A new approach for combining content-based and collaborative filters

With the development of e-commerce and the proliferation of easily accessible information, recommender systems have become a popular technique to prune large information spaces so that users are directed toward those items that best meet their needs and preferences. A variety of techniques have been proposed for performing recommendations, including content-based and collaborative techniques. Content-based filtering selects information based on semantic content, whereas collaborative filtering combines the opinions of other users to make a prediction for a target user. In this paper, we describe a new filtering approach that combines the content-based filter and collaborative filter to capitalize on their respective strengths, and thereby achieves a good performance. We present a series of recommendations on the selection of the appropriate factors and also look into different techniques for calculating user-user similarities based on the integrated information extracted from user profiles and user ratings. Finally, we experimentally evaluate our approach and compare it with classic filters, the result of which demonstrate the effectiveness of our approach.     

Logarithmic regret algorithms for online convex optimization

In an online convex optimization problem a decision-maker makes a sequence of decisions, i.e., chooses a sequence of points in Euclidean space, from a fixed feasible set. After each point is chosen, it encounters a sequence of (possibly unrelated) convex cost functions. Zinkevich (ICML 2003) introduced this framework, which models many natural repeated decision-making problems and generalizes many existing problems such as Prediction from Expert Advice and Cover’s Universal Portfolios. Zinkevich showed that a simple online gradient descent algorithm achieves additive regret , for an arbitrary sequence of T convex cost functions (of bounded gradients), with respect to the best single decision in hindsight. In this paper, we give algorithms that achieve regret O(log (T)) for an arbitrary sequence of strictly convex functions (with bounded first and second derivatives). This mirrors what has been done for the special cases of prediction from expert advice by Kivinen and Warmuth (EuroCOLT 1999), and Universal Portfolios by Cover (Math. Finance 1:1–19, 1991). We propose several algorithms achieving logarithmic regret, which besides being more general are also much more efficient to implement. The main new ideas give rise to an efficient algorithm based on the Newton method for optimization, a new tool in the field. Our analysis shows a surprising connection between the natural follow-the-leader approach and the Newton method. We also analyze other algorithms, which tie together several different previous approaches including follow-the-leader, exponential weighting, Cover’s algorithm and gradient descent. Editors: Hans Ulrich Simon, Gabor Lugosi, Avrim Blum. E. Hazan and S. Kale supported by Sanjeev Arora’s NSF grants MSPA-MCS 0528414, CCF 0514993, ITR 0205594.     

Applying artificial immune systems to collaborative filtering for movie recommendation

Collaborative filtering is a widely used recommendation technique and many collaborative filtering techniques have been developed, each with its own merits and drawbacks. In this study, we apply an artificial immune network to collaborative filtering for movie recommendation. We propose new formulas in calculating the affinity between an antigen and an antibody and the affinity of an antigen to an immune network. In addition, a modified similarity estimation formula based on the Pearson correlation coefficient is also developed. A series of experiments based on MovieLens and EachMovie datasets are conducted, and the results are very encouraging.     

Multi-domain collaborative exploration mechanisms for query expansion in an agent-based filtering framework

Novice users often do not have enough domain knowledge to create good queries for searching information on-line. To help alleviate the situation, exploration techniques have been used to increase the diversity of the search results so that not only those explicitly asked will be returned, but also those potentially relevant ones will be returned too. Most existing approaches, such as collaborative filtering, do not allow the level of exploration to be controlled. Consequently, the search results can be very different from what is expected. We propose an exploration strategy that performs intelligent query processing by first searching usable old queries, and then utilising them to adapt the current query, with the hope that the adapted query will be more relevant to the user’s areas of interest. We applied the proposed strategy to the implementation of a personal information assistant (PIA) set up for user evaluation for 3 months. The experimental results showed that the proposed exploration method outperformed collaborative filtering, and mutation and crossover methods by around 25% in terms of the elimination of off-topic results.     

Using collaborative filtering to support college students’ use of online forum for English learning

This study examined the impact of collaborative filtering (the so-called recommender) on college students’ use of an online forum for English learning. The forum was created with an open-source software, Drupal, and its extended recommender module. This study was guided by three main questions: 1) Is there any difference in online behaviors between students who use a traditional forum and students who use a forum with a recommender?; 2) Is there any difference in learning motivation between students who use a traditional forum and students who use a forum with a recommender?; 3) Is there any difference in learning achievement between students who use a traditional forum and students who use a forum with a recommender?.     

