KEYWORD EXTRACTION STRATEGY FOR ITEM BANKS TEXT CATEGORIZATION

We proposed a feature selection approach, Patterned Keyword in Phrase ( PKIP ), to text categorization for item banks. The item bank is a collection of textual question items that are short sentences. Each sentence does not contain enough relevant words for directly categorizing by the traditional approaches such as "bag-of-words." Therefore, PKIP was designed to categorize such question item using only available keywords and their patterns. PKIP identifies the appropriate keywords by computing the weight of all words. In this paper, two keyword selection strategies are suggested to ensure the categorization accuracy of PKIP. PKIP was implemented and tested with the item bank of Thai high primary mathematics questions. The test results have proved that PKIP is able to categorize the question items correctly and the two keyword selection strategies can extract the very informative keywords.     

Categorization for grouping associative items using data mining in item-based collaborative filtering

Recommendation systems have been investigated and implemented in many ways. In particular, in the case of a collaborative filtering system, the most important issue is how to manipulate the personalized recommendation results for better user understandability and satisfaction. A collaborative filtering system predicts items of interest for users based on predictive relationships discovered between each item and others. This paper proposes a categorization for grouping associative items discovered by mining, for the purpose of improving the accuracy and performance of item-based collaborative filtering. It is possible that, if an associative item is required to be simultaneously associated with all other groups in which it occurs, the proposed method can collect associative items into relevant groups. In addition, the proposed method can result in improved predictive performance under circumstances of sparse data and cold-start initiation of collaborative filtering starting from a small number of items. In addition, this method can increase prediction accuracy and scalability because it removes the noise generated by ratings on items of dissimilar content or level of interest. The approach is empirically evaluated by comparison with k-means, average link, and robust, using the MovieLens dataset. The method was found to outperform existing methods significantly.     

On a serendipity-oriented recommender system based on folksonomy

This paper proposes a recommendation method that focuses on not only predictive accuracy but also serendipity. On many of the conventional recommendation methods, items are categorized according to their attributes (a genre, an authors, etc.) by the recommender in advance, and recommendation is made using the results of the categorization. In this study, impressions of users to items are adopted as a feature of the items, and each item is categorized according to the feature. Impressions used in such categorization are prepared using folksonomy, which classifies items using tags given by users. Next, the idea of “concepts” was introduced to avoid synonym and polysemy problems of tags. “Concepts” are impressions of users on items inferred from attached tags of folksonomy. The inferring method was also devised. A recommender system based on the method was developed in java language, and the effectiveness of the proposed method was verified through recommender experiments.     

面向冷启动用户偏好获取的自适应物品询问列表生成方法

新用户进入系统之初,由于缺少必要的行为信息,推荐系统无法给出高质量的推荐,为了提高推荐精度,可以使用主动学习技术,使系统主动向用户发起相关物品的评分请求,从而通过用户的交互信息了解用户的偏好信息.现有的主动学习方法基于不确定性、熵等信息选择物品让用户提供反馈.然而由于用户的差异性,其适合的最佳策略是不一样的.因此,本文针对新用户冷启动问题,将要向用户发起询问的物品分阶段产生,通过用户对上一阶段问题的不同反馈,通过决策树模型个性化的动态选择下一阶段适合的策略来生成物品列表以征求其评价.同时,采用遗传算法来优化各个阶段的最优物品个数.在实际数据集上的实验表明,通过这种自适应生成询问物品列表的方法能够改进推荐效果.     

基于多维特征交叉的深度协同过滤算法

基于深度学习的推荐算法最初以用户和物品的ID信息作为输入,但是ID无法很好地表现用户与物品的特征。在原始数据中,用户对物品的评分数据在一定程度上能表现出用户和物品的特征,但是未考虑用户的评分偏好以及物品的热门程度。在评分任务中使用隐式反馈和ID信息作为用户与物品的特征,在消除用户主观性对特征造成的噪声的同时在一定程度上缓解冷启动问题,利用单层神经网络对原始高维稀疏特征降维,使用特征交叉得到用户与物品的低阶交互,再利用神经网络捕获用户与物品的高阶交互,有效提取了特征间的高低阶交互。在四个公开数据集上的实验表明,该算法能有效提高推荐精度。     

