A Novel Cascade Model for Learning Latent Similarity from Heterogeneous Sequential Data of MOOC

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2768–2773 Copenhagen, Denmark, September 7–11, 2017. c⃝2017 Association for Computational Linguistics A Novel Cascade Model for Learning Latent Similarity from Heterogeneous Sequential Data of MOOC Zhuoxuan Jiang1 and Shanshan Feng2 and Gao Cong2 and Chunyan Miao3 and Xiaoming Li1 1School of Electronics Engineering and Computer Science, Peking University, China 2School of Computer Science and Engineering, Nanyang Technological University, Singapore 3Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly (LILY), Nanyang Technological University, Singapore {jzhx,lxm}@pku.edu.cn, {sfeng003@e.,gaocong@,ascymiao@}ntu.edu.sg Abstract Recent years have witnessed the prolif- eration of Massive Open Online Courses (MOOCs). With massive learners being offered MOOCs, there is a demand that the forum contents within MOOCs need to be classified in order to facilitate both learners and instructors. Therefore we in- vestigate a significant application, which is to associate forum threads to subti- tles of video clips. This task can be re- garded as a document ranking problem, and the key is how to learn a distinguish- able text representation from word se- quences and learners’ behavior sequences. In this paper, we propose a novel cascade model, which can capture both the latent semantics and latent similarity by mod- eling MOOC data. Experimental results on two real-world datasets demonstrate that our textual representation outperforms state-of-the-art unsupervised counterparts for the application. 1 Introduction With the rapid development of Massive Open On- line Courses (MOOCs), more and more learners participate in MOOCs (Anderson et al., 2014). Due to the lack of effective management, most of the discussion forums within MOOCs are over- loaded and in chaos (Huang et al., 2014). There- fore, a key problem is how to manage the forum contents. To manage the forum contents, threads of fo- rums can be regarded as documents and be classi- fied to groups. There are several straightforward methods, such as defining sub-forums accord- ing to weeks and asking learners to tag threads. However their effectiveness is limited (Rossi and Gnawali, 2014), because learners have few incen- tives to tag threads. Recently, machine learning solutions have been proposed, e.g., content-related thread identification (Wise et al., 2016), confusion classification (Agrawal et al., 2015) and sentiment classification (Ramesh et al., 2015). However they are developed for specific research problems and cannot be applied to our problem. Moreover, they require labeling data which needs domain experts to label data for different courses. We observe that the video clips of a MOOC would have many well-formed subtitles composed by instructors. Moreover, within MOOC settings, the course contents can be broken down to knowl- edge points, and each video clip just corresponds to a knowledge point. Consequently, we pro- pose to fulfill the application, which is to asso- ciate threads to subtitles of video clips, i.e., thread- subtitle matching. By this way, the relevant videos to the threads can be recommended to learners, and the chaotic threads in discussion forums can also be well grouped. However, it is challenging to identify the rele- vant …

Content truncated to 500 words