Recommendation Model Research Articles

SeSMR: Secure and Efficient Session-based Multimedia Recommendation in Edge Computing

Session-based multimedia recommendation in edge computing remains an important issue for boosting the utilization of services since service composition has increasingly attracted attention. Existing session-based recommendations (SBRs) model the session sequence with multilevel feature extraction in graph neural networks (GNNs). However, multilevel feature extraction in disentangled graph neural networks causes over-smoothing and privacy leakage. To address the aforementioned problems, Secure and Efficient Session-based Multimedia Recommendation (SeSMR) model is proposed. In the proposed SeSMR model, based on BGV homomorphic encryption, a ciphertext training submodel is proposed to address the privacy leakage, ensuring the security in session-based recommendation. Furthermore, based on the reinforcement of feature activation, a residual attention mechanism is proposed to mitigate over-smoothing while maintaining the independence of multiple features. Finally, based on location coding, a soft attention mechanism is proposed to improve the recommendation accuracy, by introducing the position difference information between items into intra-session and inter-session scenarios. Experiments demonstrate that both Recall and MRR metrics exhibit nearly 2%∼5% improvement.

Personalized process–type learning path recommendation based on process mining and deep knowledge tracing

Personalized learning path recommendation considers learning goals, learning abilities, and other personalized characteristics of learners to generate a suitable learning path. Existing approaches include global optimal and local iterative path recommendation, which recommend a sequence of learning objects. Consequently, the learner can only learn in the order specified by the learning path, which provides limited flexibility for the learner. In addition, existing studies cannot both present the complete path and handle changes in the learner's knowledge state while learning along the path. This study proposes a process-type learning path model and its recommendation approach, which presents a learning path in the form of a flowchart and dynamically recommends path branches according to the knowledge states of the learner during the learning process. Specifically, deep knowledge tracing is used to annotate the knowledge states of learners in historical logs, and process mining is used to generate a personalized process–type learning path that contains sequences, parallel relationships, and selection relationships between learning objects. In addition, the correlation between the knowledge state and the selection of different branches of a learning path in historical logs can be obtained via decision mining. Thus, a branch recommendation model is trained and used to recommend a path branch in a process-type path with the highest probability of mastering the target learning object of the learner based on the learner's knowledge state. The experimental results demonstrate that the learning effectiveness and efficiency of the proposed approach are better than those of the existing approaches.

Personalized movie recommendations based on probabilistic linguistic sentiment and integrated DEMATEL-TODIM methods

With the development of society, online reviews are increasingly becoming a crucial factor in decision-making. Especially for entertainment products such as movies, they are preferred for their affordability and high entertainment factor. Therefore, this paper proposes a movie recommendation model that considers user personalization using a probabilistic linguistic approach based on online reviews. Firstly, the method constructs a quantitative sentiment framework that transforms comments into a multi-granular probabilistic sentiment language. Secondly, we build the decision-making trial and evaluation laboratory (DEMATEL) method for probabilistic linguistic environments to explore interrelationships between product attributes, and improve the distance measure and score function to better integrate probabilistic linguistic information into DEMATEL weight calculations. Furthermore, to account for risk preferences, the model employs the extended TODIM (an acronym in Portuguese for interactive and multicriteria decision making) methodology to determine the ranking of alternatives. Finally, we design Douban movie ranking experiments to demonstrate the validity of the model. Compared with other methods, this paper incorporates the emotional tendency of movie attributes and user preference into the decision-making process leading to more reasonable results.

Establishment of a one-dimensional model for CO2 Pipeline rupture process and design recommendations

