A fine-grained and multi-context-aware learning path recommendation model over knowledge graphs for online learning communities

Existing approaches to learning path recommendation for online learning communities mainly rely on the individual characteristics of users or the historical records of their learning processes, but pay less attention to the semantics of users’ postings and the context. To facilitate the knowledge understanding and personalized learning of users in online learning communities, it is necessary to conduct a fine-grained analysis of user data to capture their dynamical learning characteristics and potential knowledge levels, so as to recommend appropriate learning paths. In this paper, we propose a fine-grained and multi-context-aware learning path recommendation model for online learning communities based on a knowledge graph. First, we design a multidimensional knowledge graph to solve the problem of monotonous and incomplete entity information presentation of the single layer knowledge graph. Second, we use the topic preference features of users’ postings to determine the starting point of learning paths. We then strengthen the distant relationship of knowledge in the global context using the multidimensional knowledge graph when generating and recommending learning paths. Finally, we build a user background similarity matrix to establish user connections in the local context to recommend users with similar knowledge levels and learning preferences and synchronize their subsequent postings. Experiment results show that the proposed model can recommend appropriate learning paths for users, and the recommended similar users and postings are effective.     

Collaborative filtering with facial expressions for online video recommendation

Online video recommender systems help users find videos suitable for their preferences. However, they have difficulty in identifying dynamic user preferences. In this study, we propose a new recommendation procedure using changes of users’ facial expressions captured every moment. Facial expressions portray the users’ actual emotions about videos. We can utilize them to discover dynamic user preferences. Further, because the proposed procedure does not rely on historical rating or purchase records, it properly addresses the new user problem, that is, the difficulty in recommending products to users whose past rating or purchase records are not available. To validate the recommendation procedure, we conducted experiments with footwear commercial videos. Experiment results show that the proposed procedure outperforms benchmark systems including a random recommendation, an average rating approach, and a typical collaborative filtering approach for recommendation to both new and existing users. From the results, we conclude that facial expressions are a viable element in recommendation.     

A group recommendation system for online communities

Online communities are virtual spaces over the Internet in which a group of people with similar interests or purposes interact with others and share information. To support group activities in online communities, a group recommendation procedure is needed. Though there have been attempts to establish group recommendation, they focus on off-line environments. Further, aggregating individuals’ preferences into a group preference or merging individual recommendations into group recommendations—an essential component of group recommendation—often results in dissatisfaction of a small number of group members while satisfying the majority. To support group activities in online communities, this paper proposes an improved group recommendation procedure that improves not only the group recommendation effectiveness but also the satisfaction of individual group members. It consists of two phases. The first phase was to generate a recommendation set for a group using the typical collaborative filtering method that most existing group recommendation systems utilize. The second phase was to remove irrelevant items from the recommendation set in order to improve satisfaction of individual members’ preferences. We built a prototype system and performed experiments. Our experiment results showed that the proposed system has consistently higher precision and individual members are more satisfied.     

Multi-feature fused collaborative attention network for sequential recommendation with semantic-enriched contrastive learning

Recently, graph neural network (GNN) has been widely used in sequential recommendation because of its powerful ability to capture high-order collaborative relations, greatly promoting recommendation performance. However, some existing GNN-based methods fail to make full use of multiple relevant features of nodes and ignore the impact of semantic association between nodes on extracting user preferences. To this end, we propose a multi-feature fused collaborative attention network MASR, which sufficiently learns the temporal and positional features of nodes, and innovatively measures the importance of these two features for analyzing the nodes’ dynamic patterns. In addition, we incorporate semantic-enriched contrastive learning into collaborative filtering to enhance the semantic association between nodes and reduce the noise from the structural neighborhood, which has a positive effect on the sequential recommendation. Compared with the baseline models, the performance of MASR on MovieLens, CDs and Beauty datasets is improved by 2.0%, 2.1% and 1.7% respectively, proving its effectiveness in the sequential recommendation.     

