Item Recommendation Research Articles

Sequential ensemble learning for next item recommendation

Predicting the next item that users may engage in is a key task of recommender systems, and many methods have been proposed to deal with this task from different aspects. Theoretically, the proper ensemble of multiple different methods (a.k.a. base models) can make more accurate and stable recommendations. However, most of the existing ensemble methods rely on static aggregation strategies, which fail to capture base models’ dynamic predictive abilities for each user over time. In addition, most of the existing diversity measures used in regression or classification ensemble methods rely on a distance metric of base models’ outputs, which makes it intractably apply for next-item recommendation whose base models output sequential ranking lists.To solve the above problems, we propose a Sequential Ensemble Method, named SEM, to aggregate different base models for next-item recommendation. We assume users’ concentration and base models’ expertise can be inferred from users’ sequential behaviors and base models’ prediction results. Therefore, we propose to explicitly model base models’ dynamic predictive abilities on different users over time based on users’ concentration and base models’ expertise. In addition, we propose a new diversity measure for sequential ranking ensemble, which can perform diversity-based learning over time for better performance of next-item recommendation. Extensive experiments on six real-world data sets show that our method consistently outperforms state-of-the-art methods.

Cold-start item recommendation for representation learning based on heterogeneous information networks with fusion side information

In recommendation systems, the cold-start problem refers to the situation that users or items have too few interaction events with the system to be recommended, such as clicks or ratings. The cold-start problem leads to the loss of new users and difficulty in recommending new items and weakens the performance of recommendations. In this paper, we focus on alleviating the cold-start problem of items. Previous work has used the structure of item homogeneous graphs to establish connections between items and generate item embeddings. However, the homogeneous graph of a single node cannot fully explore the potential structural features between users and goods. Due to the flexibility of heterogeneous data, recommendation systems can also employ the Heterogeneous Information Network(HIN) to characterize complex, heterogeneous data. In this paper, a novel Heterogeneous Graph Embedding Learning based Skip-Gram (HSG) model is proposed to mitigate the cold-start problem. First, we introduce three different types of subnetworks and combine them to construct heterogeneous graphs. Then meaningful sequences of nodes are generated via meta-path-based random walk for network embedding. Finally, a Skip-Gram model incorporating side information is used to learn item representations. The learned node embeddings are used as pre-training for the Matrix Factorization (MF) model. The HSG model is proven to be effective by extensive experiments on four publicly available datasets.

Symmetric Bayesian Personalized Ranking With Softmax Weight

Preference learning, especially pairwise preference learning, is an efficient method for modeling implicit feedback in item recommendation. However, it is insufficient and not always valid to work for a basic pairwise preference model, which assumes that users prefer interacted (i.e., bought or viewed) items to un-interacted (i.e., not bought or not viewed) items. Recently, the state-of-the-art approaches have emerged as two separate but powerful methods, namely, pairwise preferences over item-sets and asymmetric pairwise preference models, respectively, to address limitations of the basic models. In spite of the success achieved by these methods, the assumption that the horizontal pairwise preference is with respect to two items does not always hold in asymmetric pairwise preference. Hence, it is appealing to integrate them into a uniform approach. In this article, we propose a novel symmetric pairwise preference assumption. We use a weighted average through a softmax function and define the overall preferences that can better discover users’ preference patterns. With the new assumption and the weighted average method, we propose a novel recommendation algorithm to improve the recommendation quality. Extensive empirical studies show that our new algorithms can significantly outperform several state-of-the-art and baseline methods over a number of public datasets.

Customer-based Recommendation Model for Next Merchant Recommendation

In the recommendation system of the credit card company, it is necessary to understand the customer patterns to predict a customer’s next merchant based on their histories. The data we want to model is much more complex and there are various patterns that customers choose. In such a situation, it is necessary to use an effective model that not only shows the relevance of the merchants, but also the relevance of the customers relative to these merchants. The proposed model aims to predict the next merchant for the customer. To improve prediction performance, we propose a novel model, called Customer-based Recommendation Model (CRM), to produce a more efficient representation of customers. For the next merchant recommendation system, we use a synthetic credit card usage dataset, BC’17. To demonstrate the applicability of the proposed model, we also apply it to the next item recommendation with another real-world transaction dataset, IJCAI’16.

