Service Recommendation Research Articles

Smartgrid-based hybrid digital twins framework for demand side recommendation service provision in distributed power systems

Electricity consumers face challenges in selecting an optimal energy-saving plan, and this is a sustainability problem. To set consumers focus on sustainable energy management, developments around ”Industry 4.0” are needed to achieve an optimal balance between cost and energy consumption with a focus on cutting-edge machine-learning models and smart services introduction. Energy modelling is crucial for confronting the challenges introduced by the energy transition. This research is to promote digital twin (DT) technology in smart energy grids. The hybrid digital twin model is developed based on smart grid end-users’ electrical components. This approach emulates the component specification of the reference smart grid system and encourages a reduction in net energy consumption. Additionally, the application of AI connects smart meters, IoT devices, and the assets of the smart grids to the DT recreation of demand-side end-users for energy efficiency recommendation provision; this setup improves energy management, energy efficiency, and the usage of renewable energy resources. The novelty of this study is that the recent scope of demand-side recommendation schemes has been extended in the direction of cutting-edge Industry 4.0 hybrid DTs. Additionally, a prototype system and its functional characteristics have been proposed with the potential to pave the path for utilization in microgrid environments. Simulation results show a unique level of parallelism between the reference system and the developed model, along with a 36.8% reduction in net energy consumption following implementation of the recommendation advice.

Combination of autochthonous Lactobacillus strains and trans-Cinnamaldehyde in water reduces Salmonella Heidelberg in turkey poults.

Reducing the colonization of Salmonella in turkeys is critical to mitigating the risk of its contamination at later stages of production. Given the increased susceptibility of newly hatched poults to Salmonella colonization, it is crucial to implement interventions that target potential transmission routes, including drinking water. As no individual intervention explored to date is known to eliminate Salmonella, the United States Department of Agriculture-Food Safety Inspection Service (USDA-FSIS) recommends employing multiple hurdles to achieve a more meaningful reduction and minimize the potential emergence of resistance. Probiotics and plant-derived antimicrobials (PDAs) have demonstrated efficacy as interventions against Salmonella in poultry. Therefore, this study aimed to investigate the use of turkey-derived Lactobacillus probiotics (LB; a mixture of Lactobacillus salivarius UMNPBX2 and L. ingluviei UMNPBX19 isolated from turkey ileum) and a PDA, trans-cinnamaldehyde (TC), alone and in combination (CO), against S. Heidelberg in turkey drinking water and poults. The presence of 5% nutrient broth or cecal contents as contaminants in water resulted in S. Heidelberg growth. TC eliminated S. Heidelberg, regardless of the contaminants present. In contrast, the cecal contents led to increased survival of Lactobacillus in the CO group. Unlike TC, LB was most effective against S. Heidelberg when the nutrient broth was present, suggesting the role of secondary metabolites in its mechanism of action. In the experiments with poults, individual TC and LB supplementation reduced cecal S. Heidelberg in challenged poults by 1.2- and 1.7-log10 colony-forming units (CFU)/g cecal contents, respectively. Their combination yielded an additive effect, reducing S. Heidelberg by 2.7 log10 CFU/g of cecal contents compared to the control (p ≤ 0.05). However, the impact of TC and LB on the translocation of S. Heidelberg to the liver was more significant than CO. TC and LB are effective preharvest interventions against S. Heidelberg in poultry production. Nonetheless, further investigations are needed to determine the optimum application method and its efficacy in adult turkeys.

Inappropriate use of proton pump inhibitors in hospitalized patients with lower gastrointestinal bleeding