Time and sequence awareness in similarity metrics for recommendation

Modeling the temporal context efficiently and effectively is essential to provide useful recommendations to users. In this work, we focus on improving neighborhood-based approaches where we integrate three different mechanisms to exploit temporal information. We first present an improved version of a similarity metric between users using a temporal decay function, then, we propose an adaptation of the Longest Common Subsequence algorithm to be used as a time-aware similarity metric, and we also redefine the neighborhood-based recommenders to be interpreted as ranking fusion techniques where the neighbor interaction sequence can be exploited by considering the last common interaction between the neighbor and the user.We demonstrate the effectiveness of these approaches by comparing them with other state-of-the-art recommender systems such as Matrix Factorization, Neural Networks, and Markov Chains under two realistic time-aware evaluation methodologies (per user and community-based). We use several evaluation metrics to measure both the quality of the recommendations – in terms of ranking relevance – and their temporal novelty or freshness. According to the obtained results, our proposals are highly competitive and obtain better results than the rest of the analyzed algorithms, producing improvements under the two evaluation dimensions tested consistently through three real-world datasets.     

A rough set-based association rule approach for a recommendation system for online consumers

Increasing use of the Internet gives consumers an evolving medium for the purchase of products and services and this use means that the determinants for online consumers’ purchasing behaviors are more important. Recommendation systems are decision aids that analyze a customer's prior online purchasing behavior and current product information to find matches for the customer's preferences. Some studies have also shown that sellers can use specifically designed techniques to alter consumer behavior. This study proposes a rough set based association rule approach for customer preference analysis that is developed from analytic hierarchy process (AHP) ordinal data scale processing. The proposed analysis approach generates rough set attribute functions, association rules and their modification mechanism. It also determines patterns and rules for e-commerce platforms and product category recommendations and it determines possible behavioral changes for online consumers.     

Listening to the investors: A novel framework for online lending default prediction using deep learning neural networks

Online peer-to-peer (P2P) lending has developed dramatically over the last decade in China. But this rapid boom carries potential risks. Investors have incurred incalculable losses due to the recent increase in fraudulent and/or unreliable online P2P platforms. Hence, predicting and identifying potential default risk platforms is crucial at this juncture. To achieve this end, we propose a two-step method which employs a deep learning neural network to extract keywords from investor comments and then utilizes a bidirectional long short-term memory (BiLSTM) based model to predict the default risk of platforms. Experimental results on real-world datasets of about 1000 platforms show that in the keyword extraction phase, our model can better capture semantic features from highly colloquial comment-text and achieve significant improvement over other baselines. Additionally, in the default platform prediction stage, our model achieves an F1 value of 80.34% in identifying potential problem platforms, outperforming four baselines by 23.37%, 5.71%, 8.93%, and 4.98% of improvement and comprehensively verifying the effectiveness of our method. Our study provides an alternative solution for platform default risk prediction issues and validates the effectiveness of investor comments in revealing the risk situation of online lending platforms.     

A new deep graph attention approach with influence and preference relationship reconstruction for rate prediction recommendation

Graph neural networks have been frequently applied in recommender systems due to their powerful representation abilities for irregular data. However, these methods still suffer from the difficulties such as the inflexible graph structure, sparse and highly imbalanced data, and relatively shallow networks, limiting rate prediction ability for recommendations. This paper presents a novel deep dynamic graph attention framework based on influence and preference relationship reconstruction (DGA-IPR) for recommender systems to learn optimal latent representations of users and items. The entire framework involves a user branch and an item branch. An influence-based dynamic graph attention (IDGA) module, a preference-based dynamic graph attention (PDGA) module, and an adaptive fine feature extraction (AFFE) module are respectively constructed for each branch. Concretely, the first two attention modules concentrate on reconstructing influence and preference relationship graphs, breaking imbalanced and fixed constraints of graph structures. Then a deep feature aggregation block and an adaptive feature fusion operation are built, improving the network depth and capturing potential high-order information expressions. Besides, AFFE is designed to acquire finer latent features for users and items. The DGA-IPR architecture is formed by integrating IDGA, PDGA, and AFFE for users and items, respectively. Experiments reveal the superiority of DGA-IPR over existing recommendation models.     

