Differently from the general online social network(OSN),locationbased mobile social network(LMSN),which seamlessly integrates mobile computing and social computing technologies,has unique characteristics of temporal,s...Differently from the general online social network(OSN),locationbased mobile social network(LMSN),which seamlessly integrates mobile computing and social computing technologies,has unique characteristics of temporal,spatial and social correlation.Recommending friends instantly based on current location of users in the real world has become increasingly popular in LMSN.However,the existing friend recommendation methods based on topological structures of a social network or non-topological information such as similar user profiles cannot well address the instant making friends in the real world.In this article,we analyze users' check-in behavior in a real LMSN site named Gowalla.According to this analysis,we present an approach of recommending friends instantly for LMSN users by considering the real-time physical location proximity,offline behavior similarity and friendship network information in the virtual community simultaneously.This approach effectively bridges the gap between the offline behavior of users in the real world and online friendship network information in the virtual community.Finally,we use the real user check-in dataset of Gowalla to verify the effectiveness of our approach.展开更多
Citation network is often used for academic recommendation. However, it is difficult to achieve high recommendation accuracy and low time complexity because it is often very large and sparse and different citations ha...Citation network is often used for academic recommendation. However, it is difficult to achieve high recommendation accuracy and low time complexity because it is often very large and sparse and different citations have different purposes. What's more, some citations include unreasonable information, such as in case of intentional self-citation. To improve the accuracy of citation network-based academic recommendation and reduce the time complexity, we propose an academic recommendation method for recommending authors and papers. In which, an author-paper bilayer citation network is built, then an enhanced topic model, Author Community Topic Time Model(ACTTM) is proposed to detect high quality author communities in the author layer, and a set of attributes are proposed to comprehensively depict the author/paper nodes in the bilayer citation network. Experimental results prove that the proposed ACTTM can detect high quality author communities and facilitate low time complexity, and the proposed academic recommendation method can effectively improve the recommendation accuracy.展开更多
With the development of the Internet of Things(Io T), people's lives have become increasingly convenient. It is desirable for smart home(SH) systems to integrate and leverage the enormous information available fro...With the development of the Internet of Things(Io T), people's lives have become increasingly convenient. It is desirable for smart home(SH) systems to integrate and leverage the enormous information available from IoT. Information can be analyzed to learn user intentions and automatically provide the appropriate services. However, existing service recommendation models typically do not consider the services that are unavailable in a user's living environment. In order to address this problem, we propose a series of semantic models for SH devices. These semantic models can be used to infer user intentions. Based on the models, we proposed a service recommendation probability model and an alternative-service recommending algorithm. The algorithm is devoted to providing appropriate alternative services when the desired service is unavailable. The algorithm has been implemented and achieves accuracy higher than traditional Hidden Markov Model(HMM). The maximum accuracy achieved is 68.3%.展开更多
In this paper, we examine methods that can provide accurate results in a form of a recommender system within a social networking framework. The social networking site of choice is Twitter, due to its interesting socia...In this paper, we examine methods that can provide accurate results in a form of a recommender system within a social networking framework. The social networking site of choice is Twitter, due to its interesting social graph connections and content characteristics. We built a recommender system which recommends potential users to follow by analyzing their tweets using the CRM114 regex engine as a basis for content classification. The evaluation of the recommender system was based on a dataset generated from real Twitter users created in late 2009.展开更多
Location based social networks（ LBSNs） provide location specific data generated from smart phone into online social networks thus people can share their points of interest（ POIs）. POI collections are complex and c...Location based social networks（ LBSNs） provide location specific data generated from smart phone into online social networks thus people can share their points of interest（ POIs）. POI collections are complex and can be influenced by various factors,such as user preferences,social relationships and geographical influence. Therefore,recommending new locations in LBSNs requires to take all these factors into consideration. However,one problem is how to determine optimal weights of influencing factors in an algorithm in which these factors are combined. The user similarity can be obtained from the user check-in data,or from the user friend information,or based on the different geographical influences on each user＇s check-in activities. In this paper,we propose an algorithm that calculates the user similarity based on check-in records and social relationships,using a proposed weighting function to adjust the weights of these two kinds of similarities based on the geographical distance between users. In addition,a non-parametric density estimation method is applied to predict the unique geographical influence on each user by getting the density probability plot of the distance between every pair of user＇s check-in locations. Experimental results,using foursquare datasets,have shown that comparisons between the proposed algorithm and the other five baseline recommendation algorithms in LBSNs demonstrate that our proposed algorithm is superior in accuracy and recall,furthermore solving the sparsity problem.展开更多
