Evaluating collaborative filtering recommendations inside large learning object repositories

Collaborative filtering (CF) algorithms are techniques used by recommender systems to predict the utility of items for users based on the similarity among their preferences and the preferences of other users. The enormous growth of learning objects on the internet and the availability of preferences of usage by the community of users in the existing learning object repositories (LORs) have opened the possibility of testing the efficiency of CF algorithms on recommending learning materials to the users of these communities. In this paper we evaluated recommendations of learning resources generated by different well known memory-based CF algorithms using two databases (with implicit and explicit ratings) gathered from the popular MERLOT repository. We have also contrasted the results of the generated recommendations with several existing endorsement mechanisms of the repository to explore possible relations among them. Finally, the recommendations generated by the different algorithms were compared in order to evaluate whether or not they were overlapping. The results found here can be used as a starting point for future studies that account for the specific context of learning object repositories and the different aspects of preference in learning resource selection.     

Privacy-preserving multi-criteria collaborative filtering

Privacy-preserving collaborative filtering schemes focus on eliminating the privacy threats inherent in single preference values, and the privacy risks in the multi-criteria preference domain are disregarded. In this work, we introduce randomized perturbation-based privacy-preserving approaches for multi-criteria collaborative filtering systems. Initially, the privacy protection methods efficiently used in traditional single-criterion systems are adapted onto multi-criteria ratings. However, these systems require intelligent protection mechanisms that are flexible and adapting to the structure of each sub-criterion. To achieve such a goal, we introduce a novel privacy-preserving protocol by adapting an entropy-based randomness determination procedure that can recover accuracy losses. The proposed protocol adjusts privacy-controlling parameters concerning the information inherent in each criterion. We experimentally evaluate the proposed schemes on three subsets of Yahoo!Movies multi-criteria preference dataset to demonstrate the effects of the proposed privacy-preserving schemes on both user privacy levels and prediction accuracy for differing sparsity rates. According to the obtained experimental outcomes, the proposed entropy-based privacy-preserving scheme can produce significantly more accurate predictions while maintaining an identical level of privacy provided by the traditional privacy protection scenario. The experimental results also confirm that the novel entropy-based privacy-preserving scheme maintains the confidentiality of personal preferences without severely compromising prediction accuracy.     

Social network data to alleviate cold-start in recommender system: A systematic review

Recommender Systems are currently highly relevant for helping users deal with the information overload they suffer from the large volume of data on the web, and automatically suggest the most appropriate items that meet users needs. However, in cases in which a user is new to Recommender System, the system cannot recommend items that are relevant to her/him because of lack of previous information about the user and/or the user-item rating history that helps to determine the users preferences. This problem is known as cold-start, which remains open because it does not have a final solution. Social networks have been employed as a good source of information to determine users preferences to mitigate the cold-start problem. This paper presents the results of a Systematic Literature Review on Collaborative Filtering-based Recommender System that uses social network data to mitigate the cold-start problem. This Systematic Literature Review compiled the papers published between 2011–2017, to select the most recent studies in the area. Each selected paper was evaluated and classified according to the depth which social networks used to mitigate the cold-start problem. The final results show that there are several publications that use the information of the social networks within the Recommender System; however, few research papers currently use this data to mitigate the cold-start problem.     

Recommender systems effect on the evolution of users’ choices distribution

Recommender systems’ (RSs) research has mostly focused on algorithms aimed at improving platform owners’ revenues and user’s satisfaction. However, RSs have additional effects, which are related to their impact on users’ choices. In order to avoid an undesired system behaviour and anticipate the effects of an RS, the literature suggests employing simulations.In this article we present a novel, well grounded and flexible simulation framework. We adopt a stochastic user’s choice model and simulate users’ repeated choices for items in the presence of alternative RSs. Properties of the simulated choices, such as their diversity and their quality, are analysed. We state four research questions, also motivated by identified research gaps, which are addressed by conducting an experimental study where three different data sets and five alternative RSs are used. We identify some important effects of RSs. We find that non-personalised RSs result in choices for items that have a larger predicted rating compared to personalised RSs. Moreover, when a user’s awareness set, which is the set containing the items that she can choose from, increases, then choices are more diverse, but the average quality (rating) of the choices decreases. Additionally, in order to achieve a higher choice diversity, increasing the awareness of the users is shown to be a more effective remedy than increasing the number of recommendations offered to the users.     

基于协同过滤的数字图书馆推荐系统研究

在海量的数字图书馆中,准确迅速地找到符合自身需要的图书是需要解决的主要问题。通过阐述传统的协同过滤算法,分析其特点以及存在的不足,并基于此提出一种改进的协同过滤算法,建立了推荐系统模型并应用到数字图书馆中。     

一种引入内容特征的混合推荐方法

本文使用聚类算法将项目和用户进行分组,从而引入内容特征,再结合协同过滤方法,构造一种混合的推荐方法。实验结果表明,本文的推荐方法在较高稀疏度下优于一般的协同过滤算法。     

Dual disentanglement of user–item interaction for recommendation with causal embedding