Creating cognitive tutors for collaborative learning: steps toward realization

Our long-term research goal is to provide cognitive tutoring of collaboration within a collaborative software environment. This is a challenging goal, as intelligent tutors have traditionally focused on cognitive skills, rather than on the skills necessary to collaborate successfully. In this paper, we describe progress we have made toward this goal. Our first step was to devise a process known as bootstrapping novice data (BND), in which student problem-solving actions are collected and used to begin the development of a tutor. Next, we implemented BND by integrating a collaborative software tool, Cool Modes, with software designed to develop cognitive tutors (i.e., the cognitive tutor authoring tools). Our initial implementation of BND provides a means to directly capture data as a foundation for a collaboration tutor but does not yet fully support tutoring. Our next step was to perform two exploratory studies in which dyads of students used our integrated BND software to collaborate in solving modeling tasks. The data collected from these studies led us to identify five dimensions of collaborative and problem-solving behavior that point to the need for abstraction of student actions to better recognize, analyze, and provide feedback on collaboration. We also interviewed a domain expert who provided evidence for the advantage of bootstrapping over manual creation of a collaboration tutor. We discuss plans to use these analyses to inform and extend our tools so that we can eventually reach our goal of tutoring collaboration.     

TEFRCF:标签熵特征表示的协同过滤个性化推荐算法

标签作为Web 2.0时代信息分类和检索的有效方式,已经成为近年的热点研究对象。标签推荐系统旨在利用标签数据为用户提供个性化推荐。现有的基于标签的推荐方法在预测用户对物品的兴趣度时往往倾向于赋予热门标签及其对应的热门物品较大的权重,导致权重偏差,降低了推荐结果的新颖性,未能充分反映用户个性化的兴趣。针对上述问题,定义了标签熵的概念来度量标签的不确定性,提出了标签熵特征表示的协同过滤个性化推荐算法。该算法通过引入标签熵来解决权重偏差问题,利用三分图形式描述用户-标签-项目之间的关系;构建基于标签熵特征表示的用户和项目特征表示,并通过特征相似性度量方法计算项目的相似性;最后利用用户标签行为和项目的相似性线性组合预测用户对项目的偏好值,并根据预测偏好值排序生成最终的推荐列表。在Last.fm数据集上的实验结果表明,该方法能够提高推荐准确性和新颖性,满足用户的个性化需求。     

A primal-dual perspective of online learning algorithms

We describe a novel framework for the design and analysis of online learning algorithms based on the notion of duality in constrained optimization. We cast a sub-family of universal online bounds as an optimization problem. Using the weak duality theorem we reduce the process of online learning to the task of incrementally increasing the dual objective function. The amount by which the dual increases serves as a new and natural notion of progress for analyzing online learning algorithms. We are thus able to tie the primal objective value and the number of prediction mistakes using the increase in the dual. Editors: Hans Ulrich Simon, Gabor Lugosi, Avrim Blum. A preliminary version of this paper appeared at the 19th Annual Conference on Learning Theory under the title “Online learning meets optimization in the dual”.     

Recommender system in collaborative learning environment using an influence diagram

Giving useful recommendations to students to improve collaboration in a learning experience requires tracking and analyzing student team interactions, identifying the problems and the target student. Previously, we proposed an approach to track students and assess their collaboration, but it did not perform any decision analysis to choose a recommendation for the student. In this paper, we propose an influence diagram, which includes the observable variables relevant for assessing collaboration, and the variable representing whether the student collaborates or not. We have analyzed the influence diagram with two machine learning techniques: an attribute selector, indicating the most important attributes that the model uses to recommend, and a decision tree algorithm revealing four different scenarios of recommendation. These analyses provide two useful outputs: (a) an automatic recommender, which can warn of problematic circumstances, and (b) a pedagogical support system (decision tree) that provides a visual explanation of the recommendation suggested.     