Depression is a common disorder worldwide. In the West, depression is also called ＂the blues＂ informally. It is a psychological disorder characterized by an all-encompassing low mood. Slightly depressed people may b...Depression is a common disorder worldwide. In the West, depression is also called ＂the blues＂ informally. It is a psychological disorder characterized by an all-encompassing low mood. Slightly depressed people may behave normally, while experiencing feelings of pain at heart. In mild depression cases, people may feel gloomy or distressed, which is usually associated with memory impairment, insomnia,展开更多
The performance of deep recommendation models degrades significantly under data poisoning attacks.While adversarial training methods such as Vulnerability-Aware Training(VAT)enhance robustness by injecting perturbatio...The performance of deep recommendation models degrades significantly under data poisoning attacks.While adversarial training methods such as Vulnerability-Aware Training(VAT)enhance robustness by injecting perturbations into embeddings,they remain limited by coarse-grained noise and a static defense strategy,leaving models susceptible to adaptive attacks.This study proposes a novel framework,Self-Purification Data Sanitization(SPD),which integrates vulnerability-aware adversarial training with dynamic label correction.Specifically,SPD first identifies high-risk users through a fragility scoring mechanism,then applies self-purification by replacing suspicious interactions with model-predicted high-confidence labels during training.This closed-loop process continuously sanitizes the training data and breaks the protection ceiling of conventional adversarial training.Experiments demonstrate that SPD significantly improves the robustness of both Matrix Factorization(MF)and LightGCN models against various poisoning attacks.We show that SPD effectively suppresses malicious gradient propagation and maintains recommendation accuracy.Evaluations on Gowalla and Yelp2018 confirmthat SPD-trainedmodels withstandmultiple attack strategies—including Random,Bandwagon,DP,and Rev attacks—while preserving performance.展开更多
Contemporary Introduction to Chinese Medicine in Comparison with Western Medicine,recently published by China,is exceptional and worthwhile reading,especially for Western readers.It is authored by Professor Xie Zhu-fa...Contemporary Introduction to Chinese Medicine in Comparison with Western Medicine,recently published by China,is exceptional and worthwhile reading,especially for Western readers.It is authored by Professor Xie Zhu-fan in collaboration with Dr. Xie Fang.Professor Xie is a renowned scholar of both Western and Chinese medicine in China,and the Director Emeritus of Peking University Institute of integrated Chinese and Western Medicine.展开更多
Recommendation systems are key to boosting user engagement,satisfaction,and retention,particularly on media platforms where personalized content is vital.Sequential recommendation systems learn from user-item interact...Recommendation systems are key to boosting user engagement,satisfaction,and retention,particularly on media platforms where personalized content is vital.Sequential recommendation systems learn from user-item interactions to predict future items of interest.However,many current methods rely on unique user and item IDs,limiting their ability to represent users and items effectively,especially in zero-shot learning scenarios where training data is scarce.With the rapid development of Large Language Models(LLMs),researchers are exploring their potential to enhance recommendation systems.However,there is a semantic gap between the linguistic semantics of LLMs and the collaborative semantics of recommendation systems,where items are typically indexed by IDs.Moreover,most research focuses on item representations,neglecting personalized user modeling.To address these issues,we propose a sequential recommendation framework using LLMs,called CIT-Rec,a model that integrates Collaborative semantics for user representation and Image and Text information for item representation to enhance Recommendations.Specifically,by aligning intuitive image information with text containing semantic features,we can more accurately represent items,improving item representation quality.We focus not only on item representations but also on user representations.To more precisely capture users’personalized preferences,we use traditional sequential recommendation models to train on users’historical interaction data,effectively capturing behavioral patterns.Finally,by combining LLMs and traditional sequential recommendation models,we allow the LLM to understand linguistic semantics while capturing collaborative semantics.Extensive evaluations on real-world datasets show that our model outperforms baseline methods,effectively combining user interaction history with item visual and textual modalities to provide personalized recommendations.展开更多
Recommending personalized travel routes from sparse,implicit feedback poses a significant challenge,as conventional systems often struggle with information overload and fail to capture the complex,sequential nature of...Recommending personalized travel routes from sparse,implicit feedback poses a significant challenge,as conventional systems often struggle with information overload and fail to capture the complex,sequential nature of user preferences.To address this,we propose a Conditional Generative Adversarial Network(CGAN)that generates diverse and highly relevant itineraries.Our approach begins by constructing a conditional vector that encapsulates a user’s profile.This vector uniquely fuses embeddings from a Heterogeneous Information Network(HIN)to model complex user-place-route relationships,a Recurrent Neural Network(RNN)to capture sequential path dynamics,and Neural Collaborative Filtering(NCF)to incorporate collaborative signals from the wider user base.This comprehensive condition,further enhanced with features representing user interaction