ABSTRACT Objectives Use of proton pump inhibitors (PPIs) is a mainstay in treating upper gastrointestinal bleeding (UGIB). However, the beneficial effects of PPIs are not anticipated to extend beyond the duodenum and may actually contribute to the risk of lower gastrointestinal bleeding (LGIB). However, in practice, PPIs are often used for inpatients with LGIB where no benefit exists. Methods A retrospective chart review was performed on inpatients during a 2-year period at an urban academic teaching hospital. Inpatients with consults to the gastroenterology (GI) service with confirmed or highly suspected LGIB were included. Outcomes regarding PPI use and the GI consulting service recommendations in these 225 patients were evaluated. Results About 37.8% of patients were started on a PPI during their inpatient course. Of those, 46% patients started on a PPI had no indication for PPI and 85% had no recommendation by the GI consultants to start a PPI. Of the 85 patients started on PPI, the GI consultants recommended stopping it in two (2.3%) patients. Lastly, 20 patients (9%) were discharged on PPI without an indication for PPI. Conclusion To our knowledge, this is the first study that looked at the inappropriate utilization of PPIs in patients admitted for LGIBs utilizing GI consultant recommendations. Given the large proportion of patients started on PPI without a clinical indication and continued at discharge and the paucity of GI recommendations to discontinue inappropriate use, we found that clinical care may be improved with formal GI recommendations regarding use of PPI.

Hybrid knowledge-infused collaborative filtering for enhanced movie clustering and recommendation

This article proposes an enhanced knowledge-based collaborative filtering model for movie recommendation services to address the limitations of collaborative filtering in capturing the diverse preferences and specific characteristics of movies. The proposed model integrates external knowledge sources, such as movie plots and reviews, to enrich the recommendation process. By leveraging this additional information, the model can better understand movies’ unique features and attributes, improving recommendation accuracy and relevance. The knowledge-based features are extracted and incorporated into the collaborative filtering framework, enhancing the model’s ability to match user preferences with movie characteristics. Experiments are conducted using the MovieLens dataset to evaluate the proposed model. The MAE and RMSE metrics are employed to assess the quality of recommendations. Comparative analyses are conducted against various baseline approaches, including popularity-based, CF-based, content-based, and hybrid recommendation models. The experimental results demonstrate the effectiveness of the proposed knowledge-based collaborative filtering model. The proposed model consistently outperforms the baselines, providing more accurate and personalized recommendations.

Dynamic matrix factorization-based collaborative filtering in movie recommendation services

Movies are a primary source of entertainment, but finding specific content can be challenging given the exponentially increasing number of movies produced each year. Recommendation systems are extremely useful for solving this problem. While various approaches exist, Collaborative Filtering (CF) is the most straightforward. CF leverages user input and historical preferences to determine user similarity and suggest movies. Matrix Factorization (MF) is one of the most popular Collaborative Filtering (CF) techniques. It maps users and items into a joint latent space, using a vector of latent features to represent each user or item. However, traditional MF techniques are static, while user cognition and product variety are constantly evolving. As a result, traditional MF approaches struggle to accommodate the dynamic nature of user-item interactions. To address this challenge, we propose a Dynamic Matrix Factorization CF model for movie recommendation systems (DMF-CF) that considers the dynamic changes in user interactions. To validate our approach, we conducted evaluations using the standard MovieLens dataset and compared it to state-of-the-art models. Our preliminary findings highlight the substantial benefits of DMF-CF, which outperforms recent models on the MovieLens-100K and MovieLens-1M datasets in terms of Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) metrics.

Recommendations of the Spanish-Portuguese Internal Medicine services in the fight against climate change and environmental degradation.

Facing the severity of the impact of climate change and environmental degradation on human health, 32 Internal Medicine societies, colleges, and associations of 29 Spanish and Portuguese-speaking countries issue a consensus document in which they call for the implication of doctors and all health professionals in the global fight against the causes of these changes. This commitment requires the cooperation of health-related organizations, elaboration and implementation of good environmental sustainability practices, greater awareness of professionals and population, promotion of education and research in this area, increasing climate resilience and environmental sustainability of health systems, combating inequalities and protecting the most vulnerable populations, adopting behaviors that protect the environment, and claiming Internal Medicine as a core specialty for empowerment of the health system to respond to these challenges.