Dynamic Multi-Behavior Sequence Modeling for Next Item Recommendation

Sequential Recommender Systems (SRSs) aim to predict the next item that users will consume, by modeling the user interests within their item sequences. While most existing SRSs focus on a single type of user behavior, only a few pay attention to multi-behavior sequences, although they are very common in real-world scenarios. It is challenging to effectively capture the user interests within multi-behavior sequences, because the information about user interests is entangled throughout the sequences in complex relationships. To this end, we first address the characteristics of multi-behavior sequences that should be considered in SRSs, and then propose novel methods for Dynamic Multi-behavior Sequence modeling named DyMuS, which is a light version, and DyMuS+, which is an improved version, considering the characteristics. DyMuS first encodes each behavior sequence independently, and then combines the encoded sequences using dynamic routing, which dynamically integrates information required in the final result from among many candidates, based on correlations between the sequences. DyMuS+, furthermore, applies the dynamic routing even to encoding each behavior sequence to further capture the correlations at item-level. Moreover, we release a new, large and up-to-date dataset for multi-behavior recommendation. Our experiments on DyMuS and DyMuS+ show their superiority and the significance of capturing the characteristics of multi-behavior sequences.

ICU Nursing Team Mental Health in The Face of the COVID-19 Pandemic: An Integrative Review

Introduction: During the COVID-19 pandemic, health professionals were at an increased risk of developing psychological conditions, especially in Intensive Care Units. A gap is verified in the knowledge about ICU Nursing teams’ mental health in the care provided to patients affected by COVID-19. Objective: To review the literature to identify the mental health conditions of ICU Nursing professionals during the COVID-19 pandemic. Materials and method: This is an integrative literature review with data collection from January to April 2022 in the Embase, Cumulative Index to Nursing and Allied Health Literature, PubMed, Scopus Content Overview and Web of Science databases. The following descriptors were listed: “Intensive Care Units,” “COVID-19,” “Mental Health,” and “Nurse.” The materials included were studies produced since 2020, available in full in the Spanish, English and Portuguese languages, and which answered the research question. Search and selection of the articles were the duty of two duly trained researchers, who worked independently through peer discussion. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses recommendations were followed. Results: From a sample of 16 articles, it was observed that the critical care of COVID-19 patients affected Nursing professionals’ mental health with an emphasis on depression, anxiety, stress and fear. This was associated with a lack of evidence-based training, scarcity of human and physical resources, work overload and insufficient work experience. Conclusions: ICU Nursing professionals are mentally shaken; therefore, it is imperative to implement policies, strategies and methods to improve the psychological conditions of these professionals, to ensure the quality of services.

M2GCF: A multi-mixing strategy for graph neural network based collaborative filtering

Graph Neural Networks (GNNs) have been successfully used to learn user and item representations for Collaborative Filtering (CF) based recommendations (GNN-CF). Besides the main recommendation task in a GNN-CF model, contrastive learning is taken as an auxiliary task to learn better representations. Both the main task and the auxiliary task face the noise problem and the distilling hard negative problem. However, existing GNN-CF models only focus on one of them and ignore the other. Aiming to solve the two problems in a unified framework, we propose a Multi-Mixing strategy for GNN-based CF (M2GCF). In the main task, M2GCF perturbs embeddings of users, items and negative items with sample-noise by a mixing strategy. In the auxiliary task, M2GCF utilizes a contrastive learning mechanism with a two-step mixing strategy to construct hard negatives. Extensive experiments on three benchmark datasets demonstrate the effectiveness of the proposed model. Further experimental analysis shows that M2GCF is robust against interaction noise and is accurate for long-tail item recommendations.

