Heterojen Bilgi Ağlarında Zamana Duyarlı Bilimsel Takım Önerileri Üretimi

Öz

Geniş kapsamlı disiplinlerarası bilimsel araştırma projelerinin artması, farklı uzmanlıklara sahip araştırmacıların proje takımlarında bir araya getirilmesini gerektirmektedir. Literatürde sosyal ağlara veya heterojen bilgi ağlarına dayalı olarak takım üretmeyi hedefleyen çalışmalar olmakla birlikte, mevcut çalışmalar çoğunlukla ortak yazarlık veya kişi-uzmanlık/yayın yeri ilişkilerine dayanmakta, kurumdaşlık ilişkilerini göz ardı etmektedir. Bu çalışmada, YÖK Akademik Platformu'ndan alınan bilgiler hem ortak yazarlık hem de kurumdaşlık ilişkilerini kapsayan bir heterojen bilgi ağına dönüştürülmüş ve bu ağ üzerinde ortak yazarlık, kurumdaşlık ve uzak aramayı kullanan üç aşamalı bir stratejiyle, çalışmaları güncel, uzmanlık derecesi yüksek, iletişim maliyeti düşük takımlar oluşturan zamana duyarlı bir algoritma (ZDA) önerilmiştir. Kurumdaşlık ilişkilerinde aynı kurumda geçirilen zamanın çakışmasını dikkate alan versiyonun (ZDA-3 Yıl) yanı sıra, bu çakışmayı kontrol etmeyen (ZDA-Zamansız) ve yalnızca ortak yazarlık ilişkilerine dayanan (OYDA) versiyonları da geliştirilmiştir. 200 farklı anahtar kelime listesiyle yapılan testlerde ZDA algoritmaları %94,5 tamamlanma oranına ulaşırken, daha geniş arama mesafesine sahip OYDA algoritması %88'de kalmıştır. Sonuçlar, heterojen bilgi ağlarındaki çok katmanlı ilişki yapısının takım oluşturmada etkili olduğunu, kurumdaşlık ilişkilerinin hem tamamlanma oranını artırdığını hem de daha düşük maliyetli takımlar oluşturmaya imkân verdiğini ortaya koymaktadır.

Anahtar Kelimeler

• takım oluşturma
• heterojen bilgi ağları
• sosyal ağ analizi
• ağ veritabanı
• kurumsal bağlantı
• neo4j

Time Sensitive Scientific Team Recommendation on Heterogeneous Information Network

Öz

The increasing number of large-scale interdisciplinary scientific research projects requires researchers with diverse expertise to come together in project teams. While there are studies in the literature that aim to suggest teams based on social networks or Heterogeneous Information Networks (HIN), existing studies mostly rely on co-authorship or author-skill/venue relationships and disregard institutional affiliations. In this study, the data obtained from the YÖK Academic Platform is transformed into a HIN encompassing both co-authorship and institutional affiliations. A time-sensitive algorithm (TSA) is proposed on this network, using a three-stage strategy employing co-authorship, institutional affiliation, and remote search to create teams with up-to-date work, high levels of expertise, and low communication costs. In addition to a version that considers the overlap of time spent at the same institution in institutional affiliations (TSA-3 Years), versions that do not control for this overlap (TSA-Timeless) and that rely solely on co-authorship relationships (COAT) have also been developed. In tests conducted with 200 different keyword lists, TSA algorithms achieved a 94.5% completion rate, while COAT, despite having a wider search distance, remained at 88%. The results reveal that the multi-layered relational structure in HINs is effective in team formation, and that institutional affiliation relationships both increase the completion rate and enable the creation of lower-cost teams.

Anahtar Kelimeler

• team formation
• heterogeneous information network
• social network analysis
• graph database
• institutional affiliation
• neo4j