A collaboration team to build social service partnerships within a safety-net health system

BackgroundTo facilitate safety-net healthcare system partnerships with community social service providers, the Los Angeles County Department of Health Services (LAC DHS) created a new collaboration team to spur cross-agency social and medical referral networks and engage communities affected by health disparities as part of a Sect. 1115 Medicaid waiver in Los Angeles County entitled Whole Person Care-Los Angeles (WPC-LA).MethodsThis observational research reviews three years of collaboration team implementation (2018–2020) through Medicaid-reportable engagement reports, a collaboration team qualitative survey on challenges, facilitators, and recommendations for community engagement. Member reflections for survey findings were conducted with the collaboration team and LAC DHS WPC-LA leadership.ResultsCollaboration team Medicaid engagement reports (n = 144) reported > 2,700 events, reaching > 70,000 individuals through cross-agency and community-partnered meetings. The collaboration team survey (n = 9) and member reflection sessions portrayed engagement processes through outreach, service assessments, and facilitation of service partnerships. The collaboration team facilitated community engagement processes through countywide workgroups on justice-system diversion and African American infant and maternal health. Recommendations for future safety net health system engagement processes included assessing health system readiness for community engagement and identifying strategies to build mutually beneficial social service partnerships.ConclusionsA dedicated collaboration team allowed for bi-directional knowledge exchange between county services, populations with lived experience, and social services, identifying service gaps and recommendations. Engagement with communities affected by health disparities resulted in health system policy recommendations and changes.

Evaluating the effectiveness of trauma-informed care training delivered to staff in community Older Adult Mental Health Services

PurposeBy older adulthood, the majority of individuals will have experienced at least one traumatic event. Trauma-informed care (TIC) is proposed to improve effectivity of health-care provision and to reduce likelihood of services causing retraumatisation. This study aims to assess the effectiveness of staff training in TIC in older adult services.Design/methodology/approachTIC training was delivered across eight Older Adult Community Mental Health Teams in the same UK organisation. Questionnaires were administered before and after training: a psychometrically robust measure, the Attitudes Related to Trauma-Informed Care, was used to assess TIC-related attitudes, and a service-developed scale was used to measure changes in TIC competence. Data was analysed using linear mixed effects modelling (LMM). Qualitative data regarding the impact of training was gathered one month after training through a free-text questionnaire.FindingsThere were 45 participants, all of whom were white British. LMM on pre- and post-data revealed that staff training significantly increased competencies across all measured TIC domains. Overall, staff attitudes were also significantly more trauma-informed after training. Qualitatively, staff identified time as the only additional resource required to deliver the skills and knowledge gained from training.Practical implicationsTraining was found to be effective in increasing TIC-related skills and attitudes. Organisations aiming to become trauma-informed should consider staff training as one aspect of a wider development plan.Originality/valueTo the best of the authors’ knowledge, this paper is the first to examine TIC training for staff working in Older Adults Mental Health Services. Recommendations for services aiming to develop a trauma-informed culture have been provided.

Parents’ experience of children with acquired brain injury undergoing neurorehabilitation: a systematic review

BackgroundThis systematic review aims to synthesise the qualitative evidence exploring parents’ experiences of children with acquired brain injury (ABI) undergoing neurorehabilitation during the first year post-injury.MethodsA systematic review of qualitative...

Frontline Perspectives on Buprenorphine for the Management of Chronic Pain.

Due to the prevalence of chronic pain and high-impact chronic pain in the US, a significant percentage of the population is prescribed opioids for pain management. However, opioid use disorder is associated with reduced quality of life, along with fatal opioid overdoses, and is a significant burden on the US economy. Considering the clinical needs of patients with intractable chronic pain and the potential harms associated with prescribed and illicit opioids in our communities, having a deep understanding of current treatment options, supporting evidence, and clinical practice guidelines is essential for optimizing treatment selections. Buprenorphine is a Schedule III opioid with a unique mechanism of action, allowing effective and long-lasting analgesia at microgram doses with fewer negative side effects and adverse events, including respiratory depression, when compared with other immediate-release, long-acting, and extended-release prescription opioids. Due to its relatively lower risk for overdose and misuse, buprenorphine was recently added to the Clinical Practice Guideline for the Use of Opioids in the Management of Chronic Pain as a first-line treatment for chronic pain managed by opioids by the US Departments of Defense and Veterans Affairs, and the Department of Health and Human Services recommends that buprenorphine be made available for the treatment of chronic pain. In this narrative review, we discuss the different buprenorphine formulations, clinical efficacy, advantages for older adults and other special populations, clinical practice guideline recommendations, and payer considerations of buprenorphine and suggest that buprenorphine products approved for chronic pain should be considered as a first-line treatment for this indication.