Pipelines serve as a critical means of transporting CO2 within Carbon Capture, Utilization, and Storage (CCUS) technology. Elevated transport pressures may lead to unforeseen pipeline ruptures. To provide practical recommendations for the design of CO2 pipelines and mitigate the risk of ruptures, this study establishes a one-dimensional Homogeneous Equilibrium Mixture (HEM) model. The model is currently used to predict the decompression behavior in existing pure CO2 rupture tests. Future work will address rupture tests and predictions involving impure CO2. The results indicate that due to differences in pressure reduction caused by decompression waves, the likelihood of long-range ruptures occurring in density CO2 pipelines is lower compared to supercritical CO2 pipelines. The crack propagation velocity and crack arrest pressure of pipelines designed according to ASME B31.3 are calculated using the High Strength Line Pipe Committee (HLP) model. Although the crack propagation velocity is lower than the decompression wave speed, the crack arrest pressure may fall below the rupture pressure, with the pressure at 1.6 m from the rupture point remaining above the crack arrest level for 1 ms post-rupture, a duration that is especially prolonged in larger-diameter pipelines. It is recommended to consider additional thickness beyond the calculated minimum thickness.

A product recommendation model based on online reviews: Improving PageRank algorithm considering attribute weights

Consumer reviews play a crucial role in evaluating products on online e-commerce platforms. Unlike numerical ratings, online reviews provide valuable information and sentiment. However, existing studies often overlook the unique interrelationships between products on e-commerce platforms, and fail to adequately capture the psychological behavior of consumers during online shopping. To address these gaps, this study presents a novel product recommendation model based on online reviews that evaluates products’ multi-attribute performances. The study first identifies the product attributes that are most important to consumers by analyzing review texts. Then, this study calculates the attribute performance scores of each product by considering consumer sentiment and the usefulness of online reviews. Next, it identifies competitors for the target product using a weighted Euclidean distance function and ranks all products employing an improved PageRank algorithm. Finally, to illustrate the validity of the proposed model, the study conducts a case study using a dataset of 41,352 online reviews obtained from Best Buy, and segments the data into three categories according to price. Comparative results with traditional MCDM models show that among the three categories, our results achieved a maximum improvement of 18.3% in the Spearman correlation coefficient.

Zero-Shot Content-Based Crossmodal Recommendation System

Information Recommendation (IR) systems are conventionally designed to operate within a single modality at a time, such as Text2Text or Image2Image. However, the concept of cross-modality aims to facilitate a versatile recommendation experience across different modalities, such as Text2Image. In recent years, significant strides have been made in developing neural recommender models that are multimodal and capable of generalizing across a broad spectrum of domains in a zero-shot manner, thanks to the robust representation capabilities of neural networks. These architectures enable the generation of embeddings for assets (i.e., content uploaded on a platform by users), presenting a concise representation of their semantics and allowing for comparisons through similarity ranking. In this paper, we present ZCCR, a Zero-shot Content-based Crossmodal Recommendation System that leverages knowledge from large-scale pretrained Vision-Language Models (VLMs) such as CLIP and ALBEF to redefine the recommendation task as a zero-shot retrieval task, eliminating the need for labeled data or prior knowledge about the recommended content. Furthermore, ZCCR performs crossmodal similarity search on an optimized index, such as FAISS, to improve the speed of recommendation. The goal is to recommend to the user assets that are similar to those they have previously uploaded, commonly referred to as their user profile. Within the user profile, we identify “areas of interest”—groups of assets associated with specific user interests, such as cooking, sports, or cars. To identify these areas of interest and construct the search query for the retrieval operation, ZCCR employs an innovative use of Agglomerative Clustering. This technique groups user past assets by similarity without requiring prior knowledge of the number of clusters. Once the areas of interest, or clusters, are identified, the centroid is utilized as the search seed to find similar assets. Experimental results demonstrate the efficiency of the selected components in terms of search time and retrieval performance on modified MSCOCO and FLICKR30k datasets tailored for the recommendation task. Furthermore, ZCCR outperforms both a baseline tagging system (BT) and a more advanced tag system which utilizes a Large Language Model (LLM) to extract embeddings from tags. The results show that, even compared with the latter, embeddings directly extracted from raw assets yield superior outcomes compared to relying on intermediate tags generated by other tools. The code implementation to reproduce all experiments and results shown in this paper is provided at the following link: ZCCR-experiments.