Kaynakça

1. [1] Stavrou, G.,Adamidis, P.A., Papathanasiou, J. B.,&Tarabanis, K.A. (2023). Team formation: A systematic literature review[Article]. International Journal of Business Science and Applied Management, 18(2), 17–34. doi: 10.69864/ijbsam.18-2.174
2. [2] Zakarian, A., & Kusiak, A. (1999). Forming teams: An analytical approach. IIE Transactions, 31(1), 85–97. doi: 10.1080/07408179908969808
3. [3] Baykasoglu, A., Dereli, T., & Das, S. (2007). Project Team Selection Using Fuzzy Optimization Approach. Cybernetics and Systems, 38(2), 155–185. doi: 10.1080/01969720601139041
4. [4] Gaston, M.E., Simmons, R., desJardins, M., 2004. Adapting Network Structure for Efficient Team Formation. Technical Report. AAAI Technical Report WS-04-02. Menlo Park, CA. In AAAI Technical Report (WS-04-02), pp. 1-8.
5. [5] Fitzpatrick, E.L., Askin, R.G., 2005. Forming effective worker teams with multi-functional skill requirements. Computers & Industrial Engineering 48, 593–608. doi:10.1016/j.cie.2004.12.014.
6. [6] Backstrom, L., Huttenlocher, D., Kleinberg, J., Lan, X., 2006. Group formation in large social networks: Membership, growth, and evolution, in: Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, Philadelphia PA USA. pp. 44–54. doi:10.1145/1150402.1150412.
7. [7] Cheatham, M., Cleereman, K., 2006. Application of Social Network Analysis to Collaborative Team Formation, in: International Symposium on Collaborative Technologies and Systems (CTS’06), IEEE, Las Vegas, NV, USA. pp. 306–311. doi:10.1109/CTS.2006.18.
8. [8] Wi, H., Oh, S., Mun, J., Jung, M., 2009. A team formation model based on knowledge and collaboration. Expert Systems with Applications 36, 9121–9134. doi:10.1016/j.eswa.2008.12.031.