GCNSLIM: Graph convolutional network with sparse linear methods for e-government service recommendation

E-government service recommendation aims to find target service items for users based on their preferences, behaviors and other information, which is one of the key technologies for information overload in e-government cloud platforms. However, it has not received adequate attention in comparison to other recommendation scenarios like news and music recommendation. Graph Convolutional Networks (GCNs) have made significant strides in Collaborative Filtering (CF) recommendations. However, existing GCN-based CF methods are mainly based on matrix factorization (MF) and incorporate some optimization techniques (e.g., contrastive learning) to enhance performance, which are not enough to handle the complexities of diverse real-world recommendation scenarios. We empirically find that when existing GCN-based CF methods are directly applied to e-government service recommendation, they are limited by the MF framework and showing poor performance. This is attributed to the fact that the equal treatment of users and items by MF is not appropriate for scenarios where the number of users and items is unbalanced. In this work, we propose a new model, GCNSLIM, which combines GCN and sparse linear methods (SLIM) instead of combining GCN and MF to accommodate e-government service recommendation. In particular, GCNSLIM explicitly injects high-order collaborative signals obtained from multi-layer light graph convolutions into the item similarity matrix in the SLIM framework, effectively improving the recommendation accuracy. In addition, we propose two optimization measures, removing layer 0 embedding and adding nonlinear activation, to further adapt to the characteristics of e-government service recommendation scenarios. Furthermore, we propose a joint optimization mode to enhance the adaptability for more diverse recommendation scenarios. We conduct extensive experiments on a real e-government service dataset and two common public datasets and demonstrate the effectiveness of GCNSLIM in recommendation accuracy and operational performance. Our implementation is available at https://github.com/songkk5/GCNSLIM.

Report: China’s New AI Regulations

The People’s Republic of China (PRC) has moved ahead of other jurisdictions in regulating Artificial Intelligence (AI) by enacting a set of regulations aimed at addressing risks associated with AI. This article reviews three key AI specific regulations enacted in the PRC in 2022 and 2023. Together these regulations set out comprehensive obligations on the use of generative AI and algorithmic recommendation services in the PRC. The obligations include general requirements such as to comply with the law and principles of fairness as well as to protect against incorrect political direction and violation of social morality. There are also more specific obligations to achieve those objectives. These include, ensuring the regulator has visibility of certain algorithms by requiring the filing of information about the relevant algorithms. Further there are specific obligations on service providers such as to label training data appropriately, take measures to screen illegal or harmful content, verify the identities of users for some services, protect against fake news and label artificial intelligence generated content as synthetic. The regulations also provide a suite of rights for users, including the right not to be targeted on the basis of their personal characteristics by services using AI algorithms. The obligations under these regulations are significant and it will be interesting to see how the PRC’s regulatory approach impacts the AI industry in the PRC in contrast to other jurisdictions where regulators are still looking at how to regulate AI or taking a different approach. PRC, China, Artificial Intelligence, AI Regulations

Multi-Modal Deep Learning based Metadata Extensions for Video Clipping

General video search and recommendation systems primarily rely on metadata and personal information. Metadata includes file names, keywords, tags, and genres, among others, and is used to describe the video's content. The video platform assesses the relevance of user search queries to the video metadata and presents search results in order of highest relevance. Recommendations are based on videos with metadata judged to be similar to the one the user is currently watching. Most platforms offer search and recommendation services by employing separate algorithms for metadata and personal information. Therefore, metadata plays a vital role in video search. Video service platforms develop various algorithms to provide users with more accurate search results and recommendations. Quantifying video similarity is essential to enhance the accuracy of search results and recommendations. Since content producers primarily provide basic metadata, it can be abused. Additionally, the resemblance between similar video segments may diminish depending on its duration. This paper proposes a metadata expansion model that utilizes object recognition and Speech-to-Text (STT) technology. The model selects key objects by analyzing the frequency of their appearance in the video, extracts audio separately, transcribes it into text, and extracts the script. Scripts are quantified by tokenizing them into words using text-mining techniques. By augmenting metadata with key objects and script tokens, various video content search and recommendation platforms are expected to deliver results closer to user search terms and recommend related content.