confidence and uncertainty,steers a CGAN stabilized by spectral normalization to generate high-fidelity latent route representations,effectively mitigating the data sparsity problem.Recommendations are then formulated using an Anchor-and-Expand algorithm,which selects relevant starting Points of Interest(POI)based on user history,then expands routes through latent similarity matching and geographic coherence optimization,culminating in Traveling Salesman Problem(TSP)-based route optimization for practical travel distances.Experiments on a real-world check-in dataset validate our model’s unique generative capability,achieving F1 scores ranging from 0.163 to 0.305,and near-zero pairs−F1 scores between 0.002 and 0.022.These results confirm the model’s success in generating novel travel routes by recommending new locations and sequences rather than replicating users’past itineraries.This work provides a robust solution for personalized travel planning,capable of generating novel and compelling routes for both new and existing users by learning from collective travel intelligence.展开更多
Recommendation systems have become indispensable for providing tailored suggestions and capturing evolving user preferences based on interaction histories.The collaborative filtering(CF)model,which depends exclusively...Recommendation systems have become indispensable for providing tailored suggestions and capturing evolving user preferences based on interaction histories.The collaborative filtering(CF)model,which depends exclusively on user-item interactions,commonly encounters challenges,including the cold-start problem and an inability to effectively capture the sequential and temporal characteristics of user behavior.This paper introduces a personalized recommendation system that combines deep learning techniques with Bayesian Personalized Ranking(BPR)optimization to address these limitations.With the strong support of Long Short-Term Memory(LSTM)networks,we apply it to identify sequential dependencies of user behavior and then incorporate an attention mechanism to improve the prioritization of relevant items,thereby enhancing recommendations based on the hybrid feedback of the user and its interaction patterns.The proposed system is empirically evaluated using publicly available datasets from movie and music,and we evaluate the performance against standard recommendation models,including Popularity,BPR,ItemKNN,FPMC,LightGCN,GRU4Rec,NARM,SASRec,and BERT4Rec.The results demonstrate that our proposed framework consistently achieves high outcomes in terms of HitRate,NDCG,MRR,and Precision at K=100,with scores of(0.6763,0.1892,0.0796,0.0068)on MovieLens-100K,(0.6826,0.1920,0.0813,0.0068)on MovieLens-1M,and(0.7937,0.3701,0.2756,0.0078)on Last.fm.The results show an average improvement of around 15%across all metrics compared to existing sequence models,proving that our framework ranks and recommends items more accurately.展开更多
Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstan...Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstanding performance across various domains,thereby prompting researchers to investigate their applicability in recommendation systems.However,due to the lack of task-specific knowledge and an inefficient feature extraction process,LLMs still have suboptimal performance in recommendation tasks.Therefore,external knowledge sources,such as knowledge graphs(KGs)and knowledge bases(KBs),are often introduced to address the issue of data sparsity.Compared to KGs,KBs possess higher retrieval efficiency,making them more suitable for scenarios where LLMs serve as recommenders.To this end,we introduce a novel framework integrating LLMs with KBs for enhanced retrieval generation,namely LLMKB.LLMKB initially leverages structured knowledge to create mapping dictionaries,extracting entity-relation information from heterogeneous knowledge to construct KBs.Then,LLMKB achieves the embedding calibration between user information representations and documents in KBs through retrieval model fine-tuning.Finally,LLMKB employs retrievalaugmented generation to produce recommendations based on fused text inputs,followed by post-processing.Experiment results on two public CRS datasets demonstrate the effectiveness of our framework.Our code is publicly available at the link:https://anonymous.4open.science/r/LLMKB-6FD0.展开更多
Attending one of China’s most prestigious institutions of higher learning, Peking University (PKU), is a dream for the vast majority of China’s middle school graduates. On November 16, the university released a list...Attending one of China’s most prestigious institutions of higher learning, Peking University (PKU), is a dream for the vast majority of China’s middle school graduates. On November 16, the university released a list of 39 senior middle schools across the country whose principals have展开更多
Abstract Instagram is a popular photo-sharing social ap- plication. It is widely used by tourists to record their journey information such as location, time and interest. Consequently, a huge volume of get-tagged phot...Abstract Instagram is a popular photo-sharing social ap- plication. It is widely used by tourists to record their journey information such as location, time and interest. Consequently, a huge volume of get-tagged photos with spatio-temporal in- formation are generated along tourist＇s travel trajectories. Such Instagram photo trajectories consist of travel paths, travel density distributions, and traveller behaviors, prefer- ences, and mobility patterns. Mining Instagram photo trajec- tories is thus very useful for many mobile and location-based social applications, including tour guide and recommender systems. However, we have not found any work that extracts interesting group-like travel trajectories from Instagram pho- tos asynchronously taken by different tourists. Motivated by this, we propose a novel concept： coterie, which reveals representative travel trajectory patterns hidden in Instagram photos taken by users at shared locations and paths. Our work includes the discovery of （1） coteries, （2） closed co- teries, and （3） the recommendation of popular travel routes based on closed coteries. For this, we first build a statistically reliable trajectory database from Instagram get-tagged pho- tos. These trajectories are then clustered by the DBSCAN method to find tourist density. Next, we transform each raw spatio-temporal trajectory into a sequence of clusters. All dis- criminative closed coteries are further identified by a Cluster- Growth algorithm. Finally, distance-aware and conformity- aware recommendation strategies are applied on closed co- teries to recommend popular tour routes. Visualized demosand extensive experimental results demonstrate the effective- ness and efficiency of our methods.展开更多
