人格特質預測一直是一項重要的研究議題。一個人的個性可以反應他的思想、感覺以及行為。近年來，E-learning system 因為提供了比傳統教育更彈性的學習方式，而吸引了許多使用者使用。越來越多使用者在E-learning system留下了大量的活動紀錄，而這些活動紀錄可以代表使用者的興趣、偏好和學習特色習慣。

因此，分析這些使用者的學習紀錄是非常有價值的。由於這些活動紀錄代表了使用者的學習過程，因此我們專注在推測使用者的嚴謹性特質。然而如何使活動紀錄對應到一個人的特性是非常具挑戰性的。為了解決此問題，我們主要從活動紀錄中找出使用者的學習模式。同時，為了能夠更仔細地比對使用者之間的個性差異，我們考慮了學習模式中動作的階層性。再者，我們比較使用者差異時，也考慮了使用者各自的學習能力。因此，在本論文中我們提出HAPE框架，同時考慮了使用者的學習模式和學習能力，且在比對差異性時保留了學習模式的階層性。

在實驗中，我們使用了現實的E-learning system的資料以顯示HAPE的效能，結果顯示我們的方法的確可以更準確地推測使用者的嚴謹性特質。同時我們也從多個面向來進行探討，包括我們的方法在不同參數設定下的表現，以及作用在不同特性的學生群上的表現。

Personality computing is an essential requirement in many applications. A person's characteristic patterns of thoughts, feelings and behaviors can be reflected in his personality. The E-learning system has been more popular in recent years because of the its flexibility. More and more users leave extensive activity logs which can represent users' interests, preferences and characteristics. Therefore, it is very valuable and promising to predict users' personalities by analyzing their digital footprints. Since the user-generated digital traces in E-learning systems can represent as users' leaning processes, we focus on inferring the personality trait Conscientiousness which is importantly related to successful academic performance and workplace performance. However, how to let activity logs correspond to a person's characteristics is extremely challenging. To deal with this problem, we aim at figuring out the users' learning styles (or working styles) from their activity logs. Meanwhile, to make a detailed comparison between users' learning styles, we consider the hierarchical relations between the actions users take in their learning styles. Furthermore, we also take a person's aptitude of all courses into account to enhance the correctness when mapping activity logs to a person's characteristics. Hence, in our work, a novel framework HAPE is proposed to aggregate a person's learning styles and aptitudes,
and preserve the hierarchy between learning styles at the same time. Empirically, our experimental studies on real data show that HAPE outperform all baselines, which means that our method is able to mapping the activity logs to human's characteristics successfully.

[1]  A. Vinciarelli and G. Mohammadi, “A survey of personality computing,” IEEE Transac- tions on A↵ective Computing, vol. 5, no. 3, pp. 273–291, July 2014. 

[2]  J. Golbeck, C. Robles, M. Edmondson, and K. Turner, “Predicting personality from twit- ter,” 2011 IEEE Third Int’l Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third Int’l Conference on Social Computing, pp. 149–156, 2011. 

[3]  M. Skowron, M. Tkalcic, B. Ferwerda, and M. Schedl, “Fusing social media cues: Person- ality prediction from twitter and instagram,” in WWW, 2016. 

[4]  C. J. Thoresen, J. C. Bradley, P. D. Bliese, and J. D. Thoresen, “The big five personality traits and individual job performance growth trajectories in maintenance and transitional job stages.” The Journal of applied psychology, vol. 89 5, pp. 835–53, 2004. 

[5]  S. Ghazi, G. Shahzada, and S. Ullah, “Relationship between students’ personality traits and their academic achievement in khyber pakhtunkhwa, pakistan,” Journal of Educational and Social Research, vol. 3, no. 2, 2013. [Online]. Available: https://www.mcser.org/journal/index.php/jesr/article/view/269 

[6]  E. C. Tupes and R. E. Christal, “Recurrent personality factors based on trait ratings.” Journal of personality, vol. 60 2, pp. 225–51, 1992. 

[7]  L. R. Goldberg, “The structure of phenotypic personality traits.” The American psychol- ogist, vol. 48 1, pp. 26–34, 1993. 

[8]  P. T. Costa and R. R. McCrae, “Domains and facets: hierarchical personality assessment using the revised neo personality inventory.” Journal of personality assessment, vol. 64 1, pp. 21–50, 1995. 