보험비교플랫폼 규율 방향

Even after the insurance comparison and recommendation services of fintech companies, including big tech companies, were allowed as a regulatory special case, it took more time for the service to start. Now, big tech companies are allowed to enter the insurance market. This is desirable in that it has a net function of bringing competition to the insurance market, rationally adjusting the price of insurance products according to the market mechanism and increasing the convenience of insurance for consumers. However, it is now time to pay attention to prevent consumer damage by framing business conduct regulations well.
 Referring to overseas cases that have experienced such services and faced regulatory concerns before us, we can come to the conclusion that we do not need to be bound by the principle of same function, same regulation. Even among traditional insurance brokers and platform companies, there are significant differences between small fintechs and big techs in terms of market power and group structure. Even if they provide the same service, there is a need to establish an entity-based dual regulatory system. The Korean insurance market is expected to have a greater impact than in the West because it has not experienced traditional insurance brokerage services, and there will be a need for a stronger regulatory framework. As big tech is allowed to enter the new financial market of insurance, it should not accept the strengthened behavioral regulation as an unfair regulation. On the contrary, there is an advantage to expanding insurance services within the framework of legislative regulation if it proves its role in promoting the convenience and welfare of insurance consumers under enhanced behavioral regulation.
 For the time being, the system will be guided by the financial authorities, but we should prepare for legislation. While the focus will be on business conduct regulation, the framework for prudential regulation should also be discussed in stages.

Research on the application of personalized recommendation

With the development of the Internet, people have more opportunities to be exposed to personalized recommendations, and the heat of related topics has increased. The purpose of this paper is to summarize the application principle and use of personalized recommendations by analyzing the relevant data, so that people can have a better understanding of personalized recommendations and make preparations for future related research. This paper first introduces the general development context of the personalized recommendation system and its uniqueness. The analysis finds that personalized recommendations can provide users with the most relevant and valuable information, goods, or services according to their interests, preferences, and behaviors. Secondly, this paper explores the application scenarios of personalized recommendation systems in different fields, including but not limited to video content, e-commerce, and online learning. In addition, this paper also lists the application of YouTube and Netflix in video content recommendation, as well as personalized recommendation services in e-commerce and online learning platforms. In particular, the rise of intelligent adaptive online learning systems and their adaptability in the field of education and training are analyzed in detail. Through this research, it can be found that a personalized recommendation system is very beneficial to people's production and life, but at present, personalized recommendation is still in the development stage, and there are many problems to be solved.

Knowledge distillation representation and DCNMIX quality prediction‐based Web service recommendation

SummaryWeb service recommendation as an emerging topic attracts increasing attention due to its important practical significance. As the number of available Web services continues to grow, users face the challenge of searching the most suitable services that meet their specific needs. Quality of service (QoS)‐based service recommendation becomes a popular approach to address this issue. However, existing QoS‐based service recommendation methods are inability to effectively capture valuable content and structural information from services. These methods often rely solely on low‐order explicit feature intersections in QoS information, do not fully utilize the high‐order implicit feature intersections, and ignore the rich semantic information existing in service descriptions and user preferences. To address this problem, this paper proposes a Web service recommendation method via combining knowledge distillation representation and DCNMIX quality prediction. This method combines content‐based and structure‐based service classification and service prediction based on multi‐dimensional service quality information. First, it builds a service relationship network using semantic features extracted from service descriptions. Second, it designs a graph neural network knowledge distillation framework. The teacher model extracts the knowledge of the graph neural network model, and the student model learns the structure‐based and feature‐based prior knowledge of the service relationship network. Then the student model is used to learn the knowledge of the teacher model, classify Web services, and obtain service representations. Finally, based on service representations and multi‐dimensional QoS information, it exploits the DCNMIX model to learn the explicit and implicit features intersections of Web services and obtain the prediction score and ranking of Web services. The experimental results on the ProgrammableWeb dataset show that the proposed method outperforms the state‐of‐the‐art baselines in terms of Recall, F1, Logloss, and AUC_ROC.

