摘要
移动互联网技术的飞速发展和智能终端设备的广泛普及为支持泛在学习(U-Learning)提供了可能.泛在学习是一种随时随地都能发生的在线学习.当前,互联网涌现了一大批可供学习的平台和资源,类似平台包括网易云课堂、Coursera、知乎网、简书网等.手动检索和个性化推荐是在线学习系统提供给用户获取学习资源的常用方式.个性化推荐因其能主动建模学习者偏好,为学习者提供个性化的学习资源推荐服务而受到了广泛关注并成为在线学习领域的研究热点.本文在传统协同过滤个性化推荐系统的基础上提出一种基于迁移学习的学习资源的推荐方法,该方法从已有的数据中学习知识,然后迁移到目标任务中,解决了目标任务中数据过少从而导致学习特征的能力不足的问题.
With the rapid development of mobile Internet technology and the wide popularity of intelligent terminal devices,it is possible to support ubiquitous learning(U-Learning). Ubiquitous learning is a kind of online learning that can take place anytime,anywhere. At present,the Internet has emerged a large number of platforms and resources for learning,such as NetEase Cloud classroom,Coursera,Zhihu.com,Jianshu.com and so on. Manual retrieval and personalized recommendation are commonly used in online learning systems to provide users with access to learning resources.Personalized recommendation,which can actively model learners' preferences and provide personalized learning resources recommendation service for learners,has been widely concerned and become a research hotspot in the field of online learning. Based on the traditional collaborative filtering personalized recommendation system,this paper proposes a recommendation method of learning resources based on transfer learning. The method learns knowledge from existing data,and then transfers to the target task. It solves the problem that there is too little data in the target task,which leads to the lack of the ability to learn the features.
引文
[1]Button D,Harrington A,Belan I.E-learning&information communication technology(ICT)in nursing education:Areview of the literature[J].Nurse Education Today,2014,34(10):1311-1323.
[2]Tarhini A,Hone K,Liu X.The effects of individual differences on e-learning users’behaviour in developing countries:A structural equation model[J].Computers in Human Behavior,2014,41:153-163.
[3]Burke R.Hybrid recommender systems:Survey and experiments[J].User Modeling and User-adapted Interaction,2002,12(4):331-370.
[4]Pan W,Yang Q.Transfer learning in heterogeneous collaborative filtering domains[J].Artificial Intelligence,2013,197:39-55.
[5]Koren Y.Factorization meets the neighborhood:a multifaceted collaborative filtering model[C]//Proceedings of the14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Las Vegas,Nevada,USA:ACM,2008:426-434.
[6]Singh A P,Gordon G J.Relational learning via collective matrix factorization[C]//Proceedings of the 14 th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining.Las Vegas,Nevada,USA:ACM,2008:650-658.
[7]Mnih A,Salakhutdinov R R.Probabilistic matrix factorization[C]//Advances in Neural Information Processing Systems.Vancouver,Canada:NIPS,2008:1257-1264.
[8]Ziegler C N,McNee S M,Konstan J A,et al.Improving recommendation lists through topic diversification[C]//Proceedings of the 14th International Conference on World Wide Web.Chiba,Japan:ACM,2005:22-32.
[9]Zesch T,Müller C,Gurevych I.Extracting lexical semantic knowledge from wikipedia and wiktionary[C]//Proceedings of the 6th International Conference on Language Resources and Evaluation.Marrakech,Morocco:LREC Press,2008:1646-1652.
[10]Sarwar B,Karypis G,Konstan J,et al.Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the 10 th International Conference on World Wide Web.Hong Kong,China:ACM,2001:285-295.
[11]Pan W,Xiang E W,Yang Q.Transfer learning in collaborative filtering with uncertain ratings[C]//Proceedings of the26th AAAI Conference on Artificial Intelligence.Toronto,Ontario,Canada:AAAI,2012,12:662-668.
[12]HerlockerJ L,Konstan J A,Riedl J.Explaining collaborative filtering recommendations[C]//Proceedings of the 2000ACM Conference on Computer Supported Cooperative Work.Philadelphia,Pennsylvania,USA:ACM,2000:241-250.