[9]  S. E. Argamon, S. Dhawle, M. Koppel, and J. W. Pennebaker, “Lexical predictors ofper- sonality type,” 2005. 

[10]  J. Oberlander and S. Nowson, “Whose thumb is it anyway? classifying author personality from weblog text,” in ACL, 2006.
[11]  A. J. Gill, S. Nowson, and J. Oberlander, “What are they blogging about? personality, topic and motivation in blogs,” in ICWSM, 2009. 

[12]  L. Flekova and I. Gurevych, “Personality profiling of fictional characters using sense-level links between lexical resources,” in EMNLP, 2015. 

[13]  T. Polzehl, S. Moller, and F. Metze, “Automatically assessing personality from speech,” in 2010 IEEE Fourth International Conference on Semantic Computing, Sep. 2010, pp. 134–140. 

[14]  O. Kampman, E. J. Barezi, D. Bertero, and P. Fung, “Investigating audio, visual, and text fusion methods for end-to-end automatic personality prediction,” in ACL, 2018. 

[15]  F. Pianesi, N. Mana, A. Cappelletti, B. Lepri, and M. Zancanaro, “Multimodal recognition of personality traits in social interactions,” in ICMI, 2008. 

[16]  N. Mana, B. Lepri, P. Chippendale, A. Cappelletti, F. Pianesi, P. Svaizer, and M. Zancanaro, “Multimodal corpus of multi-party meetings for automatic social behavior analysis and personality traits detection,” in Proceedings of the 2007 Workshop on Tagging, Mining and Retrieval of Human Related Activity Information, ser. TMR ’07. New York, NY, USA: ACM, 2007, pp. 9–14. [Online]. Available: http://doi.acm.org/10.1145/1330588.1330590 

[17]  A. Bera, T. Randhavane, and D. Manocha, “Aggressive, tense, or shy? identifying personality traits from crowd videos,” in Proceedings of the 26th International Joint Conference on Artificial Intelligence, ser. IJCAI’17. AAAI Press, 2017, pp. 112–118. [Online]. Available: http://dl.acm.org/citation.cfm?id=3171642.3171660 

[18]  S. Bai, T. Zhu, and L. Cheng, “Big-five personality prediction based on user behaviors at social network sites,” CoRR, vol. abs/1204.4809, 2012. 

[19]  D. O. Olgu ́ın, P. A. Gloor, and A. Pentland, “Capturing individual and group behavior with wearable sensors,” in AAAI Spring Symposium: Human Behavior Modeling, 2009. 

[20]  G. Chittaranjan, J. Blom, and D. Gatica-Perez, “Who’s who with big-five: Analyzing and classifying personality traits with smartphones,” 2011 15th Annual International Sympo- sium on Wearable Computers, pp. 29–36, 2011. 

[21]  A. Bordes, N. Usunier, A. Garc ́ıa-Dur ́an, J. Weston, and O. Yakhnenko, “Translating embeddings for modeling multi-relational data,” in NIPS, 2013. 

[22]  Z. Wang, J. Zhang, J. Feng, and Z. Chen, “Knowledge graph embedding by translating on hyperplanes,” in AAAI, 2014. 

[23]  Y. Lin, Z. Liu, M. Sun, Y. Liu, and X. Zhu, “Learning entity and relation embeddings for knowledge graph completion,” in AAAI, 2015. 

[24]  G. Ji, S. He, L. Xu, K. Liu, and J. Zhao, “Knowledge graph embedding via dynamic mapping matrix,” in ACL, 2015. 

[25]  K. Do, T. Tran, and S. Venkatesh, “Knowledge graph embedding with multiple relation projections,” 2018 24th International Conference on Pattern Recognition (ICPR), pp. 332– 337, 2018. 

[26]  A. Narayanan, M. Chandramohan, R. Venkatesan, L. Chen, Y. P. Liu, and S. Jaiswal, “graph2vec: Learning distributed representations of graphs,” CoRR, vol. abs/1707.05005, 2017. 

[27]  J. A. Johnson, “Measuring thirty facets of the five factor model with a 120-item public domain inventory: Development of the ipip-neo-120,” 2014.