學術研究發展日益興盛，傑出人才也隨之輩出，豐富的研究成果因應而生，然而如何在多樣的研究領域與擁有不同專業領域的學者之間，挖掘技術趨勢與人才關係，以求有效地找到值得投入的研究與合作夥伴，成為值得探討的一個議題。本研究藉由探勘不同來源的學術界開放資料，蒐集政府各部會補助之研究計畫、各學者指導之學位論文等資訊，經由識別學者身份與比對對應研究成果後，我們更可針對學者研究成果的標題、摘要與關鍵詞等進行文字前處理步驟，從而建立足跡引擎以追蹤各學者歷年於學術產出留下的研究技術名詞，並據以進行後續各項可能的分析與應用，諸如 ：運用資料視覺化技術以得知熱門技術與學者社群之發展、運用資料檢索介面讓使用者可以根據指定條件找到符合目標的學者等。而在學者合作預測此一分析課題中，我們採用異質資訊網路來表現學者與研究技術之間的整體關係，以預測學者潛在合作的可能性為目標，採用社群網路分析中的連結預測技術來推薦適合的合作對象，藉由兩位學者過往是否擁有同樣研究領域的學術成果，以路徑的概念詮釋此間接關係，並從中擷取出拓撲特徵，再以監督式學習演算法建立預測模型並進行各項實驗評估。整體而言，本研究的貢獻有兩點：其一是建立了一整套的資料處理流程，以供未來 多種可能應用之研發；其二則為深入探究學者合作預測此一研究課題，而能做為橋接合作契機的重要參考 。

With the development of science and technology, there are more and more outstanding scholars and brilliant research works. Nevertheless, it is challenging and worthwhile to explore the complex relationship among numerous scholars and their corresponding research works from various research fields. In this work, we collect information of scholars, research projects, theses and dissertations from several open data sources. Steps of identifying scholars and matching their corresponding research documents are firstly processed. Furthermore, steps of text preprocessing are then conducted on the titles, abstracts, and keywords of a research document to establish the footprint engine. Past footprints of a scholar are then carefully tracked in details. Possible applications can then be developed accordingly, including the visualization of hot topic trends and scholar communities, and the retrieval interface to help users find desired scholars. In view of the problem of collaboration prediction, we utilize heterogeneous information networks to present the overall relationship among scholars and their owned key terms. The goal is to estimate the possibility of future collaboration between two scholars. Techniques of link prediction is thus used in this work. Based on two scholars having similar footprints in the same research field, their indirect relationship is represented as a meta-path. Topological features of the meta-path are extracted to establish a prediction model using a supervised learning algorithm. Experimental studies are also conducted to evaluate the performance of our proposed approach. In summary, the contributions of this work is two-fold. Firstly, we have carefully devised a complete data flow to open the possibilities of future applications. Secondly, we have thoroughly explored the problem of collaboration prediction to bridge possible collaboration chances among scholars.

[1] Golder, Scott A., and Michael W. Macy. "Digital footprints: Opportunities and challenges for online social research." Annual Review of Sociology 40.1 (2014): 129-152.
[2] Azucar, Danny, Davide Marengo, and Michele Settanni. "Predicting the Big 5 personality traits from digital footprints on social media: A meta-analysis." Personality and individual differences 124 (2018): 150-159.
[3] Madani, Farshad, and C. Weber, "The evolution of patent mining: Applying bibliometrics analysis and keyword network analysis." World Patent Information, 46 (2016): 32-48
[4] Yoon, Byungun, and C. L. Magee, "Exploring technology opportunities by visualizing patent information based on generative topographic mapping and link prediction." Technological Forecasting and Social Change, 132 (2018): 105-117.
[5] Cheng, Qikai, et al., "Keyword-citation-keyword network: a new perspective of discipline knowledge structure analysis." Scientometrics, 124.3 (2020): 1923-1943.
[6] Li, Zijian, et al. "Transn: Heterogeneous network representation learning by translating node embeddings." 2020 IEEE 36th International Conference on Data Engineering (ICDE). IEEE, 2020.