Nowadays,software engineers use a variety of online media to search and become informed of new and interesting technologies,and to learn from and help one another.We refer to these kinds of online media which help sof...Nowadays,software engineers use a variety of online media to search and become informed of new and interesting technologies,and to learn from and help one another.We refer to these kinds of online media which help software engineers improve their performance in software development,maintenance,and test processes as software information sites.In this paper,we propose TagCombine,an automatic tag recommendation method which analyzes objects in software information sites.TagCombine has three different components:1)multi-label ranking component which considers tag recommendation as a multi-label learning problem;2)similarity-based ranking component which recommends tags from similar objects;3)tag-term based ranking component which considers the relationship between different terms and tags,and recommends tags after analyzing the terms in the objects.We evaluate TagCombine on four software information sites,Ask Different,Ask Ubuntu,Feecode,and Stack Overflow.On averaging across the four projects,TagCombine achieves recall@5 and recallS10 to 0.6198 and 0.7625 respectively,which improves TagRec proposed by Al-Kofahi et al.by 14.56%and 10.55%respectively,and the tag recommendation method proposed by Zangerle et al.by 12.08%and 8.16%respectively.展开更多
Azoospermia,defined as the absence of sperm in the ejaculate,is a well-documented consequence of exogenous testosterone(ET)and anabolic–androgenic steroid(AAS)use.These agents suppress the hypothalamic–pituitary–go...Azoospermia,defined as the absence of sperm in the ejaculate,is a well-documented consequence of exogenous testosterone(ET)and anabolic–androgenic steroid(AAS)use.These agents suppress the hypothalamic–pituitary–gonadal(HPG)axis,leading to reduced intratesticular testosterone levels and impaired spermatogenesis.This review examines the pathophysiological mechanisms underlying azoospermia and outlines therapeutic strategies for recovery.Azoospermia is categorized into pretesticular,testicular,and post-testicular types,with a focus on personalized treatment approaches based on the degree of HPG axis suppression and baseline testicular function.Key strategies include discontinuing ET and monitoring for spontaneous recovery,particularly in patients with shorter durations of ET use.For cases of persistent azoospermia,gonadotropins(human chorionic gonadotropin[hCG]and follicle-stimulating hormone[FSH])and selective estrogen receptor modulators(SERMs),such as clomiphene citrate,are recommended,either alone or in combination.The global increase in exogenous testosterone use,including testosterone replacement therapy and AAS,underscores the need for improved management of associated azoospermia,which can be temporary or permanent depending on individual factors and the type of testosterone used.Additionally,the manuscript discusses preventive strategies,such as transitioning to short-acting testosterone formulations or incorporating low-dose hCG to preserve fertility during ET therapy.While guidelines for managing testosterone-related azoospermia remain limited,emerging research indicates the potential efficacy of hormonal stimulation therapies.However,there is a notable lack of well-structured,controlled,and long-term studies addressing the management of azoospermia related to exogenous testosterone use,highlighting the need for such studies to inform evidence-based recommendations.展开更多
A personalized outfit recommendation has emerged as a hot research topic in the fashion domain.However,existing recommendations do not fully exploit user style preferences.Typically,users prefer particular styles such...A personalized outfit recommendation has emerged as a hot research topic in the fashion domain.However,existing recommendations do not fully exploit user style preferences.Typically,users prefer particular styles such as casual and athletic styles,and consider attributes like color and texture when selecting outfits.To achieve personalized outfit recommendations in line with user style preferences,this paper proposes a personal style guided outfit recommendation with multi-modal fashion compatibility modeling,termed as PSGNet.Firstly,a style classifier is designed to categorize fashion images of various clothing types and attributes into distinct style categories.Secondly,a personal style prediction module extracts user style preferences by analyzing historical data.Then,to address the limitations of single-modal representations and enhance fashion compatibility,both fashion images and text data are leveraged to extract multi-modal features.Finally,PSGNet integrates these components through Bayesian personalized ranking(BPR)to unify the personal style and fashion compatibility,where the former is used as personal style features and guides the output of the personalized outfit recommendation tailored to the target user.Extensive experiments on large-scale datasets demonstrate that the proposed model is efficient on the personalized outfit recommendation.展开更多
