Panelist atama problemi için bir algoritma ve karar destek sistemi: TÜBİTAK örneği

Öz

Proje önerilerinin sağlıklı bir şekilde ve proje konusunda uzmanlığa sahip kişiler tarafından değerlendirilmesi hem kaynakların etkin bir şekilde kullanılması hem de bu hizmetleri yürüten kurumların güvenilirliği açısından önem taşımaktadır. Bu çalışmada, birden fazla proje önerisinin değerlendirildiği panellerde en uygun panelist kümesinin oluşturulması için potansiyel panelist adaylarını dinamik bir şekilde listeleyen bir algoritma ve bu algoritmayı kullanan bir karar destek sistemi (PaneLIST) geliştirilmiştir. MS Excel VBA tabanlı PaneLIST, ülkemizde araştırma desteklerininin önemli bir bölümünü sağlayan ve bunu gerçekleştirirken her yıl çok sayıda panel düzenleyen TÜBİTAK’ın verileri kullanılarak oluşturulan test panellerinde sınanmış ve elde edilen sonuçların öngörülen kriterlerin tamamını önemli ölçüde sağladığı görülmüştür. Ayrıca, PaneLIST’in sunduğu çözümler, uygunluk skorları toplamını en büyükleyen (EBSkT) ve proje önerilerinin değerlendirme seviyeleri arasındaki sapmaları en küçükleyen (EKSp) tamsayılı programlama modelleri ile birlikte bu iki durumu bir arada ele alan üçüncü bir modelden (EBSkT-5) elde edilen kesin ve en iyi çözümleri ile kıyaslanmıştır. PaneLIST’in, yüksek uygunluk skorları toplamını, projeler arasında dengeli bir dağılım gözeterek gerçekleştirdiği ve bu yönüyle EBSkT ve EKSp’de yer alan iki hedefi bir arada gözettiği; %5’lik bir sapma kısıtı altında en büyük skoru bulmayı amaçlayan problem (EBSkT-5) ile hemen hemen aynı (%1’in altında yakınlık) performansı gösterdiği belirlenmiştir.

Anahtar Kelimeler

• Akran değerlendirmesi
• panel değerlendirmesi
• karar destek sistemi
• hakem atama problemi
• tamsayılı programlama
• sezgisel algortima
• dinamik panelist atama

An Algorithm and a Decision Support System for the Panelist Assignment Problem: The Case of TUBITAK

Öz

Evaluation of project proposals in a proper manner and by the people who have expertise on the topics of the proposals is crucial not only for efficient deployment of resources, but also for credibility of the funding organizations. In this study, an algorithm and a decision support system (PaneLIST) are developed to provide a dynamic list of potential panelists from which the most appropriate set of panelists will be selected. PaneLIST, which is based on MS Excel VBA, has been validated by using the data of TUBITAK, primary organization responsible for research funding and conducts the comprehensive peer review activities for a long time. The results showed that the PaneLIST satisfies the required criteria to a great extent. Moreover, PaneLIST’s performance was compared with the results of the two integer programming models having the objectives of maximizing the sum of relevance scores (EBSkT) and minimizing the total deviation among the evalution levels of the proposals (EKSp) as well as a third model (EBSkT-5) which couples the two. The numerical experiments showed that PaneLIST attains high sum of relevance scores with a balanced distribution in terms of evaluation levels of proposals, thus shows regard to objectives of both EBSkT and EKSp at the same time; moreover, the results are so close (less than 1%) to the results of EBSkT-5 in which sum of relevance scores is maximized under a 5% constraint on the total deviation among the evalution levels of the proposals.

