values and perform the final evaluation of alternatives using the identified weights. Results confirm the high potential of the library for identification preferences and modeling customer behavior.
Paper Type
Full Paper
DOI
10.62036/ISD.2022.35
Python Library for Consumer Decision Support System with Automatic Identification of Preferences
The development of information systems (IS) has increased in the e-commerce field. The need for continuous improvement of decision support systems implies the integration of multiple methodologies such as expert knowledge, data mining, big data, artificial intelligence, and multicriteria decision analysis (MCDA) methods. Artificial intelligence algorithms have proven their effectiveness as an engine for data-driven information systems. MCDA methods demonstrated usefulness in domains dealing with multiple dimensions. One of the most critical points of any MCDA procedure is criteria weighting using subjective or objective methods. However, both approaches have several limitations when there is a need to map the preferences of unavailable experts. EVO-SPOTIS library integrating a stochastic evolutionary algorithm with the MCDA method, introduced in this paper, attempts to address this problem. In this approach, the Differential Evolution (DE) algorithm is used to identify decision-makers’ preferences based on datasets evaluated by experts in the past. The Stable Preference Ordering Towards Ideal Solution (SPOTIS) method is used to compute the DE objective function’s values and perform the final evaluation of alternatives using the identified weights. Results confirm the high potential of the library for identification preferences and modeling customer behavior.
Recommended Citation
Watróbski, J., Baczkiewicz, A., & Rudawska, I. (2022). Python Library for Consumer Decision Support System with Automatic Identification of Preferences. In R. A. Buchmann, G. C. Silaghi, D. Bufnea, V. Niculescu, G. Czibula, C. Barry, M. Lang, H. Linger, & C. Schneider (Eds.), Information Systems Development: Artificial Intelligence for Information Systems Development and Operations (ISD2022 Proceedings). Cluj-Napoca, Romania: Risoprint. ISBN: 978-973-53-2917-4. https://doi.org/10.62036/ISD.2022.35