The rapid development of short video platforms poses new challenges for traditional recommendation systems.Recommender systems typically depend on two types of user behavior feedback to construct user interest profile...The rapid development of short video platforms poses new challenges for traditional recommendation systems.Recommender systems typically depend on two types of user behavior feedback to construct user interest profiles:explicit feedback(interactive behavior),which significantly influences users’short-term interests,and implicit feedback(viewing time),which substantially affects their long-term interests.However,the previous model fails to distinguish between these two feedback methods,leading it to predict only the overall preferences of users based on extensive historical behavior sequences.Consequently,it cannot differentiate between users’long-term and shortterm interests,resulting in low accuracy in describing users’interest states and predicting the evolution of their interests.This paper introduces a video recommendationmodel calledCAT-MFRec(CrossAttention Transformer-Mixed Feedback Recommendation)designed to differentiate between explicit and implicit user feedback within the DIEN(Deep Interest Evolution Network)framework.This study emphasizes the separate learning of the two types of behavioral feedback,effectively integrating them through the cross-attention mechanism.Additionally,it leverages the long sequence dependence capabilities of Transformer technology to accurately construct user interest profiles and predict the evolution of user interests.Experimental results indicate that CAT-MF Rec significantly outperforms existing recommendation methods across various performance indicators.This advancement offers new theoretical and practical insights for the development of video recommendations,particularly in addressing complex and dynamic user behavior patterns.展开更多
In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performan...In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performance of recommendation systems.Due to the imbalance,machine learning algorithms tend to classify inputs into the positive(majority)class every time to achieve high prediction accuracy.Imbalance can be categorized such as by features and classes,but most studies consider only class imbalance.In this paper,we propose a recommendation system that can integrate multiple networks to adapt to a large number of imbalanced features and can deal with highly skewed and imbalanced datasets through a loss function.We propose a loss aware feature attention mechanism(LAFAM)to solve the issue of feature imbalance.The network incorporates an attention mechanism and uses multiple sub-networks to classify and learn features.For better results,the network can learn the weights of sub-networks and assign higher weights to important features.We propose suppression loss to address class imbalance,which favors negative loss by penalizing positive loss,and pays more attention to sample points near the decision boundary.Experiments on two large-scale datasets verify that the performance of the proposed system is greatly improved compared to baseline methods.展开更多
Mycophenolic acid(MPA),the active moiety of both mycophenolate mofetil(MMF)and enteric-coated mycophenolate sodium(EC-MPS),serves as a primary immunosuppressant for maintaining solid organ transplants.Therapeutic drug...Mycophenolic acid(MPA),the active moiety of both mycophenolate mofetil(MMF)and enteric-coated mycophenolate sodium(EC-MPS),serves as a primary immunosuppressant for maintaining solid organ transplants.Therapeutic drug monitoring(TDM)enhances treatment outcomes through tailored approaches.This study aimed to develop an evidence-based guideline for MPA TDM,facilitating its rational application in clinical settings.The guideline plan was drawn from the Institute of Medicine and World Health Organization(WHO)guidelines.Using the Delphi method,clinical questions and outcome indicators were generated.Systematic reviews,Grading of Recommendations Assessment,Development,and Evaluation(GRADE)evidence quality evaluations,expert opinions,and patient values guided evidence-based suggestions for the guideline.External reviews further refined the recommendations.The guideline for the TDM of MPA(IPGRP-2020CN099)consists of four sections and 16 recommendations encompassing target populations,monitoring strategies,dosage regimens,and influencing factors.High-risk populations,timing of TDM,area under the curve(AUC)versus trough concentration(C0),target concentration ranges,monitoring frequency,and analytical methods are addressed.Formulation-specific recommendations,initial dosage regimens,populations with unique considerations,pharmacokinetic-informed dosing,body weight factors,pharmacogenetics,and drug–drug interactions are covered.The evidence-based guideline offers a comprehensive recommendation for solid organ transplant recipients undergoing MPA therapy,promoting standardization of MPA TDM,and enhancing treatment efficacy and safety.展开更多
基金National Key Basic Research Program of China (973 Program) under Grant No.2012CB315802 and No.2013CB329102.National Natural Science Foundation of China under Grant No.61171102 and No.61132001.New generation broadband wireless mobile communication network Key Projects for Science and Technology Development under Grant No.2011ZX03002-002-01,Beijing Nova Program under Grant No.2008B50 and Beijing Higher Education Young Elite Teacher Project under Grant No.YETP0478
文摘Differently from the general online social network(OSN),locationbased mobile social network(LMSN),which seamlessly integrates mobile computing and social computing technologies,has unique characteristics of temporal,spatial and social correlation.Recommending friends instantly based on current location of users in the real world has become increasingly popular in LMSN.However,the existing friend recommendation methods based on topological structures of a social network or non-topological information such as similar user profiles cannot well address the instant making friends in the real world.In this article,we analyze users' check-in behavior in a real LMSN site named Gowalla.According to this analysis,we present an approach of recommending friends instantly for LMSN users by considering the real-time physical location proximity,offline behavior similarity and friendship network information in the virtual community simultaneously.This approach effectively bridges the gap between the offline behavior of users in the real world and online friendship network information in the virtual community.Finally,we use the real user check-in dataset of Gowalla to verify the effectiveness of our approach.