A deep learning-based multi-turn conversation modeling for diagnostic Q&A document recommendation

Online healthcare communities (OHCs) have become producers of medical information. Solving the issue of how to effectively reuse such a large amount of medical data and discover its potential value is of the utmost importance for alleviating the shortage of medical resources. Online consultation has received widespread attention and population since its first appearance in 1999, and as a result, many diagnostic multi-turn questions and answers (Q&A) documents have become available. This type of document is formed by multiple rounds of patient questions and doctors’ diagnostic answers and contains massive medical knowledge and doctors’ diagnostic experience. Few studies concentrate on the modeling and recommendation of this type of document, yet making these documents convenient for reuse reduces the cost of medical consultation for patients and saves time addressing common diseases for doctors. In this paper, we focus on the modeling and understanding of diagnostic multi-turn Q&A records and propose a deep-learning recommendation framework based on patient medical information needs, the contents of Q&A records and doctor background information. With the evaluation based on a real dataset that contains pediatric consultation dialogues fetched from DingXiangYuan, a famous online consultation application in China, we found that the proposed model achieved a good performance on the recommendation of diagnostic multi-turn Q&A records and outperformed baseline models. In addition, we discussed a potential application scenario of the recommendation model, suggesting that the proposed model can promote the reduction of patient costs and doctors’ work pressure in countries or regions with insufficient medical resources.     

Analyzing review sentiments and product images by parallel deep nets for personalized recommendation

Recommendation is an effective marketing tool widely used in the e-commerce business, and can be made based on ratings predicted from the rating data of purchased items. To improve the accuracy of rating prediction, user reviews or product images have been used separately as side information to learn the latent features of users (items). In this study, we developed a hybrid approach to analyze both user sentiments from review texts and user preferences from item images to make item recommendations more personalized for users. The hybrid model consists of two parallel modules to perform a procedure named the multiscale semantic and visual analyses (MSVA). The first module is designated to conduct semantic analysis on review documents in various aspects with word-aware and scale-aware attention mechanisms, while the second module is assigned to extract visual features with block-aware and visual-aware attention mechanisms. The MSVA model was trained, validated and tested using Amazon Product Data containing sampled reviews varying from 492,970 to 1 million records across 22 different domains. Three state-of-the-art recommendation models were used as the baselines for performance comparisons. Averagely, MSVA reduced the mean squared error (MSE) of predicted ratings by 6.00%, 3.14% and 3.25% as opposed to the three baselines. It was demonstrated that combining semantic and visual analyses enhanced MSVA's performance across a wide variety of products, and the multiscale scheme used in both the review and visual modules of MSVA made significant contributions to the rating prediction.     

Evaluating deep transfer learning for whole-brain cognitive decoding

Research in many fields has shown that transfer learning (TL) is well-suited to improve the performance of deep learning (DL) models in datasets with small numbers of samples. This empirical success has triggered interest in the application of TL to cognitive decoding analyses with functional neuroimaging data. Here, we systematically evaluate TL for the application of DL models to the decoding of cognitive states (e.g., viewing images of faces or houses) from whole-brain functional Magnetic Resonance Imaging (fMRI) data. We first pre-train two DL architectures on a large, public fMRI dataset and subsequently evaluate their performance in an independent experimental task and a fully independent dataset. The pre-trained DL models consistently achieve higher decoding accuracies and generally require less training time and data than model variants that were not pre-trained, while also outperforming linear baseline models trained from scratch, clearly underlining the benefits of pre-training. We demonstrate that these benefits arise from the ability of the pre-trained models to reuse many of their learned features when training with new data, providing deeper insights into the mechanisms giving rise to the benefits of pre-training. Yet, we also surface nuanced challenges for whole-brain cognitive decoding with DL models when interpreting the decoding decisions of the pre-trained models, as these have learned to utilize the fMRI data in unforeseen and counterintuitive ways to identify individual cognitive states.     

HOG-assisted deep feature learning for pedestrian gender recognition

Pedestrian gender recognition is a very challenging problem, since the viewpoint variations, illumination changes, occlusion, and poor quality are usually encountered in the pedestrian images. To address this problem, an effective HOG-assisted deep feature learning (HDFL) method is proposed in this paper. The key novelty lies in the design of HDFL network to effectively explore both deep-learned feature and weighted histogram of oriented gradient (HOG) feature for the pedestrian gender recognition. Specifically, the deep-learned and weighted HOG feature extraction branches are simultaneously performed on the input pedestrian image. A feature fusion process is subsequently conducted to obtain a more robust and discriminative feature, which is then fed to a softmax classifier for pedestrian gender recognition. Extensive experiments on multiple existing pedestrian image datasets have shown that the proposed HDFL method is able to effectively recognize the pedestrian gender, and consistently outperforms the state-of-the-art methods.     