一种基于本体语义相似度的协同过滤推荐方法

协同过滤推荐是一种基于用户偏好的个性化推荐方法,一般包含两个步骤:首先根据用户或项目的标注信息计算出用户或项目的相似度,确定邻居集合;然后根据相似度进行排序推荐,其核心问题在于相似度的计算。为了更好地达到这一目的,近年来关于将用户社交网络信息融入相似度计算的方法受到广泛关注。用户的注册信息、项目评分和社交信息都可以作为用户比较的依据。基于此提出了通过构建用户本体,计算本体之间的语义相似度,从而找到相似用户集合,最终实现目标用户的推荐方法。该方法为本体技术与推荐系统的结合提供了一种思路,实验表明 它能够在一定程度上提高推荐的准确度。     

融合类别信息和用户兴趣度的协同过滤推荐算法

协同过滤直接根据用户的行为记录去预测其可能感兴趣的项目,是现今最成功、应用最广泛的推荐技术。推荐的准确度受相似性度量方法效果的影响。传统的相似性度量方法主要关注用户共同评分项之间的相似度,忽视了评分项目中的类别信息,在面对数据稀疏性问题时存在一定的不足。针对上述问题,提出基于分类信息 的评分矩阵填充方法,结合用户兴趣相似度计算方法并充分考虑到评分项目的类别信息,使得兴趣度的度量更加符合推荐系统应用的实际情况。实验结果表明,该算法可以弥补传统相似性度量方法的不足,缓解评分数据稀疏对协同过滤算法的影响,能够提高推荐的准确性、多样性和新颖性。     

Analyzing collaborative learning processes automatically: Exploiting the advances of computational linguistics in computer-supported collaborative learning

In this article we describe the emerging area of text classification research focused on the problem of collaborative learning process analysis both from a broad perspective and more specifically in terms of a publicly available tool set called TagHelper tools. Analyzing the variety of pedagogically valuable facets of learners’ interactions is a time consuming and effortful process. Improving automated analyses of such highly valued processes of collaborative learning by adapting and applying recent text classification technologies would make it a less arduous task to obtain insights from corpus data. This endeavor also holds the potential for enabling substantially improved on-line instruction both by providing teachers and facilitators with reports about the groups they are moderating and by triggering context sensitive collaborative learning support on an as-needed basis. In this article, we report on an interdisciplinary research project, which has been investigating the effectiveness of applying text classification technology to a large CSCL corpus that has been analyzed by human coders using a theory-based multi-dimensional coding scheme. We report promising results and include an in-depth discussion of important issues such as reliability, validity, and efficiency that should be considered when deciding on the appropriateness of adopting a new technology such as TagHelper tools. One major technical contribution of this work is a demonstration that an important piece of the work towards making text classification technology effective for this purpose is designing and building linguistic pattern detectors, otherwise known as features, that can be extracted reliably from texts and that have high predictive power for the categories of discourse actions that the CSCL community is interested in.

基于标签信息特征相似性的协同过滤个性化推荐

标签推荐系统旨在利用标签数据为用户提供个性化推荐。已有的基于标签的推荐方法往往忽视了用户和资源本身的特征,而且在相似性度量时仅针对项目相似性或用户相似性进行计算,并未充分考虑二者之间的有效融合,推荐结果的准确性较低。为了解决上述问题,将标签信息融入到结合用户相似性和项目相似性的协同过滤中,提出融合标签特征与相似性的协同过滤个性化推荐方法。该方法在充分考虑用户、项目以及标签信息的基础上,利用二维矩阵来定义用户-标签以及标签-项目之间的行为。构建用户和项目的标签特征表示,通过基于标签特征的相似性度量方法计算用户相似性和项目相似性。基于用户标签行为和用户与项目的相似性线性组合来预测用户对项目的偏好值,并根据预测偏好值排序,生成最终的推荐列表。在Last.fm数据集上的实验结果表明,该方法能够提高推荐的准确度,满足用户的个性化需求。     

用户和项目联合度对二分网络个性化推荐的影响

首先采用物质流动算法进行二部图相似系数投影,然后利用随机游走模型得到协同过滤结果。在计算相似系数时,采用了考虑用户和项目联合度分布特征的改进算法。通过数据模拟可知,在最优情况下推荐项目准确率提高了18. 19%,推荐项目多样性提高了21. 90%。对用户和项目联合度的分布进行了统计分析,结果表明,在最优情况下,其符合指数为--2. 33的指数分布。     