基金supported by the grants from Natural Science Foundation of China (Project No.61471060)
文摘Citation network is often used for academic recommendation. However, it is difficult to achieve high recommendation accuracy and low time complexity because it is often very large and sparse and different citations have different purposes. What's more, some citations include unreasonable information, such as in case of intentional self-citation. To improve the accuracy of citation network-based academic recommendation and reduce the time complexity, we propose an academic recommendation method for recommending authors and papers. In which, an author-paper bilayer citation network is built, then an enhanced topic model, Author Community Topic Time Model(ACTTM) is proposed to detect high quality author communities in the author layer, and a set of attributes are proposed to comprehensively depict the author/paper nodes in the bilayer citation network. Experimental results prove that the proposed ACTTM can detect high quality author communities and facilitate low time complexity, and the proposed academic recommendation method can effectively improve the recommendation accuracy.
基金supported by the National Key Research and Development Program(No.2016YFB0800302)
文摘With the development of the Internet of Things(Io T), people's lives have become increasingly convenient. It is desirable for smart home(SH) systems to integrate and leverage the enormous information available from IoT. Information can be analyzed to learn user intentions and automatically provide the appropriate services. However, existing service recommendation models typically do not consider the services that are unavailable in a user's living environment. In order to address this problem, we propose a series of semantic models for SH devices. These semantic models can be used to infer user intentions. Based on the models, we proposed a service recommendation probability model and an alternative-service recommending algorithm. The algorithm is devoted to providing appropriate alternative services when the desired service is unavailable. The algorithm has been implemented and achieves accuracy higher than traditional Hidden Markov Model(HMM). The maximum accuracy achieved is 68.3%.
文摘In this paper, we examine methods that can provide accurate results in a form of a recommender system within a social networking framework. The social networking site of choice is Twitter, due to its interesting social graph connections and content characteristics. We built a recommender system which recommends potential users to follow by analyzing their tweets using the CRM114 regex engine as a basis for content classification. The evaluation of the recommender system was based on a dataset generated from real Twitter users created in late 2009.
文摘Location based social networks（ LBSNs） provide location specific data generated from smart phone into online social networks thus people can share their points of interest（ POIs）. POI collections are complex and can be influenced by various factors,such as user preferences,social relationships and geographical influence. Therefore,recommending new locations in LBSNs requires to take all these factors into consideration. However,one problem is how to determine optimal weights of influencing factors in an algorithm in which these factors are combined. The user similarity can be obtained from the user check-in data,or from the user friend information,or based on the different geographical influences on each user＇s check-in activities. In this paper,we propose an algorithm that calculates the user similarity based on check-in records and social relationships,using a proposed weighting function to adjust the weights of these two kinds of similarities based on the geographical distance between users. In addition,a non-parametric density estimation method is applied to predict the unique geographical influence on each user by getting the density probability plot of the distance between every pair of user＇s check-in locations. Experimental results,using foursquare datasets,have shown that comparisons between the proposed algorithm and the other five baseline recommendation algorithms in LBSNs demonstrate that our proposed algorithm is superior in accuracy and recall,furthermore solving the sparsity problem.