[7] Zhang, J.; Li, T.; Jiang, Z.; Hu, X.; Jazayeri, A. A Noval Weighted Meta Graph Method for Classification in Heterogeneous Information Networks. Appl. Sci. 2020, 10, 1603. https://doi.org/10.3390/app10051603
[8] Shahmohammadi, Amin, Ehsan Khadangi, and Alireza Bagheri. "Presenting new collaborative link prediction methods for activity recommendation in Facebook." Neurocomputing 210 (2016): 217-226.
[9] Radhakrishnan, Srinivasan, et al. "Novel keyword co-occurrence network-based methods to foster systematic reviews of scientific literature." PloS one 12.3 (2017): e0172778.
[10] Tabassum, Shazia, et al. "Social network analysis: An overview." Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 8.5 (2018): e1256.
[11] Sun, Huaping, et al. "Measuring China's new energy vehicle patents: A social network analysis approach." Energy 153 (2018): 685-693.
[12] Liu, Linqing, and Shiye Mei. "Visualizing the GVC research: a co-occurrence network based bibliometric analysis." Scientometrics 109.2 (2016): 953-977.
[13] González, Luis-Millán, et al. "An author keyword analysis for mapping Sport Sciences." PLoS One 13.8 (2018): e0201435.
[14] Min, Kyunghun, Moonyoung Yoon, and Katsunori Furuya. "A Comparison of a smart city’s trends in urban planning before and after 2016 through keyword network analysis." Sustainability 11.11 (2019): 3155.
[15] Bütün, Ertan, and Mehmet Kaya. "Predicting citation count of scientists as a link prediction problem." IEEE transactions on cybernetics 50.10 (2019): 4518-4529.
[16] Shahmohammadi, Amin, Ehsan Khadangi, and Alireza Bagheri. "Presenting new collaborative link prediction methods for activity recommendation in Facebook." Neurocomputing 210 (2016): 217-226.
[17] Li, Shugang, et al. "Friend recommendation for cross marketing in online brand community based on intelligent attention allocation link prediction algorithm." Expert Systems with Applications 139 (2020): 112839.
[18] Wang, Xiao, et al. "Dynamic heterogeneous information network embedding with meta-path based proximity." IEEE Transactions on Knowledge and Data Engineering (2020).
[19] Sun, Yizhou, et al. "Pathsim: Meta path-based top-k similarity search in heterogeneous information networks." Proceedings of the VLDB Endowment 4.11 (2011): 992-1003.
[20] Jalili, Mahdi, et al. "Link prediction in multiplex online social networks." Royal Society open science 4.2 (2017): 160863.
[21] Li, Ji-chao, et al. "A link prediction method for heterogeneous networks based on BP neural network." Physica A: Statistical Mechanics and its Applications 495 (2018): 1-17.
[22] Martínez, Víctor, Fernando Berzal, and Juan-Carlos Cubero. "A survey of link prediction in complex networks." ACM computing surveys (CSUR) 49.4 (2016): 1-33.
[23] Wang, Peng, et al. "Link prediction in social networks: the state-of-the-art." Science China Information Sciences 58.1 (2015): 1-38.
[24] Zhang, Muhan, and Yixin Chen. "Link prediction based on graph neural networks." Advances in neural information processing systems 31 (2018).
[25] Academic Research Service Portal Researcher Query from MOST, https://arspb.most.gov.tw/NSCWebFront/modules/talentSearch/talentSearch.do?action=initSearchList&LANG=ch
[26] Daud, Nur Nasuha, et al. "Applications of link prediction in social networks: A review." Journal of Network and Computer Applications 166 (2020): 102716.
[27] Y. Sun, R. Barber, M. Gupta, C. C. Aggarwal, and J. Han, “Co-Author Relationship Prediction in Heterogeneous Bibliographic Networks,” Proceeding of International Conference on Advances in Social Networks Analysis and Mining, pages 121-128, July 2011