The incidence and prevalence of medical device-related pressure injuries in pediatric patients: Systematic review and meta-analysis

ProblemTo determine the pooled incidence and prevalence rate of medical device-related pressure injuries(MDRPIs) using the Braden QD scale, medical devices that frequently cause MDRPIs, and anatomical locations that are vulnerable to them. Eligibility criteriaUsing the Braden QD scale, being published in English between 01/01/2018-and 01/03/2023. MethodsThis study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis recommendations and registered in the International Prospective Register of Systematic Reviews database (No: CRD42021276501). SampleA total of 7 studies with 25,742 pediatric patients were included. ResultsThe pooled prevalence and cumulative incidence of hospital-acquired pressure injuries(HAPIs) were 7.8% (95% CI: 5.2–11.4%) and 3.9% (95% CI: 0.5–24.6%) respectively, and the incidence rate was 8.2/1000 person-days (95% CI: 2.4–14.2/1000 person-days). The pooled prevalence and cumulative incidence of MDRPIs were 7% (95% CI: 5.5–8.8%) and 5% (95% CI: 3.2–7.8%) respectively, and the incidence rate was 6.7/1000 person-days (95% CI, 0.11–13.4/1000 person-days). The most affected anatomical locations were the face (29.1%), ankle/foot (20.1%), and head (15.7%). Medical devices that frequently caused MDRPIs were external monitoring devices (24.5%), respiratory devices (22.8%), and supportive/securing devices (14.9%). ConclusionsAccording to the current systematic review and meta-analyses, the incidence and prevalence of HAPIs and MDRPIs are moderate to high. ImplicationsThe findings suggested that healthcare providers should pay more attention to reducing HAPIs and MDRPIs and future studies should be conducted to understand their characteristics and risk factors.

Neighbor importance-aware graph collaborative filtering for item recommendation

The emerging topic of Graph Neural Networks (GNN) has attracted increasing attention and achieved state-of-the-art (SOTA) performance in many recommendation problems, due to its strong ability in node representation with exploring high-order information. To learn a node’s representation, previous methods usually linearly combine the embeddings of node features, amusing the equal importance of neighbors. However, due to the intrinsic differences (i.e., degree, create time) over neighbors, we argue that these differences carry important signals for node representation. Ignoring them will lead to a suboptimal in node representation and thus weaken the effectiveness of the follow-up graph-based operations. To address it, we propose BIG-SAGE@ for item recommendation with rating prediction task, which is a neighbor importance-aware graph neural network. Specifically, its main idea is twofold: 1) A rating confidence-based neighborhood sampling method is introduced, making the sampling process biased to those more valuable nodes. 2) An attention network is integrated to achieve the rating prediction task, by flexibly incorporating information from user and item embedding features. Finally, we verified the effectiveness of the proposed model on six public data sets. Extensive experimental results demonstrate the superior performance of BIG-SAGE@ in the rating prediction and TopN ranking tasks, compared to the SOTA methods.

On the problem of recommendation for sensitive users and influential items: Simultaneously maintaining interest and diversity

Recommender systems, in real-world circumstances, tend to limit user exposure to certain topics and to overexpose them to others to maximize performance. However, repeated exposure to biased content could lead to the so-called echo chamber phenomenon: especially in social network environments, people encounter only information that reflects their previous beliefs and opinions, reinforcing them. This phenomenon could have worrying consequences for society, including the spread of aggressive, unhealthy, or risky behaviors. Some persons can be more affected than others by echo-chambers. We define as sensitive the users whose behavior could be influenced by the over- or under-exposure to certain items due to the echo-chamber effect, and as influential the items that could influence the behavior of such users. In this paper, we address the problem of recommending influential items to sensitive users. We formalize the problem and propose three techniques that can be used to diversify the distributions of influential items in order to positively affect sensitive users’ behavior. Recommendations that meet this diversity criterion could potentially avoid dangerous societal consequences and simultaneously promote healthier lifestyles. We tested the proposed techniques in a real-world dataset by considering two different case studies that involved potentially aggressive and potentially depressed users. All techniques have been proven to be effective and allow high performance to be maintained while diversifying recommendations.

Food-dependent exercise-induced wheals/angioedema, anaphylaxis, or both: A systematic review of phenotypes.

