Revisit Girvan-Newman Algorithm for Research Topic Analysis: An Application on Library and Information Science Studies

Research trend analysis gives the research community an essential chance to learn the past to support sustainable development. The topic of evolution analysis presents a chance to position the current research, linkages among research topics, and identify the research gap. In this study, the authors revisit a known mechanism, namely Girvan-Newman (GN) algorithm, and propose a new approach for research topic analysis. Based on the GN algorithm, author-keywords analysis approach, one-mode cluster, and duo GN algorithm analysis were suggested and applied to research topic analysis of Library and Information Science studies. The results show that the suggested approach could process major quantity materials and be able to avoid the possible distorted results gained by taking the small size of samples, or two-mode cluster, to ensure the validity of the results. The topics' hierarchy structure also suggests a different approach that could be used to deconstruct the linkages among research topics for future study.

關鍵字

Research Topic Evolution ； Girvan-Newman Algorithm ； Library Information Science

並列摘要

研究趨勢的分析為學術界提供一個可以了解過去並藉以支持未來持續發展的重要機會。主題演化分析能用來定位當前研究、連結研究主題間的關係，以及辨識研究主題間的落差。在本研究中，作者重新審視現有的Girvan-Newman（GN）演算法在主題演化分析的應用，提出了一個新的主題演化分析的方法。在作者－關鍵詞關係、單模叢集分析和雙重GN演算法的基礎上，作者進行圖書資訊學的研究主題分析。研究結果顯示，作者提出的方法可以處理大量資料文獻，並且能夠避免因為小樣本或雙模叢集分析導致的偏誤結果，進而確保研究結果的有效性。最後作者更提出建構研究主題的階層來衡量研究主題之間的連結關係，可作為後續深入研究的方法。

並列關鍵字

研究主題演化； Girvan-Newman演算法；圖書資訊學

參考文獻

Alghamdi, E., & Greene, D. (2019). Active semi-supervised overlapping community finding with pairwise constraints. Applied Network Science, 4, Article 63. https://doi.org/10.1007/s41109-019-0175-7

Burt, R. S. (2004). Structural holes and good ideas. American Journal of Sociology, 110(2), 349-399. https://doi.org/10.1086/421787

Chang, Y.-W., Huang, M.-H., & Lin, C.-W. (2015). Evolution of research subjects in library and information science based on keyword, bibliographical coupling, and co-citation analyses. Scientometrics, 105(3), 2071-2087. https://doi.org/10.1007/s11192-015-1762-8

Despalatović, L., Vojković, T., & Vukic̆ević, D. (2014). Community structure in networks: Girvan-Newman algorithm improvement. In P. Biljanovic, Z. Butkovic, K. Skala, S. Golubic, M. Cicin-Sain, V. Sruk, S. Ribaric, S. Gros, B. Vrdoljak, M. Mauher, & G. Cetusic (Eds.), 2014 37th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO) (pp. 997-1002). IEEE. https://doi.org/10.1109/MIPRO.2014.6859714

Figuerola, C. G., García Marco, F. J., & Pinto, M. (2017). Mapping the evolution of library and information science (1978–2014) using topic modeling on LISA. Scientometrics, 112(3), 1507-1535. https://doi.org/10.1007/s11192-017-2432-9

主題瀏覽