The concept of flow in collaborative game-based learning

Generally, high-school students have been characterized as bored and disengaged from the learning process. However, certain educational designs promote excitement and engagement. Game-based learning is assumed to be such a design. In this study, the concept of flow is used as a framework to investigate student engagement in the process of gaming and to explain effects on game performance and student learning outcome. Frequency 1550, a game about medieval Amsterdam merging digital and urban play spaces, has been examined as an exemplar of game-based learning. This 1-day game was played in teams by 216 students of three schools for secondary education in Amsterdam. Generally, these students show flow with their game activities, although they were distracted by solving problems in technology and navigation. Flow was shown to have an effect on their game performance, but not on their learning outcome. Distractive activities and being occupied with competition between teams did show an effect on the learning outcome of students: the fewer students were distracted from the game and the more they were engaged in group competition, the more students learned about the medieval history of Amsterdam. Consequences for the design of game-based learning in secondary education are discussed.     

Prototyping a collaborative learning tool for nurses

Abstract  A Nurse Collaborative Learning (Nurse Co-Learn) tool was designed and developed to help the Department of Nursing to introduce Information Technology to undergraduate nursing students. The MUCH (Many Using and Creating Hypermedia) system was used as the basis for this tool. The Nurse Co-Learn tool supported coordination between clinic personnel. Students were required to take on a role in the clinic, and to perform the relevant tasks for that role. Through participating in this role-playing game students learned word processing, database and spreadsheet skills, whilst performing authentic tasks. Three prototypes were developed and trialled; this paper highlights the features of the prototypes and the responses of the users.     

Supporting collaborative learning in engineering design

In team-based project courses, collaborative learning is the dominant learning mode. Collaborative learning has been shown to increase individual learning through co-construction and personal reflection. Rapid adoption of web-based communication and mobile computing by students provide opportunities to take advantage of computer-supported collaboration for engineering education. We present preliminary findings on a computer environment, the Kiva Web, that supports the activities of group collaboration for interdisciplinary engineering design teams. We have employed methods from human–computer interaction (HCI) to iterate on the design in the context of use. In this paper, we discuss the evolution of the Kiva Web and the implications for both student design teams and professional design teams.     

On the combination of user-based and item-based collaborative filtering

In this paper, we propose two new filtering algorithms which are a combination of user-based and item-based collaborative filtering schemes. The first one, Hybrid-Ib, identifies a reasonably large neighbourhood of similar users and then uses this subset to derive the item-based recommendation model. The second algorithm, Hybrid-CF, starts by locating items similar to the one for which we want a prediction, and then, based on that neighbourhood, it generates its user-based predictions. We start by describing the execution steps of the algorithms and proceed with extended experiments. We conclude that our algorithms are directly comparable to existing filtering approaches, with Hybrid-CF producing favorable or, in the worst case, similar results in all selected evaluation metrics.     

Collaborative reputation mechanisms for electronic marketplaces

Members of electronic communities are often unrelated to each other, they may have never met and have no information on each other's reputation. This kind of information is vital in Electronic Commerce interactions, where the potential counterpart's reputation can be a significant factor in the negotiation strategy. Collaborative reputation mechanisms can provide personalized evaluations of the various ratings assigned to each user to predict their reliabilities. While these reputation mechanisms are developed in the context of electronic commerce, they are applicable in other types of electronic communities such as chatrooms, newsgroups, mailing lists, etc.     

基于用户共现矩阵乘子的分布式协同过滤推荐

随着大数据时代的到来,应用数据量剧增,个性化推荐技术日趋重要。传统的推荐技术直接应用于大数据环境时会面临推荐精度低、推荐时延长以及网络开销大等问题,导致推荐性能急剧下降。针对上述问题,提出用户共现矩阵乘子推荐策略,将用户相似度矩阵与项目评分矩阵相乘得到用户对项目的预测评分矩阵,从而生成对每个用户的候选推荐项目集;在此基础上,根据分布式处理架构的特点对传统协同过滤算法进行并行化扩展,设计了基于用户的分布式协同过滤算法;最后通过重定义序列组合的MapReduce模式将多个子任务串联起来,自动地完成顺序化的执行。实验结果表明,该算法在分布式计算环境下具有良好的推荐精度和推荐效率。