Anahtar Kelimeler

• peer review
• panel evaluation
• decision support system
• reviewer assignment problem
• integer programming
• heuristic algorithm
• dynamic panelist assignment

Kaynakça

1. Aksop, C. (2018). Akademik Makale Değerlendirmesi Kapsamında Hakem Atma Süreçlerinin İncelenmesi ve Yeni Bir Sistem Önerisi. TÜBİTAK.
2. Bouajaja, S., & Dridi, N. (2017). A survey on human resource allocation problem and its applications. Operational Research, 17(2), 339–369. https://doi.org/10.1007/s12351-016-0247-8
3. Ceylan, D., Saatçioǧlu, Ö., & Sepil, C. (1994). An algorithm for the committee construction problem. European Journal of Operational Research, 77(1), 60–69. https://doi.org/10.1016/0377-2217(94)90028-0
4. Cook, W. D., Golany, B., Kress, M., Penn, M., & Raviv, T. (2005). Optimal allocation of proposals to reviewers to facilitate effective ranking. Management Science, 51(4), 655–661. https://doi.org/10.1287/mnsc.1040.0290
5. Daş, G., & Göçken, T. (2014). A fuzzy approach for the reviewer assignment problem. Computers & Industrial Engineering, 72, 50–57. https://doi.org/10.1016/J.CIE.2014.02.014
6. Dell’amico, M., & Martello, S. (1997). Linear assignment. In S. Martello, M. Dell’amico, & F. Maffioli (Eds.), Annotated bibliographies in combinatorial optimization. Chichester, England: John Wiley & Sons Ltd.
7. Dumais, S., & Nielsen, J. (1992). Automating the assignment of submitted manuscripts to reviewers. Proceedings of the 15th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval.
8. ESF. (2011). European Peer Review Guide. Strasbourg.