Food-dependent exercise-induced allergic reactions can manifest with wheals, angioedema, and anaphylaxis, alone or in combination. To systematically review the clinical manifestation, culprit foods and exercise, augmenting factors, comorbidities, and treatment options of each phenotype. Using predefined search terms, we assessed and analyzed the relevant literature until June 2021. Preferred Reporting Items for Systematic Reviews and Meta-analysis recommendations were applied to this systematic review. A total of 231 studies with 722 patients were included. The most common phenotype was anaphylaxis with wheals, angioedema, or both, reported in 80% of patients. This was associated with a higher number of anaphylactic episodes, augmenting factors, and use of on-demand antihistamine compared to the least common phenotype, anaphylaxis without wheals or angioedema, reported in 4% of patients. Anaphylaxis with wheals/angioedema was also associated with distinct characteristics compared to stand-alone wheals, angioedema, or both, in 17% of patients: Patients with anaphylaxis were older at the time of disease onset, less often had history of atopy, showed more positive results in response to food and exercise provocation tests, had a more restricted spectrum of culprit foods, and more often used on-demand epinephrine. The three phenotypes of allergic reactions to food and exercise differ in their clinical characteristics, triggers, and response to treatment. Knowledge of these differences may help with patient education and counseling, as well as disease management.

FaiRIR: Mitigating Exposure Bias From Related Item Recommendations in Two-Sided Platforms

Related item recommendations (RIRs) are ubiquitous in most online platforms today, including e-commerce and content streaming sites. These recommendations not only help users compare items related to a given item but also play a major role in bringing traffic to individual items, thus deciding the exposure that different items receive. With a growing number of people depending on such platforms to earn their livelihood, it is important to understand whether different items are receiving their desired exposure. To this end, our experiments on multiple real-world RIR datasets reveal that the existing RIR algorithms often result in very skewed exposure distribution of items, and the quality of items is not a plausible explanation for such skew in exposure. To mitigate this exposure bias, we introduce multiple flexible interventions [fair related item recommendation (FaiRIR)] in the RIR pipeline. We instantiate these mechanisms with two well-known algorithms for constructing RIRs--rating singular value decomposition (SVD) and item2vec--and show on real-world data that our mechanisms allow for a fine-grained control on exposure distribution, often at a small or no cost in terms of recommendation quality, measured in terms of relatedness and user satisfaction.

Diagnostic accuracy of circulating exosomal circRNAs in malignances: A meta-analysis and systematic review.

Pathogenesis of malignant tumors are often accompanied by aberrant expression of circular RNAs (circRNAs), indicating the potential diagnostic value of circRNAs in tumors. CircRNAs have been found to be enriched, stable and ubiquitous in serum and plasma exosomes. The study aims at evaluating the diagnostic performance of circulating (plasma and serum) exosomal circRNA in different types of cancer by synthesis of published data. A comprehensive literature search was conducted in PubMed, Embase, Medline and the Web of Science databases to identify potentially eligible studies published before April 2021. We conducted the meta-analysis following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses recommendations. Eleven articles comprising 21 studies were included, and a total of 1609 cases and 1498 controls were evaluated. Six types of cancer were involved in these studies, including lung cancer, hepatocellular carcinoma, colorectal cancer, gastric cancer, multiple myeloma and osteosarcoma. The pooled sensitivity and specificity were 0.72 (95% confidence interval [CI], 0.62-0.81) and 0.83 (95% CI, 0.78-0.88), respectively. Summary receiver operating characteristic curve was constructed and the pooled value of area under curve was 0.86 (95% CI, 0.83-0.89), indicating a favorable diagnostic efficacy of circulating exosomal circRNAs in malignancies. In conclusion, our study evaluated the diagnostic power of circulating exosomal circRNAs in 6 types of cancer by synthesis of published data comprising 21 studies from eleven articles. The pooled analysis provided the evidence supporting circulating exosomal circRNAs as a promising noninvasive diagnostic biomarkers for malignancies.

SMONE: A Session-based Recommendation Model Based on Neighbor Sessions with Similar Probabilistic Intentions

A session-based recommendation system (SRS) tries to predict the next possible choice of anonymous users. In recent years, graph neural network (GNN) models have been successfully applied to SRSs and have achieved great success. Using GNN models in SRSs, each session graph is processed successively to obtain the embedding of the node (i.e, each action on an item), which is then imported into the prediction module to generate recommendation results. However, solely depending on the session graph to obtain the node embeddings is not sufficient because each session only involves a few items. Therefore, neighbor sessions have been used to extend the session graph to learn more informative node representations. In this paper, we introduce a S ession-based recommendation MO del based on N eighbor sessions with similar probabilistic int E ntions(SMONE). SMONE models the intentions behind sessions in a probabilistic way and retrieves the neighbor sessions with similar intentions. After the neighbor sessions are found, the target session and its neighbor sessions are modeled as a hypyergraph to learn the contextualized embeddings, which are combined with item embeddings through GNN to produce the final item recommendations. Experiments on real-world datasets prove the effectiveness and superiority of SMONE.

