Cross-domain Recommendation (CR) has been extensively studied in recent years to alleviate the data sparsity issue in recommender systems by utilizing different domain information. In this work, we focus on the more general Non-overlapping Cross-domain Sequential Recommendation (NCSR) scenario. Non-overlapping Cross-domain Sequential Recommendation (NCSR) is challenging because there are no overlapped entities (e.g., users and items) between domains, and there is only users’ implicit feedback and no content information. Previous Cross-domain Recommendation (CR) methods cannot solve NCSR well, since (1) they either need extra content to align domains or need explicit domain alignment constraints to reduce the domain discrepancy from domain-invariant features, (2) they pay more attention to users’ explicit feedback (i.e., users’ rating data) and cannot well capture their sequential interaction patterns, (3) they usually do a single-target cross-domain recommendation task and seldom investigate the dual-target ones. Considering the above challenges, we propose Prompt Learning-based Cross-domain Recommender (PLCR), an automated prompting-based recommendation framework for the NCSR task. Specifically, to address the challenge (1), Prompt Learning-based Cross-domain Recommender (PLCR) resorts to learning domain-invariant and domain-specific representations via its prompt learning component, where the domain alignment constraint is discarded. For challenges (2) and (3), PLCR introduces a pre-trained sequence encoder to learn users’ sequential interaction patterns, and conducts a dual-learning target with a separation constraint to enhance recommendations in both domains. Our empirical study on two sub-collections of Amazon demonstrates the advance of PLCR compared with some related SOTA methods.

The task of knowledge tracing aims to monitor students’ knowledge states through their historical answer records and predict their future performance in answering questions. In recent years, knowledge tracing models based on deep learning have exhibited significantly higher accuracy in prediction compared to traditional knowledge tracing models. However, existing models often fail to fully leverage the impact of difficulty factors on students’ knowledge states. In this paper, to better exploit the difficulty factors for enhancing the discriminative power among different questions, we propose a novel Difficulty-Fusion Knowledge Tracing(DFAKT) model. Specifically, we design a DEI(Difficult-Enhance Interaction) method for extracting difficulty features by integrating skill difficulty and question-specific difficulty information. Subsequently, we develop a Difficulty-Fusion module designed to integrate the student interaction information, which includes features such as answers and knowledge points, with the difficulty feature. Finally, recognizing that different exercises have varying impacts on students’ knowledge states, we incorporate an attention-based GRU module in the model to dynamically aggregate the knowledge states from previous time steps. This allows the model to focus more on questions with a greater impact on knowledge states, thereby improving prediction accuracy. Experimental results demonstrate that the proposed model outperforms existing models in predictive performance on large-scale real-world datasets, and the effectiveness of each module is validated.

Academic performance prediction aims to leverage student-related information to predict their future academic outcomes, which is beneficial to numerous educational applications, such as personalized teaching and academic early warning. In this article, we reveal the students' behavior trajectories by mining campus smartcard records, and capture the characteristics inherent in trajectories for academic performance prediction. Particularly, we carefully design a tri-branch convolutional neural network (CNN) architecture, which is equipped with rowwise, columnwise, and depthwise convolutions and attention operations, to effectively capture the persistence, regularity, and temporal distribution of student behavior in an end-to-end manner, respectively. However, different from existing works mainly targeting at improving the prediction performance for the whole students, we propose to cast academic performance prediction as a top-k ranking problem, and introduce a top-k focused loss to ensure the accuracy of identifying academically at-risk students. Extensive experiments were carried out on a large-scale real-world dataset, and we show that our approach substantially outperforms recently proposed methods for academic performance prediction. For the sake of reproducibility, our codes have been released at https://github.com/ZongJ1111/Academic-Performance-Prediction.

Shared-account Cross-domain Sequential Recommendation (SCSR) is an emerging yet challenging task that simultaneously considers the shared-account and cross-domain characteristics in the sequential recommendation. Existing works on SCSR are mainly based on Recurrent Neural Network (RNN) and Graph Neural Network (GNN) but they ignore the fact that although multiple users share a single account, it is mainly occupied by one user at a time. This observation motivates us to learn a more accurate user-specific account representation by attentively focusing on its recent behaviors. Furthermore, though existing works endow lower weights to irrelevant interactions, they may still dilute the domain information and impede the cross-domain recommendation. To address the above issues, we propose a reinforcement learning-based solution, namely RL-ISN, which consists of a basic cross-domain recommender and a reinforcement learning-based domain filter. Specifically, to model the account representation in the shared-account scenario, the basic recommender first clusters users’ mixed behaviors as latent users, and then leverages an attention model over them to conduct user identification. To reduce the impact of irrelevant domain information, we formulate the domain filter as a hierarchical reinforcement learning task, where a high-level task is utilized to decide whether to revise the whole transferred sequence or not, and if it does, a low-level task is further performed to determine whether to remove each interaction within it or not. To evaluate the performance of our solution, we conduct extensive experiments on two real-world datasets, and the experimental results demonstrate the superiority of our RL-ISN method compared with the state-of-the-art recommendation methods.

Deep cross-modal hashing retrieval models inherit the vulnerability of deep neural networks. They are vulnerable to adversarial attacks, especially for the form of subtle perturbations to the inputs. Although many adversarial attack methods have been proposed to handle the robustness of hashing retrieval models, they still suffer from two problems: (1) Most of them are based on the white-box settings, which is usually unrealistic in practical application. (2) Iterative optimization for the generation of adversarial examples in them results in heavy computation. To address these problems, we propose an Efficient Query-based Black-Box Attack (EQB 2 A) against deep cross-modal hashing retrieval, which can efficiently generate adversarial examples for the black-box attack. Specifically, by sending a few query requests to the attacked retrieval system, the cross-modal retrieval model stealing is performed based on the neighbor relationship between the retrieved results and the query, thus obtaining the knockoffs to substitute the attacked system. A multi-modal knockoffs-driven adversarial generation is proposed to achieve efficient adversarial example generation. While the entire network training converges, EQB 2 A can efficiently generate adversarial examples by forward-propagation with only given benign images. Experiments show that EQB 2 A achieves superior attacking performance under the black-box setting.

Knowledge Tracing (KT) is an important part of intelligent online education and is the key to personalized guidance for students' learning process. It aims at dynamically estimate students' knowledge status based on history answer records and predict whether they will answer the next question correctly. Predicting students' knowledge is a difficult task because student learning is a dynamic process and students' knowledge status is constantly changing. However, existing approaches often ignore the fact of students' stage development of learning ability and do not take into account the impact of short-term knowledge states on the next moment. Existing models do not emphasize the importance of students' long-term knowledge states and short-term knowledge states. In this paper, we propose a new Long- and Short-term Attention network Knowledge Tracing model (LSAKT). Specifically, we divided the sequences into subsequences based on time stamps, with the first attention layer learning the student's long-term knowledge state based on the interactional performance with the problem, while the other attention layer learns the student's short-term knowledge state based on the last sequence. Finally, we integrate the long-term knowledge state and the short-term knowledge state to form the student's final knowledge state. We evaluated the proposed model on four publicly available datasets, and the experimental results showed that the LSAKT approach achieved a fantastic result.

With the gradual expansion scale of mobile intelligent terminals and the popularity of social networks, information exchange among mobile users shows the social characteristics. This article takes the same interest information among users to define the social relationship metrics such as contact frequency, contact strength and contact regularity etc. A ranking information forwarding algorithm with social characteristic in mobile Delay Tolerant network is presented and the simulation results show that this algorithm performs better in delivery rate, lower delay and buffer memory.