Multiheaded deep learning chatbot for increasing production and marketing

Some businesses on product development prefer to use a chatbot for judging the customer's view. Today, the ability of a chatbot to consider the context is challenging due to its technical nature. Sometimes, it may misjudge the context, making the wrong decision in predicting the product's originality in the market. This task of chatbot helps the enterprise make huge profits from accurate predictions. However, chatbots may commit errors in dialogs and bring inappropriate responses to users, reducing the confidentiality of product and marketing information. This, in turn, reduces the enterprise gain and imposes cost complications on businesses. To improve the performance of chatbots, AI models are used based on deep learning concepts. This research proposes a multi-headed deep neural network (MH-DNN) model for addressing the logical and fuzzy errors caused by retrieval chatbot models. This model cuts down on the error raised from the information loss. Our experiments extensively trained the model on a large Ubuntu dialog corpus. The recall evaluation scores showed that the MH-DNN approach slightly outperformed selected state-of-the-art retrieval-based chatbot approaches. The results obtained from the MHDNN augmentation approach were pretty impressive. In our proposed work, the MHDNN algorithm exhibited accuracy rates of 94% and 92%, respectively, with and without the help of the Seq2Seq technique.     

Online portfolio management via deep reinforcement learning with high-frequency data

Recently, models that based on Transformer (Vaswani et al., 2017) have yielded superior results in many sequence modeling tasks. The ability of Transformer to capture long-range dependencies and interactions makes it possible to apply it in the field of portfolio management (PM). However, the built-in quadratic complexity of the Transformer prevents its direct application to the PM task. To solve this problem, in this paper, we propose a deep reinforcement learning-based PM framework called LSRE-CAAN, with two important components: a long sequence representations extractor and a cross-asset attention network. Direct Policy Gradient is used to solve the sequential decision problem in the PM process. We conduct numerical experiments in three aspects using four different cryptocurrency datasets, and the empirical results show that our framework is more effective than both traditional and state-of-the-art (SOTA) online portfolio strategies, achieving a 6x return on the best dataset. In terms of risk metrics, our framework has an average volatility risk of 0.46 and an average maximum drawdown risk of 0.27 across the four datasets, both of which are lower than the vast majority of SOTA strategies. In addition, while the vast majority of SOTA strategies maintain a poor turnover rate of approximately greater than 50% on average, our framework enjoys a relatively low turnover rate on all datasets, efficiency analysis illustrates that our framework no longer has the quadratic dependency limitation.     

An adaptive online learning algorithm for distributed convex optimization with coupled constraints over unbalanced directed graphs

This paper investigates a distributed optimization problem over multi-agent networks subject to both local and coupled constraints in a non-stationary environment, where a set of agents aim to cooperatively minimize the sum of locally time-varying cost functions when the communication graphs are time-changing connected and unbalanced. Based on dual decomposition, we propose a distributed online dual push-sum learning algorithm by incorporating the push-sum protocol into dual gradient method. We then show that the regret bound has a sublinear growth of O(T^p) and the constraint violation is also sublinear with order of $O\left(T^{1?p/2}\right),$ where T is the time horizon and 0 < p ≤ 1/2. Finally, simulation experiments on a plug-in electric vehicle charging problem are utilized to verify the performance of the proposed algorithm. The proposed algorithm is adaptive without knowing the total number of iterations T in advance. The convergence results are established on more general unbalanced graphs without the boundedness assumption on dual variables. In addition, more privacy concerns are guaranteed since only dual variables related with coupled constraints are exchanged among agents.     