基于度量学习的多空间推荐系统

隐式反馈具有数据获取成本小、形式广泛的特点,因此在现代推荐系统中被广泛使用.由于用户的隐式反馈通常是稀疏,不平衡,且含义不明确的.因此,想要准确学习用户和物品之间的复杂交互具有挑战性.传统的基于矩阵分解的推荐方法只能建模用户-物品之间的相似性.同时,矩阵分解使用点积运算作为相似度评估方式,而点积运算不满足三角不等式,即不能将用户-物品相似性传递到用户-用户以及物品-物品的相似性建模.因此,矩阵分解不足以在隐式反馈中充分建模用户和物品的关系.尽管现在有基于隐式反馈使用欧式距离来度量用户-物品相似度的度量学习方法,使得对应的推荐方法能够满足三角不等式.但是,现有的度量方法通常会将每个用户或者物品表示为度量空间中的单个点,进而在单个空间内通过用户-物品之间的距离来表征用户-物品之间的相似性.由于在不同的环境下,用户对于同一种类型的物品的偏好也可能存在差异.基于单个空间的用户、物品嵌入向量有可能无法满足用户具有的多种偏好和物品具有的多种属性,进而限制了推荐系统的性能.为了充分刻画用户和物品,我们尝试从多个侧面对于用户和物品进行表示,并提出了一个基于多空间的度量学习(MML)框架.通过设计整合多个空间相似性的度量方式,我们将用户和物品投影到多个空间中进行细粒度的表示.另外,我们设计了一种经过校准的优化策略,包括经过校准的最大间隔损失函数和经过校准的采样方法.在保持多空间度量学习表示能力的同时,确保框架的有效性.最后,模型通过训练好的用户、物品向量,对于稀疏的用户-物品交互矩阵进行填补.在动态更新空间权重的同时,可以赋予模型新的训练视角,最终实现端到端的训练.通过四个真实世界推荐数据集上进行的大量实验表明,MML可以在Recall和nDCG衡量指标上将目前最优的对比算法提高40%以上.     

A hybrid recommendation algorithm adapted in e-learning environments

With the rapid development of online learning technology, a huge amount of e-learning materials have been generated which are highly heterogeneous and in various media formats. Besides, e-learning environments are highly dynamic with the ever increasing number of learning resources that are naturally distributed over the network. On the other hand, in the online learning scenario, it is very difficult for users without sufficient background knowledge to choose suitable resources for their learning. In this paper, a hybrid recommender system is proposed to recommend learning items in users’ learning processes. The proposed method consists of two steps: (1) discovering content-related item sets using item-based collaborative filtering (CF), and (2) applying the item sets to sequential pattern mining (SPM) algorithm to filter items according to common learning sequences. The two approaches are combined to recommend potentially useful learning items to guide users in their current learning processes. We also apply the proposed approach to a peer-to-peer learning environment for resource pre-fetching where a central directory of learning items is not available. Experiments are conducted in a centralized and a P2P online learning systems for the evaluation of the proposed method and the results show good performance of it.     

一种基于文本卷积神经网络的推荐算法

传统的矩阵因子分解模型不能有效提取用户和物品特征,而基于深度学习模型可以很好地提取特征信息。当前,主流的基于深度学习推荐算法只是单一地将神经网络的输出或物品特征与用户特征乘积的形式来做推荐预测,不能充分挖掘用户和物品之间的关系。基于此,本文提出一种基于文本卷积神经网络与带偏置项的奇异值分解(BiasSVD)结合的推荐算法,利用文本卷积神经网络(TextCNN)来充分提取用户和物品的特征信息,然后用奇异值分解方法来做推荐,深层次理解文档上下文信息,进一步提高推荐的准确性。将该算法在MovieLens的2个真实数据集上做广泛的评估分析,推荐的准确度要明显优于ConvMF算法及主流深度学习推荐算法。     

A weight-based item recommendation approach for electronic commerce systems

The recommendation system is a useful tool that can be employed to identify potential relationships between items and users in electronic commerce systems. Consequently, it can remarkably improve the efficiency of a business. The topic of how to enhance the accuracy of a recommendation has attracted much attention by researchers over the past decade. As such, many methods to accomplish this task have been introduced. However, more complex calculations are normally necessary to achieve a higher accuracy, which is not suitable for a real-time system. Hence, in this paper, we propose a weight-based item recommendation approach to provide a balanced formula between the recommended accuracy and the computational complexity. The proposed methods employ a newly defined distance to describe the relationship between the users and the items, after which the recommendations and predictive algorithms are developed. A data analysis based on the MovieLens datasets indicates that the methods applied can obtain suitable prediction accuracy and maintain a relatively low computational complexity.     