Efficient and adaptive secure cross-domain recommendations

In recommendation systems, cross-domain recommendation (CDR) has emerged as an effective method for solving the problem of data sparsity and cold start. However, challenges such as data heterogeneity, data security, and recommendation accuracy are faced by existing CDR methods. An efficient and adaptive secure cross-domain recommendation (EAAS-CDR) model is proposed. First, the Automatic Encoder–Decoder (AutoED) model is proposed to address variations in user–item interaction quantity and item characteristics across various domains. The AutoED model enables adaptive and secure cross-domain recommendations by modeling user preferences across various domains, unifying vector dimensions, and avoiding the release of the rating matrix. An efficient cross-domain privacy protection method is proposed to resist data leakage. To ensure feature invariance for knowledge transfer, the inherent low perturbation characteristics of the Johnson–Lindenstrauss Transform mechanism are leveraged and combined with the output of our encoder. Finally, domain relevance measurement metrics and a domain fusion algorithm are proposed to use the source domain knowledge in the target domain and improve the recommendation accuracy of the model. Experimental results demonstrate that our approach is not only more efficient but also more secure than existing CDR approaches.

PRMNBR: Personalized Recommendation Model for Next Basket Recommendation Using User’s Long-Term Preference, Short-Term Preference, and Repetition Behaviour

PRMNBR: Personalized Recommendation Model for Next Basket Recommendation Using User’s Long-Term Preference, Short-Term Preference, and Repetition Behaviour

A survey on large language models for recommendation

A survey on large language models for recommendation

Challenging unverified assumptions in causal claims: Do gas stoves increase risk of pediatric asthma?

The use of unverified models for risk estimates and policy recommendations can be highly misleading, as their predictions may not reflect real-world health impacts. For example, a recent article states that NO2 from gas stoves “likely causes ∼50,000 cases of current pediatric asthma from long-term NO2 exposure alone” annually in the United States. This explicitly causal claim, which is contrary to several methodology and review articles published in this journal, among others, reflects both (a) An unverified modeling assumption that pediatric asthma burden is approximately proportional to NO2; and (b) An unverified causal assumption that the assumed proportionality between exposure and response is causal. The article is devoid of any causal analysis showing that these assumptions are likely to be true. It does not show that reducing NO2 exposure from gas stoves would reduce pediatric asthma risk. Its key references report no significant associations – let alone causation – between NO2 and pediatric asthma. Thus, the underlying data suggests that the number of pediatric asthma cases caused by gas stoves in the United States is indistinguishable from zero. This highlights the need to rigorously validate modeling assumptions and causal claims in public health risk assessments to ensure scientifically sound foundations for policy decisions.

A graph neural network-based teammate recommendation model for knowledge-intensive crowdsourcing

Current research on teammate recommendation decision in various fields is difficult to apply to knowledge-intensive crowdsourcing, which needs to make a trade-off between solver willingness and team performance. To weigh the bilateral benefits between the solver and the seeker, this paper proposes a teammate recommendation model based on multi-criteria decision-making of solver willingness and team performance. The prediction of willingness to team up is driven by explicit competitiveness, teammate familiarity, and social capital. By adapting the graph-based deep learning model, we transform the solver performance prediction into a bipartite network link prediction problem. We jointly compute the solver willingness and team performance to make the final recommendation decisions. The experimental results based on Kaggle data prove the effectiveness of our method. The findings provide managerial implications and a practical reference for the design and operation of teammate recommender systems.

The crowd in MOOCs: a study of learning patterns at scale

ABSTRACT The increasing availability of learning activity data in Massive Open Online Courses (MOOCs) enables us to conduct a large-scale analysis of learners' learning behavior. In this paper, we analyze a dataset of 351 million learning activities from 0.8 million unique learners enrolled in over 1.6 thousand courses within two years. Specifically, we mine and identify the learning patterns of the crowd from both temporal and course enrollment perspectives leveraging mutual information theory and sequential pattern mining methods. From the temporal perspective, we find that the time intervals between consecutive learning activities of learners exhibit a mix of power-law and periodic cosine function distribution. By qualifying the relationship between course pairs, we observe that the most frequently co-enrolled courses usually fall in the same category or the same university. We demonstrate these findings can facilitate manifold applications including recommendation tasks on courses. A simple recommendation model utilizing the course enrollment patterns is competitive to the baselines with 200× faster training time.