Event-based incremental recommendation via factors mixed Hawkes process

Incremental recommendation systems have garnered significant research interest since they ideally adapt to users' ongoing events (such as clicking, browsing, and reviewing) and recommend items without retaining the model. Many methods have tracked the event generation sequences for incremental recommendation. However, most existing models treat the event as a static snapshot or a black box, ignoring the underlying factors that may trigger the event generation. Underlying such inner factors can help RS reasonably and foreseeingly evaluate the potential items for the user next time. Along this vein, we propose the Factors Mixed Hawkes Process (FMHP) for event-based incremental recommendations. First, we extend each event to a notion of factor-driven event sequence. Next, we consider three factors that may influence the occurrence of an event: intrinsic intensity, external intensity, and historical intensity. An intrinsic intensity function, multi-type temporal attention, and a hybrid time decay function are incorporated in FMHP to evaluate the intrinsic, external, and historical intensity, respectively. In addition, an incremental updating strategy is implemented in FMHP, continuously updating event intensity as new events occur. We conduct extensive experiments on four public datasets (e.g., Amazon Beauty, LastFM, Movielens, and Amazon Book). Compared with state-of-the-art incremental recommendation methods, our proposed FMHP model achieves superior performance, with up to 9.77%, 9.35%, 9.32%, 10.10% w.r.t. HR, 8.86%, 10.26%, 9.81%, 9.38% w.r.t. NDCG, and 9.64%, 9.32%, 8.97%,9.85% w.r.t. Recall in Beauty, LastFM, MovieLens, and Book, respectively. Besides, the case study shows that the three factors in our proposed FMHP method play a vital role in triggering event generation.

A split-federated learning and edge-cloud based efficient and privacy-preserving large-scale item recommendation model

The combination of federated learning and recommender system aims to solve the privacy problems of recommendation through keeping user data locally at the client device during the model training session. However, most existing approaches rely on user devices to fully compute the deep model designed for the large-scale item recommendation; therefore, imposing high calculation and communication overheads on resource-constrained user devices. Consequently, achieving efficient federated recommendations across ubiquitous mobile devices remains an open research problem. To this end, in this paper we propose an efficient and privacy-preserving federated learning framework which is based on the cloud-edge collaboration for large-scale item recommendation called SpFedRec. In our method, to reduce the computation and communication cost of the federated two-tower model, a split learning approach is applied to migrate the item model from participants’ edge devices to the computationally powerful cloud side and compress item data while transmitting. Meanwhile, to enhance the feature representation, the Squeeze-and-Excitation network mechanism is used on the backbone model to optimize the perception of dominant features. Moreover, because the gradients transmitted contain private information about the user; therefore, we propose a multi-party circular secret-sharing chain based on secret sharing for better privacy protection. Extensive experiments using plausible assumptions on two real-world datasets demonstrate that our proposed method improves the average computation time and communication cost by 23% and 49%, respectively. Furthermore, the proposed model accomplishes comparable performance with other state-of-art federated recommendation models.

Modeling Within-Basket Auxiliary Item Recommendation with Matchability and Ubiquity

Within-basket recommendation is to recommend suitable items for the current basket with some already known items. The within-basket auxiliary item recommendation ( WBAIR ) is to recommend auxiliary items based on the primary items in the basket. Such a task exists in many real-life scenarios. Unlike the associations between items that can be transmitted in both directions, primary and auxiliary relationships are unidirectional. Then, the suitable matching patterns between primary and auxiliary items cannot be explored by traditional directionless methods. Therefore, we design the Matc4Rec algorithm to integrate the primary and auxiliary factors, and finally recommend items that not only match the interests of users but also satisfy the primary and auxiliary relationships between items. Specifically, we capture the pattern from three aspects: matchability within-basket , matchability between baskets , and ubiquity . By exploiting this pattern, the designed algorithm not only achieves good results on real-world datasets but also improves the interpretability of recommendations. As a result, we can know which commodities are suitable as auxiliary items. The experiment results demonstrate that our algorithm can also alleviate the cold start problem.