钢板表面缺陷深度主动学习高效分类方法

针对钢板表面缺陷图像分类传统深度学习算法中需要大量标签数据的问题,提出一种基于主动学习的高效分类方法。该方法包含一个轻量级的卷积神经网络和一个基于不确定性的主动学习样本筛选策略。神经网络采用简化的convolutional base进行特征提取,然后用全局池化层替换掉传统密集连接分类器中的隐藏层来减轻过拟合。为了更好的衡量模型对未标签图像样本所属类别的不确定性,首先将未标签图像样本传入到用标签图像样本训练好的模型,得到模型对每一个未标签样本关于标签的概率分布(probability distribution over classes, PDC),然后用此模型对标签样本进行预测并得到模型对每个标签的平均PDC。将两类分布的KL-divergence值作为不确定性指标来筛选未标签图像进行人工标注。根据在NEU-CLS开源缺陷数据集上的对比实验,该方法可以通过44%的标签数据实现97%的准确率,极大降低标注成本。     

A deep recommendation model of cross-grained sentiments of user reviews and ratings

The matrix factorization model based on user-item rating data has been widely studied and applied in recommender systems. However, data sparsity, the cold-start problem, and poor explainability have restricted its performance. Textual reviews usually contain rich information about items’ features and users’ sentiments and preferences, which can solve the problem of insufficient information from only user ratings. However, most recommendation algorithms that take sentiment analysis of review texts into account are either fine- or coarse-grained, but not both, leading to uncertain accuracy and comprehensiveness regarding user preference. This study proposes a deep learning recommendation model (i.e., DeepCGSR) that integrates textual review sentiments and the rating matrix. DeepCGSR uses the review sets of users and items as a corpus to perform cross-grained sentiment analysis by combining fine- and coarse-grained levels to extract sentiment feature vectors for users and items. Deep learning technology is used to map between the extracted feature vector and latent factor through the rating-based matrix factorization model and obtain deep, nonlinear features to predict the user's rating of an item. Iterative experiments on e-commerce datasets from Amazon show that DeepCGSR consistently outperforms the recommendation models LFM, SVD++, DeepCoNN, TOPICMF, and NARRE. Overall, comparing with other recommendation models, the DeepCGSR model demonstrated improved evaluation results by 14.113% over LFM, 13.786% over SVD++, 9.920% over TOPICMF, 5.122% over DeepCoNN, and 2.765% over NARRE. Meanwhile, the DeepCGSR has great potential in fixing the overfitting and cold-start problems. Built upon previous studies and findings, the DeepCGSR is the state of the art, moving the design and development of the recommendation algorithms forward with improved recommendation accuracy.     

Iterative rule-guided reasoning over sparse knowledge graphs with deep reinforcement learning

In recent years, reasoning over knowledge graphs (KGs) has been widely adapted to empower retrieval systems, recommender systems, and question answering systems, generating a surge in research interest. Recently developed reasoning methods usually suffer from poor performance when applied to incomplete or sparse KGs, due to the lack of evidential paths that can reach target entities. To solve this problem, we propose a hybrid multi-hop reasoning model with reinforcement learning (RL) called SparKGR, which implements dynamic path completion and iterative rule guidance strategies to increase reasoning performance over sparse KGs. Firstly, the model dynamically completes the missing paths using rule guidance to augment the action space for the RL agent; this strategy effectively reduces the sparsity of KGs, thus increasing path search efficiency. Secondly, an iterative optimization of rule induction and fact inference is designed to incorporate global information from KGs to guide the RL agent exploration; this optimization iteratively improves overall training performance. We further evaluated the SparKGR model through different tasks on five real world datasets extracted from Freebase, Wikidata and NELL. The experimental results indicate that SparKGR outperforms state-of-the-art baseline models without losing interpretability.     

An improved event-triggered control method for fuzzy systems under membership functions online learning

This paper is concerned with the problem of adaptive event-triggered (AET) based optimal fuzzy controller design for nonlinear networked control systems (NCSs) characterized by Takagi–Sugeno (T–S) fuzzy models. An improved AET communication scheme with a memory adaptive rule is proposed to enhance the utilization of the state response vertex data. Different from the existing ET based results, the improved AET scheme can save more communication resources and acquire better system performance. The sufficient criteria of performance analysis and controller design are presented for the closed-loop control system subject to mismatched membership functions (MFs) and AET scheme. And then, a new MFs online learning algorithm on the basis of the gradient descent approach is employed to optimize the MFs of fuzzy controller and obtain optimal fuzzy controller for further improving system performance. Finally, two simulation examples are presented to verify the advantage and effectiveness of the provided controller design technique.