9. [9] Lappas, T., Liu, K., Terzi, E., 2009. Finding a team of experts in social networks, in: Proceedings of the 15thACMSIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, Paris France. pp. 467–476. doi:10.1145/1557019.1557074.
10. [10] Li, C.T., Shan, M.K., 2010. Team Formation for Generalized Tasks in Expertise Social Networks, in: 2010 IEEE Second International Conference on Social Computing, IEEE, Minneapolis, MN, USA. pp. 9–16. doi:10.1109/SocialCom.2010.12.
11. [11] Dorn, C., Dustdar, S., 2010. Composing Near-Optimal Expert Teams: A Trade-Off between Skills and Connectivity, in: Meersman, R., Dillon, T., Herrero, P. (Eds.), On the Move to Meaningful Internet Systems: OTM 2010. Springer Berlin Heidelberg, Berlin, Heidelberg. volume 6426, pp. 472–489. doi:10.1007/978-3-642-16934-2_35.
12. [12] Kargar, M., An, A., 2011. Discovering top-k teams of experts with/without a leader in social networks, in: Proceedings of the 20th ACMInternational Conference on Information and Knowledge Management, ACM, Glasgow Scotland, UK. pp. 985–994. doi:10.1145/2063576.2063718.
13. [13] Anagnostopoulos, A., Becchetti, L., Castillo, C., Gionis, A., & Leonardi, S. (2010). Power in unity: Forming teams in large-scale community systems. In Proceedings of the 19th ACM international conference on Information and knowledge management (pp. 599–608). Toronto ON Canada: ACM. doi: 10.1145/1871437.1871515
14. [14] Anagnostopoulos, A., Becchetti, L., Castillo, C., Gionis, A., & Leonardi, S. (2012). Online team formation in social networks. In Proceedings of the 21st international conference on World Wide Web (pp. 839–848). Lyon France: ACM. doi: 10.1145/2187836.2187950
15. [15] Majumder, A., Datta, S., Naidu, K., 2012. Capacitated team formation problem on social networks, in: Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, Beijing China. pp. 1005–1013. doi:10.1145/2339530.2339690.
16. [16] Rangapuram, S.S., Bühler, T., Hein, M., 2013. Towards realistic team formation in social networks based on densest subgraphs. In: Proceedings of the 22nd International Conference on World Wide Web. ACM, Rio de Janeiro, Brazil, pp. 1077–1088. doi:10.1145/2488388.2488482.
17. [17] Awal, G.K., Bharadwaj, K.K., 2014. Team formation in social networks based on collective intelligence – an evolutionary approach. Applied Intelligence, 41, 627–648. doi:10.1007/s10489-014-0528-y.
18. [18] Mahajan, Y., Cho, J.H., Chen, I., 2024. Privacy-preserving and diversity-aware trust-based team formation in online social networks. ACM Transactions on Intelligent Systems and Technology, 15. doi:10.1145/3670411.
19. [19] Selvarajah, K., Zadeh, P.M., Kobti, Z., Palanichamy, Y., Kargar, M., 2021. A unified framework for effective team formation in social networks. Expert Systems with Applications, 177, 114886. doi:10.1016/j.eswa.2021.114886.
20. [20] Yin, X., Qu, C., Wang, Q., Wu, F., Liu, B., Chen, F., Chen, X., Fang, D., 2018. Social connection aware team formation for participatory tasks. IEEE Access, 6, 20309–20319. doi:10.1109/ACCESS.2018.2819992.
21. [21] Gutiérrez, J.H., Astudillo, C.A., Ballesteros-Pérez, P., Mora-Melià, D., Candia-Véjar, A., 2016. The multiple team formation problem using sociometry. Computers & Operations Research, 75, 150–162. doi:10.1016/j.cor.2016.05.012.
22. [22] Daş, G.S., Altınkaynak, B., Göçken, T., Türker, A.K., 2022. A set partitioning based goal programming model for the team formation problem. International Transactions in Operational Research, 29, 301–322. doi:10.1111/itor.13022.
23. [23] Campêlo, M.B., Figueiredo, T.F., 2021. Integer programming approaches to the multiple team formation problem. Computers & Operations Research, 133, 105354. doi:10.1016/j.cor.2021.105354.
24. [24] Fathian, M., Saei-Shahi, M., Makui, A., 2017. A new optimization model for reliable team formation problem considering experts’ collaboration network. IEEE Transactions on Engineering Management, 64, 586–593. doi:10.1109/TEM.2017.2715825.
25. [25] Cao, N., Lin, Y.R., Li, L., Tong, H., 2015. G-Miner: Interactive visual group mining on multivariate graphs. In: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems. ACM, Seoul, South Korea, pp. 279–288. doi:10.1145/2702123.2702446.
26. [26] Hu, C., Zhou, Q., Tong, H., 2024. Genius: Subteam replacement with clustering-based graph neural networks. In: Proceedings of the 2024 SIAM International Conference on Data Mining (SDM), pp. 10–18. doi:10.1137/1.9781611978032.2.
27. [27] Li, L., Tong, H., Cao, N., Ehrlich, K., Lin, Y.R., Buchler, N., 2016. TEAMOPT: Interactive team optimization in big networks. In: Proceedings of the 25th ACM International Conference on Information and Knowledge Management. ACM, Indianapolis, USA, pp. 2485–2487. doi:10.1145/2983323.2983340. [28] Zhou, Q., Li, L., Tong, H., 2019. Towards real-time team optimization. In: 2019 IEEE International Conference on Big Data. IEEE, Los Angeles, USA, pp. 1008–1017. doi:10.1109/BigData47090.2019.9006078.
28. [29] El-Ashmawi, W.H., Ali, A.F., Tawhid, M.A., 2019. An improved particle swarm optimization with a new swap operator for team formation problem. Journal of Industrial Engineering International, 15, 53–71. doi:10.1007/s40092-018-0282-6.