A novel approach to enterprise technical collaboration: Recommending R&D partners through technological similarity and complementarity

Choosing the right partner is a key factor in the success of enterprise R&D cooperation, directly affecting innovation outcomes and market competitiveness. Technical similarity provides a common language and foundational understanding between enterprises, while technical complementarity offers opportunities for knowledge exchange and innovation. However, no previous research has effectively integrated these two features within a collaborator recommendation framework. This study aims to explore a method that combines technological similarity and complementarity for collaborator recommendations. We introduced the Technological Similarity and Complementarity Enhanced Collaborator Recommendation (TSCE-CR) model, which constructs a heterogeneous corporate collaboration network and designs a tailored loss function. This model effectively integrates features of technological similarity and complementarity, enabling the neural network to capture and elucidate the nonlinear and multidimensional relationships in corporate collaborations. Experimental validation on patent data in the field of artificial intelligence demonstrated that our TSCE-CR model excels in identifying potential collaborators, effectively confirming the critical role of technological complementarity in R&D collaboration. This research provides a flexible framework for future studies on collaborator recommendations and offers reliable decision-making support for enterprises in selecting R&D partners.

Artificial Intelligence and Blockchain Technology Drive Leadership Decision-Making Research Group Recommendation Algorithm

Artificial intelligence and blockchain can improve the effectiveness of leadership decision-making in two dimensions. Artificial intelligence technology can improve the scientificity of leadership decision-making, and blockchain technology can guarantee the democracy of leadership decision-making. Society pushes everyone to be gregarious. Group recommendation is thus one of the research focuses in recent years. Prior group recommendation algorithms fail to consider either the influence of group structure on computing scale or the impressions users of higher weights leave on other group members. To address the aforementioned challenges, this paper proposes a group recommendation model based on members’ influence and leader impact. In this paper, a model has been proposed to compute members’ influence on each other based on interactions and presence. The decisions of leaders identified by the proposed model are the basis for further group recommendation, which yields satisfactory recommendations for most group members as leaders’ judgments are more professional. Experimental results on real-world datasets demonstrate better accuracy of the proposed method compared to those of the mainstream group recommendation algorithms.

Distribution-aware hybrid noise augmentation in graph contrastive learning for recommendation

The recommender systems are one of the most effective big data tools for solving the information overload problem, but data sparsity greatly affects its performance. However, most of the existing graph-based contrastive learning methods perturb the original graph structure for data augmentation, which may lead to semantic loss and subsequently degrade the recommendation performance. Meanwhile, studies have shown that improving the uniformity of data distribution is more effective than data augmentation using graph perturbation. In this paper, to enhance the uniformity of data distribution, we propose Distribution-Aware Hybrid Noise augmentation in graph contrastive learning for Recommendation (DAHNRec). Specifically, we design distribution-aware hybrid noise augmentation (DAHN) method to generate noise suitable for the users and items, instead of simply utilizing random noise following uniform distributions. By applying DAHN, the model can learn the appropriate noise distribution during the training process to enhance the uniformity of the embedding space. Additionally, we propose Balanced Bayesian Personalized Ranking (BBPR) as the loss function for recommendation tasks to improve difference characterizing between positive and negative samples. This improves the ranking tasks’ performance. Furthermore, we conduct extensive experiments based on public benchmark datasets. The results indicate that DAHNRec outperforms several existing methods. Furthermore, DAHNRec is a general recommendation model based on user–item interactions, which can be used to improve the recommendation performance in various real-world scenarios, such as e-commerce platforms and social media.