Egyptian evidence-based consensus on clinical practice recommendations for the management of Kawasaki disease

BackgroundKawasaki disease (KD) is an acute, self-limited febrile disease of unidentified cause that mostly affects children less than 5 years of age. This work aimed to provide an appropriate Egyptian evidence-based consensus on clinical practice recommendations for the management of Kawasaki disease. The main objective of this study, which employed the Delphi method, was to reach a consensus among experts on a treat-to-target management approach for KD.ResultsThe expert panel was confined to an online survey (n=26), and all the expert completed the three rounds. At the conclusion of round 3, a total of 17 recommendation items were gathered, which were divided into two sections. The range of respondents (ranks 7–9) who agreed with the recommendations was 92.3 to 100%. All 17 clinical standards identified by the scientific committee were written in the same way. There have been algorithms proposed for managing various KD conditions.ConclusionThe developed evidence-based consensus recommendations for the diagnosis and management of KD represent an up-to-date document that focuses on clinical management questions which are generally posed to health care professionals involved in the management of KD. This guideline was developed considering experience with and availability of treatment and diagnostic options in Egypt.

Spit It Out: Is Caffeine Mouth Rinse an Effective Ergogenic Aid? A Systematic Review and Meta-Analysis

ABSTRACT The effects of caffeine mouth rinse (Caff-MR) on sports performance have been evaluated by recent studies presenting contrasting results. Therefore, this systematic review aimed to summarize the evidence regarding the effects of Caff-MR on sports performance. According to Preferred Reporting Items for Systematic Reviews and Meta-Analyses recommendations, the search through PubMed, Web of Science, and Scopus databases was performed until December 2021. A random-effects meta-analysis of standardized mean differences was performed to assess the overall effect of Caff-MR on sports performance-related outcomes. Sixteen studies met the inclusion criteria (207 participants). The meta-analysis showed no significant overall effect of Caff-MR on sports performance (p = 0.09). Moreover, no effects were shown on maximal heart rate or ratings of perceived exertion. The risk of bias for the included studies assessed using the risk of bias in randomized trials Cochrane tool presented 87.5% (n = 14) of the studies as medium and 12.5% (n = 2) as high scores. These data do not support using Caff-MR for sports performance, although more consistent and well-controlled studies are warranted. Large heterogeneity in study designs precludes making practical recommendations for exercise performance. More research is warranted to elucidate the potential of Caff-MR, specifically focusing on determining the optimal dose.

Expert consensus for Jingtong Granules in treatment of cervical radiculopathy in clinical application

With the effects of activating blood and resolving stasis, and moving Qi to relieve pain, Jingtong Granules is widely used in the treatment of cervical radiculopathy in China. Long-term clinical application and related evidence have shown that the prescription has ideal effect in alleviating the pain in neck, shoulder, and upper limbs, stiffness or scurrying numbness, and scurrying pain caused by this disease. However, there is a lack of consensus on the clinical application of Jingtong Granules. Therefore, clinical first-line experts and methodology experts from all over the country were invited to compile this expert consensus. This expert consensus is expected to guide clinicians to use Jingtong Granules in a standardized and reasonable way, improve clinical efficacy, reduce medication risks, and benefit patients. First, according to the clinical experience of experts and the standard development procedures, the indications, syndrome characteristics, clinical advantages, and possible adverse reactions of Jingtong Granules were summarized. Then, through face-to-face interview of clinical doctors in traditional Chinese medicine and western medicine and survey of the clinical application, the clinical problems were summed up, and the consensus was reached with the nominal group method to form the final clinical problems. Third, evidence retrieval was carried out for the clinical problems, and relevant evidence was evaluated. The GRADE system was employed to rate the quality of evidence. Fourth, 5 recommendation items and 3 consensuses items were summarized with the nominal group method. Opinions and peer reviews on the consensus content were solicited through expert meetings and letter reviews. The final consensus includes the summary of evidence on the clinical indications, effectiveness, and safety of Jingtong Granules, which can serve as a reference for clinicians in hospitals and primary health institutions.