结合评分比例因子及项目属性的协同过滤算法

针对传统的协同过滤算法存在用户评分矩阵稀疏及未考虑项目属性之间关系的问题,提出了结合评分比例因子及项目属性的协同过滤算法。首先利用评分矩阵得出项目之间的共同与非共同评分用户数量比矩阵,以此增加项目共同评分用户的影响度,减少用户—项目评分矩阵的稀疏性对项目相似度计算带来的误差;然后对项目属性量化得出其对项目相似度的影响权重,提高项目相似度计算的准确性,根据以上两点提出了一种结合评分比例因子及项目属性权重作为项目相似度权重的算法。实验结果表明,该算法在召回率和准确率上相比现有的方法分别提高了5.1%和4.7%,适用于电商类网站的个性化推荐。     

A Unified Framework for Providing Recommendations in Social Tagging Systems Based on Ternary Semantic Analysis

Social Tagging is the process by which many users add metadata in the form of keywords, to annotate and categorize items (songs, pictures, Web links, products, etc.). Social tagging systems (STSs) can provide three different types of recommendations: They can recommend 1) tags to users, based on what tags other users have used for the same items, 2) items to users, based on tags they have in common with other similar users, and 3) users with common social interest, based on common tags on similar items. However, users may have different interests for an item, and items may have multiple facets. In contrast to the current recommendation algorithms, our approach develops a unified framework to model the three types of entities that exist in a social tagging system: users, items, and tags. These data are modeled by a 3-order tensor, on which multiway latent semantic analysis and dimensionality reduction is performed using both the Higher Order Singular Value Decomposition (HOSVD) method and the Kernel-SVD smoothing technique. We perform experimental comparison of the proposed method against state-of-the-art recommendation algorithms with two real data sets (Last.fm and BibSonomy). Our results show significant improvements in terms of effectiveness measured through recall/precision.     

自适应用户的Item-based协同过滤推荐算法

传统Item-based协同过滤算法计算两个条目间相似性时, 将每个评分视为同等重要, 忽略了共评用户（对两个条目共同评分的用户）与目标用户间的相似性对条目间相似性的影响。针对此问题, 提出了一种自适应用户的Item-based协同过滤算法。该算法将共评用户与目标用户的相似性作为共评用户评分重要性的权重, 以实现针对不同的目标用户, 为目标条目选择不同的、适合目标用户的最近邻居集, 从而提高推荐准确性。实验结果表明, 提出的算法可以显著提高推荐系统的推荐质量。     

考虑项目属性的协同过滤推荐模型

针对传统的基于用户的协同过滤(UCF)模型在相似性度量过程中没有充分考虑项目属性的问题,提出了两种考虑项目属性的协同过滤推荐模型。模型首先对用户评分相似性进行优化;然后从项目属性的角度统计用户关于不同项目的评价次数,获得优化的基于项目属性的用户相似性;最后通过自适应平衡因子协调处理两方面的相似性结果进行项目预测与推荐。实验结果表明,在不同的数据集中,新提出的模型不仅时间花费较为合理,而且评分预测准确性明显提高,平均提高了5%,从而证明了模型在改进用户相似性度量精度方面的有效性。     

Construction and analysis of educational tests using abductive machine learning

Recent advances in educational technologies and the wide-spread use of computers in schools have fueled innovations in test construction and analysis. As the measurement accuracy of a test depends on the quality of the items it includes, item selection procedures play a central role in this process. Mathematical programming and the item response theory (IRT) are often used in automating this task. However, when the item bank is very large, the number of item combinations increases exponentially and item selection becomes more tedious. To alleviate the computational complexity, researchers have previously applied heuristic search and machine learning approaches, including neural networks, to solve similar problems. This paper proposes a novel approach that uses abductive network modeling to automatically identify the most-informative subset of test items that can be used to effectively assess the examinees without seriously degrading accuracy. Abductive machine learning automatically selects only effective model inputs and builds an optimal network model of polynomial functional nodes that minimizes a predicted squared error criterion. Using a training dataset of 1500 cases (examinees) and 45 test items, the proposed approach automatically selected only 12 items which classified an evaluation population of 500 cases with 91% accuracy. Performance is examined for various levels of model complexity and compared with that of statistical IRT-based techniques. Results indicate that the proposed approach significantly reduces the number of test items required while maintaining acceptable test quality.     

项目子相似度融合的协同过滤推荐算法

针对用户评分数据稀疏性和项目最近邻寻找的不准确性问题,提出了一种项目子相似度融合的协同过滤推荐算法.该算法根据目标用户每一属性取值,选取与该属性值一致的用户作为用户子空间,并在此空间上计算目标项目与其他项目之间的相似度(称其为项目子相似度).在此基础上,以项目子相似度为依据选取目标项目的K最近邻,计算其预测评分;最后对用户不同属性上的预测评分进行加权求和,得到目标项目的最终评分.实验结果表明,该算法能准确地选取目标项目的最近邻,明显改善了推荐质量.     