Automated recommendation model using ordinal probit regression factorization machines

Automated recommendation model using ordinal probit regression factorization machines

Ensemble-based predictive model for crop recommendation

Ensemble-based predictive model for crop recommendation

Cluster-based Graph Collaborative Filtering

Graph Convolution Networks (GCNs) have significantly succeeded in learning user and item representations for recommendation systems. The core of their efficacy is the ability to explicitly exploit the collaborative signals from both the first- and high-order neighboring nodes. However, most existing GCN-based methods overlook the multiple interests of users while performing high-order graph convolution. Thus, the noisy information from unreliable neighbor nodes ( \(e.g.\) , users with dissimilar interests) negatively impacts the representation learning of the target node. Additionally, conducting graph convolution operations without differentiating high-order neighbors suffers the over-smoothing issue when stacking more layers, resulting in performance degradation. In this paper, we aim to capture more valuable information from high-order neighboring nodes while avoiding noise for better representation learning of the target node. To achieve this goal, we propose a novel GCN-based recommendation model, termed Cluster-based Graph Collaborative Filtering (ClusterGCF). This model performs high-order graph convolution on cluster-specific graphs, which are constructed by capturing the multiple interests of users and identifying the common interests among them. Specifically, we design an unsupervised and optimizable soft node clustering approach to classify user and item nodes into multiple clusters. Based on the soft node clustering results and the topology of the user-item interaction graph, we assign the nodes with probabilities for different clusters to construct the cluster-specific graphs. To evaluate the effectiveness of ClusterGCF, we conducted extensive experiments on four publicly available datasets. Experimental results demonstrate that our model can significantly improve recommendation performance.

Optimizing co-application of starter ammoniacal-N and K-containing fertilizers for barley in the seed-furrow

It is unclear how seed-placed starter ammoniacal-N-and-K fertilizers influence barley (Hordeum vulgare L.) seed-crop emergence, and its early establishment, due to salt injury and potential ammonia production and toxicity within the soil profile under different placement rates. Keeping in this view, the study was undertaken to experiment the effects of different fertilizer types, and rates using three different soil types characterized by fine (G1), medium (G2), and coarse (G3) textures when simultaneously placed with barley seed and establish safe-starter fertilizer recommendation models. The fertilizers were applied at varying rates from 0 to 60 kg N ha−1 and 0 to 74 kg K ha−1 alongside barley seeding. Maximum predicted safe starter ammoniacal-N-fertilizer rates were as follows: urea = 15, 35 and 26 kg N ha−1 is safe for G1, G2 and G3 soils, respectively. DAP (Diammonium phosphate), 52 kg N ha−1 is safe for fine textures whilst 12 and 37 kg N ha−1 for medium and coarse textures. MAP (Monoammonium phosphate) is safely placed at 55, 11 and ≥60 kg N ha−1 for G1 – G3 soils. For MOP (Muriate of Potash) and SOP (Sulfate of Potash), 67 and ≥74 kg K ha−1 were ideal for improved barley emergence across all the soil types. Visual appearances including delayed seed emergence, leaf chlorosis, and stubby coleoptiles were observed at varying seed-placed N and K rates. Intricate relationships among various emergence parameters, root architecture, aboveground and root biomass, chlorophyll content, all showing significant interrelationships. Data indicate that cubic model could be used for predicting seed-safe rates for multiple fertilizer types.

An improved cross-domain sequential recommendation model based on intra-domain and inter-domain contrastive learning

Cross-domain recommendation aims to integrate data from multiple domains and introduce information from source domains, thereby achieving good recommendations on the target domain. Recently, contrastive learning has been introduced into the cross-domain recommendations and has obtained some better results. However, most cross-domain recommendation algorithms based on contrastive learning suffer from the bias problem. In addition, the correlation between the user’s single-domain and cross-domain preferences is not considered. To address these problems, a new recommendation model is proposed for cross-domain scenarios based on intra-domain and inter-domain contrastive learning, which aims to obtain unbiased user preferences in cross-domain scenarios and improve the recommendation performance of both domains. Firstly, a network enhancement module is proposed to capture users’ complete preference by applying a graphical convolution and attentional aggregator. This module can reduce the limitations of only considering user preferences in a single domain. Then, a cross-domain infomax objective with noise contrast is presented to ensure that users’ single-domain and cross-domain preferences are correlated closely in sequential interactions. Finally, a joint training strategy is designed to improve the recommendation performances of two domains, which can achieve unbiased cross-domain recommendation results. At last, extensive experiments are conducted on two real-world cross-domain scenarios. The experimental results show that the proposed model in this paper achieves the best recommendation results in comparison with existing models.