29. [30] El-Ashmawi, W.H., Ali, A.F., Słowik, A., 2020. An improved Jaya algorithm with a modified swap operator for solving team formation problem. Soft Computing, 24, 16627–16641. doi:10.1007/s00500-020-04965-x.
30. [31] El-Ashmawi, W.H., Sheta, A.F., Alqadi, B.S., AbdElminaam, D.S., Mohammed Alsekait, D.M., 2025. A new version of the golden eagle optimizer algorithm and its application for solving a trio-objective skillful team formation problem in a social network. Journal of Artificial Intelligence and Soft Computing Research, 15, 357–384. doi:10.2478/jaiscr-2025-0018.
31. [32] Rehman, M.Z., Zamli, K.Z., Almutairi, M.S., Chiroma, H., Aamir, M., Kader, M.A., Mohd Nawi, N., 2021. A novel state space reduction algorithm for team formation in social networks. PLOS ONE, 16. doi:10.1371/journal.pone.0259786.
32. [33] Sun, Y., Han, J., 2013. Mining heterogeneous information networks: A structural analysis approach. ACM SIGKDD Explorations Newsletter, 14, 20–28. doi:10.1145/2481244.2481248.
33. [34] Cai, D., Shao, Z., He, X., Yan, X., Han, J., 2005. Community mining from multi-relational networks. In: European Conference on Principles of Data Mining and Knowledge Discovery. Springer, pp. 445–452.
34. [35] Shafie, T., 2015. A multigraph approach to social network analysis. Journal of Social Structure, 16, 1–21. doi:10.21307/joss-2019-011.
35. [36] Yang, J., Li, M., Wu, B., Xu, C., 2016. Forming a research team of experts in expert-skill co-occurrence network of research news. In: 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), pp. 1143–1150. doi:10.1109/ASONAM.2016.7752383.
36. [37] Yang, X., Wang, P., Fang, W., 2018. Team formation with relationship strength based on meta-path in heterogeneous network. In: 2018 IEEE 9th International Conference on Software Engineering and Service Science. IEEE, pp. 797–801. doi:10.1109/ICSESS.2018.8663775.
37. [38] Hamidi Rad, R., Fani, H., Kargar, M., Szlichta, J., & Bagheri, E. (2020, October). Learning to form skill-based teams of experts. In Proceedings of the 29th ACM international conference on information & knowledge management (pp. 2049-2052). doi:10.1145/3340531.3412140.
38. [39] Hamidi Rad, R., Bagheri, E., Kargar, M., Srivastava, D., Szlichta, J., 2021. Retrieving skill-based teams from collaboration networks. In: Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 2015–2019. doi:10.1145/3404835.3463105.
39. [40] Hamidi Rad, R., Nguyen, H., Al-Obeidat, F., Bagheri, E., Kargar, M., Srivastava, D., Szlichta, J., Zarrinkalam, F., 2023. Learning heterogeneous subgraph representations for team discovery. Information Retrieval Journal, 26, 8. doi:10.1007/s10791-023-09421-6.
40. [41] Lee, O.J., Hong, S., Kim, J.T., 2021. Interinstitutional research team formation based on bibliographic network embedding. Mobile Information Systems, 2021. doi:10.1155/2021/6629520.
41. [42] Sharaf, M.A., Alghamdi, T.G., 2025. Balanced team formation using hybrid graph convolution networks and MILP. Applied Sciences, 15, 2049. doi:10.3390/app15042049.
42. [43] Dara, M., Rad, R.H., Zarrinkalam, F., Bagheri, E., 2025. Retrieval-augmented neural team formation. In: Advances in Information Retrieval. Springer Nature Switzerland, pp. 362–371. doi:10.1007/978-3-031-88714-7_35.
43. [44] Liu, X., Wu, K., Liu, B., Qian, R., 2023. HNERec: Scientific collaborator recommendation model based on heterogeneous network embedding. Information Processing & Management, 60, 103253. doi:10.1016/j.ipm.2022.103253.
44. [45] Gao, X., Wu, S., Xia, D., Xiong, H., 2023. Topic-sensitive expert finding based solely on heterogeneous academic networks. Expert Systems with Applications, 213, 119241. doi:10.1016/j.eswa.2022.119241.
45. [46] Guan, K., Huang, W., Chen, T., Chan, W.K.V., 2025. TEDRec: Transformer-based scientific collaborator recommendation via textual-edge dynamic network modeling. Information Processing & Management, 62, 104283. doi:10.1016/j.ipm.2025.104283.
46. [47] Wang, X., Huang, W., Zhao, B., Li, S., 2026. Scientific collaborator recommendation via hypergraph embedding. Information Processing & Management, 63, 104423. doi:10.1016/j.ipm.2025.104423.
47. [48] Liu, J., Zhang, Q., Xu, S., Huang, C., Gong, X., 2026. Innovation performance improvement path in scientific research teams: A perspective of network embeddedness. Information Processing & Management, 63, 104484. doi:10.1016/j.ipm.2025.104484.
48. [49] Priem, J., Piwowar, H., & Orr, R. (2022). OpenAlex: A fully-open index of scholarly works, authors, venues, institutions, and concepts. arXiv preprint arXiv:2205.01833.
49. [50] Ley, M. (2002). The DBLP Computer Science Bibliography: Evolution, Research Issues, Perspectives. In: Laender, A.H.F., Oliveira, A.L. (eds) String Processing and Information Retrieval. SPIRE 2002. Lecture Notes in Computer Science, vol 2476. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45735-6_1
50. [51] Anadolu Ajansı, “YÖK Akademik arama motoru kullanıma açılıyor,” 3 Şubat 2016. [Çevrimiçi]. https://www.aa.com.tr/tr/egitim/yokakademik-arama-motoru-kullanima-aciliyor/530885