TY - GEN
T1 - Metaheuristics Strategies for Trade Data Harmonization
T2 - 14th and 15th International Joint Conference on Computational Intelligence, IJCCI 2022 and IJCCI 2023
AU - Khargharia, Himadri Sikhar
AU - Shakya, Sid
AU - Ruta, Dymitr
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - Harmonizing trade data from diverse datasets with varied product categories poses a substantial challenge due to differences in trade volume representation. Discrepancies arise from distinct subcategory structures in datasets, leading to disparities in traded volume. This study focuses on devising an approach to harmonize product subcategory selection by comparing volumes across datasets. Metaheuristic techniques: Genetic Algorithm (GA), Population-based Incremental Learning (PBIL), Distribution Estimation using MRF (DEUM), and Simulated Annealing (SA) are employed to address the intricate challenge of aligning subcategory volumes across sources while ensuring the agreement of selected subcategories. Evaluation of solutions considers fitness, scalability, and technique-specific strengths and weaknesses. Multiple instances of trade data harmonization are examined to assess the applicability of these techniques in mitigating trade-volume disparities. The study provides insights into the efficacy of metaheuristic techniques addressing complexities of harmonizing the trade data with inconsistent subcategory structures across datasets. Results contribute to the understanding of effective strategies for achieving alignment in hierarchical trade data.
AB - Harmonizing trade data from diverse datasets with varied product categories poses a substantial challenge due to differences in trade volume representation. Discrepancies arise from distinct subcategory structures in datasets, leading to disparities in traded volume. This study focuses on devising an approach to harmonize product subcategory selection by comparing volumes across datasets. Metaheuristic techniques: Genetic Algorithm (GA), Population-based Incremental Learning (PBIL), Distribution Estimation using MRF (DEUM), and Simulated Annealing (SA) are employed to address the intricate challenge of aligning subcategory volumes across sources while ensuring the agreement of selected subcategories. Evaluation of solutions considers fitness, scalability, and technique-specific strengths and weaknesses. Multiple instances of trade data harmonization are examined to assess the applicability of these techniques in mitigating trade-volume disparities. The study provides insights into the efficacy of metaheuristic techniques addressing complexities of harmonizing the trade data with inconsistent subcategory structures across datasets. Results contribute to the understanding of effective strategies for achieving alignment in hierarchical trade data.
KW - Distribution estimation using MRF
KW - Genetic algorithm
KW - Population-based incremental learning
KW - Simulated annealing
KW - Trade data harmonisation
UR - https://www.scopus.com/pages/publications/105002018380
U2 - 10.1007/978-3-031-85252-7_14
DO - 10.1007/978-3-031-85252-7_14
M3 - Conference contribution
AN - SCOPUS:105002018380
SN - 9783031852510
T3 - Studies in Computational Intelligence
SP - 240
EP - 264
BT - Computational Intelligence - 14th and 15th International Joint Conference on Computational Intelligence IJCCI 2022 and IJCCI 2023, Revised Selected Papers
A2 - Bäck, Thomas
A2 - van Stein, Niki
A2 - Wagner, Christian
A2 - Garibaldi, Jonathan M.
A2 - Marcelloni, Francesco
A2 - Lam, H.K.
A2 - Cottrell, Marie
A2 - Doctor, Faiyaz
A2 - Filipe, Joaquim
A2 - Warwick, Kevin
A2 - Kacprzyk, Janusz
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 13 November 2023 through 15 November 2023
ER -