To achieve personalized recommendations, the recommender system selects the items that users may like by learning the collected user–item interaction data. However, the acquisition and use of data usually form a feedback loop, which leads to recommender systems suffering from popularity bias. To solve this problem, we propose a novel dual disentanglement of user–item interaction for recommendation with causal embedding (DDCE). Different from the existing work, our innovation is we take into account double-end popularity bias from the user-side and the item-side. Firstly, we perform a causal analysis of the reasons for user–item interaction and obtain the causal embedding representation of each part according to the analysis results. Secondly, on the item-side, we consider the influence of item attributes on popularity to improve the reliability of the item popularity. Then, on the user-side, we consider the effect of the time series when obtaining users’ interest. We model the contrastive learning task to disentangle users’ long–short-term interests, which avoids the bias of long–short-term interests overlapping, and use the attention mechanism to realize the dynamic integration of users’ long–short-term interests. Finally, we realize the disentanglement of user–item interaction reasons by decoupling user interest and item popularity. We experiment on two real-world datasets (Douban Movie and KuaiRec) to verify the significance of DDCE, the average improvement of DDCE in three evaluation metrics (NDCG, HR, and Recall) compared to the state-of-the-art model are 5.1106% and 4.1277% (MF as the backbone), 3.8256% and 3.2790% (LightGCN as the backbone), respectively.     

A relevance feedback mechanism for cluster-based retrieval

Contextual document clustering is a novel approach which uses information theoretic measures to cluster semantically related documents bound together by an implicit set of concepts or themes of narrow specificity. It facilitates cluster-based retrieval by assessing the similarity between a query and the cluster themes’ probability distribution. In this paper, we assess a relevance feedback mechanism, based on query refinement, that modifies the query’s probability distribution using a small number of documents that have been judged relevant to the query. We demonstrate that by providing only one relevance judgment, a performance improvement of 33% was obtained.     

A unifying and general account of fairness measurement in recommender systems

Fairness is fundamental to all information access systems, including recommender systems. However, the landscape of fairness definition and measurement is quite scattered with many competing definitions that are partial and often incompatible. There is much work focusing on specific – and different – notions of fairness and there exist dozens of metrics of fairness in the literature, many of them redundant and most of them incompatible. In contrast, to our knowledge, there is no formal framework that covers all possible variants of fairness and allows developers to choose the most appropriate variant depending on the particular scenario. In this paper, we aim to define a general, flexible, and parameterizable framework that covers a whole range of fairness evaluation possibilities. Instead of modeling the metrics based on an abstract definition of fairness, the distinctive feature of this study compared to the current state of the art is that we start from the metrics applied in the literature to obtain a unified model by generalization. The framework is grounded on a general work hypothesis: interpreting the space of users and items as a probabilistic sample space, two fundamental measures in information theory (Kullback–Leibler Divergence and Mutual Information) can capture the majority of possible scenarios for measuring fairness on recommender system outputs. In addition, earlier research on fairness in recommender systems could be viewed as single-sided, trying to optimize some form of equity across either user groups or provider/procurer groups, without considering the user/item space in conjunction, thereby overlooking/disregarding the interplay between user and item groups. Instead, our framework includes the notion of statistical independence between user and item groups. We finally validate our approach experimentally on both synthetic and real data according to a wide range of state-of-the-art recommendation algorithms and real-world data sets, showing that with our framework we can measure fairness in a general, uniform, and meaningful way.     

概率标引模型研究

本文对二值独立性标引模型、DIA模型和2—Poisson模型三种典型的概率标引模型进行了研究,分析了其原理和处理过程,评价了其性能。     

A social recommendation method based on an adaptive neighbor selection mechanism

Recommender systems are techniques to make personalized recommendations of items to users. In e-commerce sites and online sharing communities, providing high quality recommendations is an important issue which can help the users to make effective decisions to select a set of items. Collaborative filtering is an important type of the recommender systems that produces user specific recommendations of the items based on the patterns of ratings or usage (e.g. purchases). However, the quality of predicted ratings and neighbor selection for the users are important problems in the recommender systems. Selecting suitable neighbors set for the users leads to improve the accuracy of ratings prediction in recommendation process. In this paper, a novel social recommendation method is proposed which is based on an adaptive neighbor selection mechanism. In the proposed method first of all, initial neighbors set of the users is calculated using clustering algorithm. In this step, the combination of historical ratings and social information between the users are used to form initial neighbors set for the users. Then, these neighbor sets are used to predict initial ratings of the unseen items. Moreover, the quality of the initial predicted ratings is evaluated using a reliability measure which is based on the historical ratings and social information between the users. Then, a confidence model is proposed to remove useless users from the initial neighbors of the users and form a new adapted neighbors set for the users. Finally, new ratings of the unseen items are predicted using the new adapted neighbors set of the users and the $top\_N$ interested items are recommended to the active user. Experimental results on three real-world datasets show that the proposed method significantly outperforms several state-of-the-art recommendation methods.     

A new robust relevance model in the language model framework

In this paper, a new robust relevance model is proposed that can be applied to both pseudo and true relevance feedback in the language-modeling framework for document retrieval. There are at least three main differences between our new relevance model and other relevance models. The proposed model brings back the original query into the relevance model by treating it as a short, special document, in addition to a number of top-ranked documents returned from the first round retrieval for pseudo feedback, or a number of relevant documents for true relevance feedback. Second, instead of using a uniform prior as in the original relevance model proposed by Lavrenko and Croft, documents are assigned with different priors according to their lengths (in terms) and ranks in the first round retrieval. Third, the probability of a term in the relevance model is further adjusted by its probability in a background language model. In both pseudo and true relevance cases, we have compared the performance of our model to that of the two baselines: the original relevance model and a linear combination model. Our experimental results show that the proposed new model outperforms both of the two baselines in terms of mean average precision.