文摘Depression is a common disorder worldwide. In the West, depression is also called ＂the blues＂ informally. It is a psychological disorder characterized by an all-encompassing low mood. Slightly depressed people may behave normally, while experiencing feelings of pain at heart. In mild depression cases, people may feel gloomy or distressed, which is usually associated with memory impairment, insomnia,
文摘The performance of deep recommendation models degrades significantly under data poisoning attacks.While adversarial training methods such as Vulnerability-Aware Training(VAT)enhance robustness by injecting perturbations into embeddings,they remain limited by coarse-grained noise and a static defense strategy,leaving models susceptible to adaptive attacks.This study proposes a novel framework,Self-Purification Data Sanitization(SPD),which integrates vulnerability-aware adversarial training with dynamic label correction.Specifically,SPD first identifies high-risk users through a fragility scoring mechanism,then applies self-purification by replacing suspicious interactions with model-predicted high-confidence labels during training.This closed-loop process continuously sanitizes the training data and breaks the protection ceiling of conventional adversarial training.Experiments demonstrate that SPD significantly improves the robustness of both Matrix Factorization(MF)and LightGCN models against various poisoning attacks.We show that SPD effectively suppresses malicious gradient propagation and maintains recommendation accuracy.Evaluations on Gowalla and Yelp2018 confirmthat SPD-trainedmodels withstandmultiple attack strategies—including Random,Bandwagon,DP,and Rev attacks—while preserving performance.
文摘Contemporary Introduction to Chinese Medicine in Comparison with Western Medicine,recently published by China,is exceptional and worthwhile reading,especially for Western readers.It is authored by Professor Xie Zhu-fan in collaboration with Dr. Xie Fang.Professor Xie is a renowned scholar of both Western and Chinese medicine in China,and the Director Emeritus of Peking University Institute of integrated Chinese and Western Medicine.
基金supported by the National Key R&D Program of China[2022YFF0902703]the State Administration for Market Regulation Science and Technology Plan Project(2024MK033).
文摘Recommendation systems are key to boosting user engagement,satisfaction,and retention,particularly on media platforms where personalized content is vital.Sequential recommendation systems learn from user-item interactions to predict future items of interest.However,many current methods rely on unique user and item IDs,limiting their ability to represent users and items effectively,especially in zero-shot learning scenarios where training data is scarce.With the rapid development of Large Language Models(LLMs),researchers are exploring their potential to enhance recommendation systems.However,there is a semantic gap between the linguistic semantics of LLMs and the collaborative semantics of recommendation systems,where items are typically indexed by IDs.Moreover,most research focuses on item representations,neglecting personalized user modeling.To address these issues,we propose a sequential recommendation framework using LLMs,called CIT-Rec,a model that integrates Collaborative semantics for user representation and Image and Text information for item representation to enhance Recommendations.Specifically,by aligning intuitive image information with text containing semantic features,we can more accurately represent items,improving item representation quality.We focus not only on item representations but also on user representations.To more precisely capture users’personalized preferences,we use traditional sequential recommendation models to train on users’historical interaction data,effectively capturing behavioral patterns.Finally,by combining LLMs and traditional sequential recommendation models,we allow the LLM to understand linguistic semantics while capturing collaborative semantics.Extensive evaluations on real-world datasets show that our model outperforms baseline methods,effectively combining user interaction history with item visual and textual modalities to provide personalized recommendations.
基金supported by the Chung-Ang University Research Grants in 2023.Alsothe work is supported by the ELLIIT Excellence Center at Linköping–Lund in Information Technology in Sweden.
文摘Recommending personalized travel routes from sparse,implicit feedback poses a significant challenge,as conventional systems often struggle with information overload and fail to capture the complex,sequential nature of user preferences.To address this,we propose a Conditional Generative Adversarial Network(CGAN)that generates diverse and highly relevant itineraries.Our approach begins by constructing a conditional vector that encapsulates a user’s profile.This vector uniquely fuses embeddings from a Heterogeneous Information Network(HIN)to model complex user-place-route relationships,a Recurrent Neural Network(RNN)to capture sequential path dynamics,and Neural Collaborative Filtering(NCF)to incorporate collaborative signals from the wider user base.This comprehensive condition,further enhanced with features representing user interaction confidence and uncertainty,steers a CGAN stabilized by spectral normalization to generate high-fidelity latent route representations,effectively mitigating the data sparsity problem.Recommendations are then formulated using an Anchor-and-Expand algorithm,which selects relevant starting Points of Interest(POI)based on user history,then expands routes through latent similarity matching and geographic coherence optimization,culminating in Traveling Salesman Problem(TSP)-based route optimization for practical travel distances.Experiments on a real-world check-in dataset validate our model’s unique generative capability,achieving F1 scores ranging from 0.163 to 0.305,and near-zero pairs−F1 scores between 0.002 and 0.022.These results confirm the model’s success in generating novel travel routes by recommending new locations and sequences rather than replicating users’past itineraries.This work provides a robust solution for personalized travel planning,capable of generating novel and compelling routes for both new and existing users by learning from collective travel intelligence.