基于用户聚类与项目划分的优化推荐算法

针对传统协同过滤推荐算法没有充分考虑用户属性及项目类别划分等因素对相似度计算产生的影响,存在数据稀疏性,从而导致推荐准确度不高的问题.提出一种基于用户属性聚类与项目划分的协同过滤推荐算法,算法对推荐准确度有重要影响的相似度计算进行了充分考虑.先对用户采用聚类算法以用户身份属性聚类,进而再对项目进行类别划分,在相似度计算中增加类别相似度,考虑共同评分用户数通过加权系数进行综合相似度计算,最后结合平均相似度,采用阈值法综合得出最近邻.实验结果表明,所提算法能够有效提高推荐精度,为用户提供更准确的推荐项目.     

AICF: Attention-based item collaborative filtering

Item-to-item collaborative filtering (short for ICF) has been widely used in ecommerce websites due to his interpretability and simplicity in real-time personalized recommendation. The focus of ICF is to calculate the similarity between items. With the rapid development of machine learning in recent years, it takes similarity model instead of cosine similarity and Pearson coefficient to calculate the similarity between items in recommendation. However, the existing similarity models can not sufficient to express the preferences of users for different items. In this work, we propose a novel attention-based item collaborative filtering model(AICF) which adopts three different attention mechanisms to estimate the weights of historical items that users have interacted with. Compared with the state-of-the-art recommendation models, the AICF model with simple attention mechanism Self-Attention can better estimate the weight of historical items on non-sparse data sets. Due to depth models can model complex connection between items, our model with the more complex Transformer achieves superior recommendation performance on sparse data. Extensive experiments on ML-1M and Pinterest-20 show that the proposed model greatly outperforms other novel models in recommendation accuracy and provides users with personalized recommendation list more in line with their interests.     

社交网络中融合社交关系和语义信息的推荐算法

协同过滤方法广泛应用于推荐,但是数据稀疏成为模型提供高质量推荐的一大障碍.为了解决此问题,文中提出融合社交关系和语义信息的推荐算法,提高协同过滤方法的推荐性能,有机融合稀疏的用户行为记录、项目的社交信息和项目的语义信息.应用矩阵分解技术把行为矩阵和项目社交关系映射到一个低维的特征空间,提供项目社交关系信息分解的显式解释,分析关系信息对用户行为偏好产生的影响.同时,使用社会化因子正则的级联去噪自编码器模型学习项目语义特征,改进传统深度学习模型.在真实腾讯微博和Twitter数据集上的实验表明,文中方法有效提高召回率、准确率和推荐效率.     

基于图神经网络的异构信任推荐算法

传统基于图神经网络的社交推荐算法通过加强用户和项目特征的学习提升预测精度,但随着用户数据日益稀疏和社交关系趋于复杂,推荐质量提升缓慢。为挖掘用户和项目的潜在关联关系,提出一种结合图神经网络的异构信任推荐算法（GraphTrust）。在显式信任关系的基础上获取用户的潜在好友,根据动态影响力传播模型将图神经网络中的节点和边进行分类,通过不同类型的边在不同节点间进行影响力传播扩散,捕捉隐藏在高阶网络结构中的影响力扩散特征,并使用户和项目的潜在特征随着影响力传播过程达到平衡状态,最终将用户交互的项目特征作为辅助特征与用户特征聚合进行评分预测。在Yelp和Flickr数据集上的实验结果表明,当潜在特征维数为64时,GraphTrust算法相比于DiffNet++算法的命中率和归一化折损累计增益分别提升了13.2%、22.2%和20.4%、25.5%,在一定程度上提高了推荐过程的可解释性和预测精度,并且缓解了数据稀疏问题。     

Improvement of collaborative filtering using rating normalization

With the advent of the Internet, the types and amount of information one can access have increased dramatically. In today’s overwhelming information environment, recommendation systems that quickly analyze large amounts of available information and help users find items of interest are increasingly needed. This paper proposes an improvement of an existing preference prediction algorithm to increase the accuracy of recommendation systems. In a recommendation system, prediction of items preferred by users is based on their ratings. However, individual users with the same degree of satisfaction to an item may give different ratings to the item. We intend to make more precise preference prediction by perceiving differences in users’ rating dispositions. The proposed method consists of two processes of perceiving users’ rating dispositions with clustering and of performing rating normalization according to such rating dispositions. The experimental results show that our method yields higher performance than ordinary collaborative filtering approach.