Manufacturing cloud service recommendation model based on deep feature learning and user preference perception

ABSTRACT The manufacturing cloud service recommendation model is a key technology for users to quickly discover and obtain the personalized services they need from the increasingly expanding manufacturing cloud; however, the manufacturing cloud service recommendation model faces problems, such as difficulty in representing deep features of services and too sparse user preference models. To address the above problems, this paper proposes a manufacturing cloud service recommendation model framework based on deep feature learning and user preference perception and investigates the key technologies involved. On the one hand, a Convolutional Neural Network (CNN)-based deep feature extraction method for manufacturing cloud services is proposed by combining the Latent Dirichlet Allocation (LDA) topic model and word2vec. On the other hand, the concept of a category is introduced through the method of label clustering to perceive user behavior, efficiently build user preference models, and solve the sparse problem of the user-service matrix. Based on this, the execution process of the recommendation model for manufacturing cloud services is analyzed by fusing the deep features of services and user preference models. The experimental results on the actual dataset prove that the method greatly improves the recommendation effect.

Time-Aware Dual LSTM Neural Network with Similarity Graph Learning for Remote Sensing Service Recommendation.

Technological progress has led to significant advancements in Earth observation and satellite systems. However, some services associated with remote sensing face issues related to timeliness and relevance, which affect the application of remote sensing resources in various fields and disciplines. The challenge now is to help end-users make precise decisions and recommendations for relevant resources that meet the demands of their specific domains from the vast array of remote sensing resources available. In this study, we propose a remote sensing resource service recommendation model that incorporates a time-aware dual LSTM neural network with similarity graph learning. We further use the stream push technology to enhance the model. We first construct interaction history behavior sequences based on users' resource search history. Then, we establish a category similarity relationship graph structure based on the cosine similarity matrix between remote sensing resource categories. Next, we use LSTM to represent historical sequences and Graph Convolutional Networks (GCN) to represent graph structures. We construct similarity relationship sequences by combining historical sequences to explore exact similarity relationships using LSTM. We embed user IDs to model users' unique characteristics. By implementing three modeling approaches, we can achieve precise recommendations for remote sensing services. Finally, we conduct experiments to evaluate our methods using three datasets, and the experimental results show that our method outperforms the state-of-the-art algorithms.

Multi-representation web service recommendation system based on attention mechanism

Multi-representation web service recommendation system based on attention mechanism

Research on the application of bilateral collaborative algorithm in hotel tourism consumption preference recommendation

SummaryThe continuous expansion of the domestic tourism market has driven the development of the hotel related industry. It is a key focus of the hotel industry to provide accurate supporting services to tourists based on their preferences. However, the current recommendation model mainly considers tourist factors and lacks consideration for hotel factors. In order to address the concerns of recommendation services, a bilateral collaborative recommendation model is proposed, which first analyzes tourist feature data through comprehensive similarity calculation. Firstly, the comprehensive similarity calculation is used to analyze tourist feature data, and the K‐means Clustering Algorithm (K‐means) is introduced to improve the clustering effect between data. And to better distinguish tourist preferences, the Gale Shapley (G‐S) bilateral matching model is introduced to achieve preference matching of consumer features. In the Mean Absolute Error (MAE) analysis of the model, the User‐based Collaborative Filtering (UBCF) model performs the worst. When the k value is 90, the minimum MAE value is 0.7896. The Item‐based Collaborative Filtering (IBCF) model is 0.7776, with the best performance being the improved G‐S model of 0.7325. In the performance testing of three models' preference recommendations, the improved G‐S model achieves an accuracy of over 90% in all three types of feature data recommendations, with the lowest accuracy of 62.5% for UBCF and 71.3% for IBCF. It can be seen that the proposed improved G‐S model can accurately identify tourist and hotel preference data and accurately recommend corresponding projects. The research content provides important technical references for the digital transformation and development of the tourism industry.