9. Fan, Z., Chen, Y., Ma, J., & Zhu, Y. (2009). Decision support for proposal grouping: A hybrid approach using knowledge rule and genetic algorithm. Expert Systems with Applications, 36(2), 1004–1013. https://doi.org/10.1016/J.ESWA.2007.11.011
10. Garg, N., Kavitha, T., Kumar, A., Mehlhorn, K., & Mestre, J. (2010). Assigning Papers to Referees. Algorithmica, 58(1), 119–136. https://doi.org/10.1007/s00453-009-9386-0
11. Goldsmith, J., & Sloan, R. (2007). The AI conference paper assignment problem. Proc. AAAI Workshop on Preference Handling for Artificial Intelligence. Vancouver.
12. Gupta, S., & Punnen, A. (1988). Minimum deviation problems. Operations Research Letters, 7(4), 201–204. https://doi.org/10.1016/0167-6377(88)90029-6
13. Hartvigsen, D., Wei, J., & Czuchlewski, R. (1999). The Conference Paper-Reviewer Assignment Problem. Decision Sciences, 30(3), 865–876. https://doi.org/10.1111/j.1540-5915.1999.tb00910.x
14. Hettich, S., & Pazzani, M. (2006). Mining for proposal reviewers. Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD ’06, 862. https://doi.org/10.1145/1150402.1150521
15. Janak, S., Taylor, M., Floudas, C., Burka, M., & Mountziaris, T. (2006). Novel and Effective Integer Optimization Approach for the NSF Panel-Assignment Problem: A Multiresource and Preference-Constrained Generalized Assignment Problem. Industrial & Engineering Chemistry Research, 45(1), 258–265. https://doi.org/10.1021/ie050478h
16. Jin, J., Niu, B., Ji, P., & Geng, Q. (2018). An integer linear programming model of reviewer assignment with research interest considerations. Annals of Operations Research, 1–25. https://doi.org/10.1007/s10479-018-2919-7
17. Karimzadehgan, M., & Zhai, C. (2012). Integer linear programming for Constrained Multi-Aspect Committee Review Assignment. Information Processing & Management, 48(4), 725–740. https://doi.org/10.1016/J.IPM.2011.09.004
18. Karimzadehgan, M., Zhai, C., & Belford, G. (2008). Multi-aspect expertise matching for review assignment. Proceeding of the 17th ACM Conference on Information and Knowledge Mining - CIKM ’08, 1113. https://doi.org/10.1145/1458082.1458230
19. Kuhn, H. (1955). The Hungarian method for the assignment problem. Naval Research Logistics Quarterly, 2(1–2), 83–97. https://doi.org/10.1002/nav.3800020109
20. Liu, O., Wang, J., Ma, J., & Sun, Y. (2016). An intelligent decision support approach for reviewer assignment in R&D project selection. Computers in Industry, 76, 1–10. https://doi.org/10.1016/J.COMPIND.2015.11.001
21. Liu, X., Suel, T., & Memon, N. (2014). A robust model for paper reviewer assignment. Proceedings of the 8th ACM Conference on Recommender Systems - RecSys ’14, 25–32. https://doi.org/10.1145/2645710.2645749
22. Ma, J., Xu, W., Sun, Y., Turban, E., Wang, S., & Liu, O. (2012). An Ontology-Based Text-Mining Method to Cluster Proposals for Research Project Selection. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, 42(3), 784–790. https://doi.org/10.1109/TSMCA.2011.2172205
23. Moawad, M., Maher, M., Awad, A., & Sakri S. (2019). MINARET: A Recommendation Framework for Scientific Reviewers. 22nd International Conference on Extending Database Technology (EDBT).
24. Mungen, A., Gundogan, E., Alhajj, R., & Kaya, M. (2018). A Novel Local Propagation Based Expert Finding Method. 2018 International Conference on Artificial Intelligence and Data Processing (IDAP), 1–7. https://doi.org/10.1109/IDAP.2018.8620860
25. Nguyen, J., Sánchez-Hernández, G., Agell, N., Rovira, X., & Angulo, C. (2018). A decision support tool using Order Weighted Averaging for conference review assignment. Pattern Recognition Letters, 105, 114–120. https://doi.org/10.1016/J.PATREC.2017.09.020
26. Pentico, D. W. (2007). Assignment problems: A golden anniversary survey. European Journal of Operational Research, 176(2), 774–793. https://doi.org/10.1016/J.EJOR.2005.09.014
27. Protasiewicz, J., Pedrycz, W., Kozłowski, M., Dadas, S., Stanisławek, T., Kopacz, A., & Gałężewska, M. (2016). A recommender system of reviewers and experts in reviewing problems. Knowledge-Based Systems, 106, 164–178. https://doi.org/10.1016/J.KNOSYS.2016.05.041
28. Selçuk Doğan, G. H. (2012). Expert Finding in Domains with Unclear Topics. Middle East Technical University.
29. Sun, Y., Ma, J., Fan, Z., & Wang, J. (2007). A Hybrid Knowledge and Model Approach for Reviewer Assignment. 2007 40th Annual Hawaii International Conference on System Sciences (HICSS’07). https://doi.org/10.1109/HICSS.2007.17
30. Tayal, D., Saxena, P., Sharma, A., Khanna, G., & Gupta, S. (2014). New method for solving reviewer assignment problem using type-2 fuzzy sets and fuzzy functions. Applied Intelligence, 40(1), 54–73. https://doi.org/10.1007/s10489-013-0445-5
31. Üçer, S. (2011). Bilimsel Değerlendirmeler için Performans Varisi Tabanlı Bir Değerlendirme Sistemi Geliştirilmesi. TÜBİTAK.
32. Wang, F., Chen, B., & Miao, Z. (2008). A Survey on Reviewer Assignment Problem. In New Frontiers in Applied Artificial Intelligence (pp. 718–727). https://doi.org/10.1007/978-3-540-69052-8_75
33. Xu, Y., Ma, J., Sun, Y., Hao, G., Xu, W., & Zhao, D. (2010). A decision support approach for assigning reviewers to proposals. Expert Systems with Applications, 37(10), 6948–6956. https://doi.org/10.1016/J.ESWA.2010.03.027
34. Yeşilçimen, A., & Yıldırım, E. (2019). An alternative polynomial-sized formulation and an optimization based heuristic for the reviewer assignment problem. European Journal of Operational Research, 276(2), 436–450. https://doi.org/10.1016/J.EJOR.2019.01.035
35. Yıldırım, E., Aykanat, C., Oruç, A., Atmaca, A., Kayaaslan, E., & Koca, E. (2012). Geniş Kapsamlı Proje Değerlendirme ve Seçim Sistemi.
36. Yunhong, X., & Xianli, Z. (2016). A LDA model based text-mining method to recommend reviewer for proposal of research project selection. 2016 13th International Conference on Service Systems and Service Management (ICSSSM), 1–5. https://doi.org/10.1109/ICSSSM.2016.7538568
37. Zhao, H., Tao, W., Zou, R., & Xu, C. (2018). Construction and Application of Diversified Knowledge Model for Paper Reviewers Recommendation. https://doi.org/10.1007/978-981-13-2206-8_11