基金funded by Soonchunhyang University,Grant Number 20250029。
文摘Recommendation systems have become indispensable for providing tailored suggestions and capturing evolving user preferences based on interaction histories.The collaborative filtering(CF)model,which depends exclusively on user-item interactions,commonly encounters challenges,including the cold-start problem and an inability to effectively capture the sequential and temporal characteristics of user behavior.This paper introduces a personalized recommendation system that combines deep learning techniques with Bayesian Personalized Ranking(BPR)optimization to address these limitations.With the strong support of Long Short-Term Memory(LSTM)networks,we apply it to identify sequential dependencies of user behavior and then incorporate an attention mechanism to improve the prioritization of relevant items,thereby enhancing recommendations based on the hybrid feedback of the user and its interaction patterns.The proposed system is empirically evaluated using publicly available datasets from movie and music,and we evaluate the performance against standard recommendation models,including Popularity,BPR,ItemKNN,FPMC,LightGCN,GRU4Rec,NARM,SASRec,and BERT4Rec.The results demonstrate that our proposed framework consistently achieves high outcomes in terms of HitRate,NDCG,MRR,and Precision at K=100,with scores of(0.6763,0.1892,0.0796,0.0068)on MovieLens-100K,(0.6826,0.1920,0.0813,0.0068)on MovieLens-1M,and(0.7937,0.3701,0.2756,0.0078)on Last.fm.The results show an average improvement of around 15%across all metrics compared to existing sequence models,proving that our framework ranks and recommends items more accurately.
文摘Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstanding performance across various domains,thereby prompting researchers to investigate their applicability in recommendation systems.However,due to the lack of task-specific knowledge and an inefficient feature extraction process,LLMs still have suboptimal performance in recommendation tasks.Therefore,external knowledge sources,such as knowledge graphs(KGs)and knowledge bases(KBs),are often introduced to address the issue of data sparsity.Compared to KGs,KBs possess higher retrieval efficiency,making them more suitable for scenarios where LLMs serve as recommenders.To this end,we introduce a novel framework integrating LLMs with KBs for enhanced retrieval generation,namely LLMKB.LLMKB initially leverages structured knowledge to create mapping dictionaries,extracting entity-relation information from heterogeneous knowledge to construct KBs.Then,LLMKB achieves the embedding calibration between user information representations and documents in KBs through retrieval model fine-tuning.Finally,LLMKB employs retrievalaugmented generation to produce recommendations based on fused text inputs,followed by post-processing.Experiment results on two public CRS datasets demonstrate the effectiveness of our framework.Our code is publicly available at the link:https://anonymous.4open.science/r/LLMKB-6FD0.
文摘Attending one of China’s most prestigious institutions of higher learning, Peking University (PKU), is a dream for the vast majority of China’s middle school graduates. On November 16, the university released a list of 39 senior middle schools across the country whose principals have
文摘Abstract Instagram is a popular photo-sharing social ap- plication. It is widely used by tourists to record their journey information such as location, time and interest. Consequently, a huge volume of get-tagged photos with spatio-temporal in- formation are generated along tourist＇s travel trajectories. Such Instagram photo trajectories consist of travel paths, travel density distributions, and traveller behaviors, prefer- ences, and mobility patterns. Mining Instagram photo trajec- tories is thus very useful for many mobile and location-based social applications, including tour guide and recommender systems. However, we have not found any work that extracts interesting group-like travel trajectories from Instagram pho- tos asynchronously taken by different tourists. Motivated by this, we propose a novel concept： coterie, which reveals representative travel trajectory patterns hidden in Instagram photos taken by users at shared locations and paths. Our work includes the discovery of （1） coteries, （2） closed co- teries, and （3） the recommendation of popular travel routes based on closed coteries. For this, we first build a statistically reliable trajectory database from Instagram get-tagged pho- tos. These trajectories are then clustered by the DBSCAN method to find tourist density. Next, we transform each raw spatio-temporal trajectory into a sequence of clusters. All dis- criminative closed coteries are further identified by a Cluster- Growth algorithm. Finally, distance-aware and conformity- aware recommendation strategies are applied on closed co- teries to recommend popular tour routes. Visualized demosand extensive experimental results demonstrate the effective- ness and efficiency of our methods.
文摘Nowadays,software engineers use a variety of online media to search and become informed of new and interesting technologies,and to learn from and help one another.We refer to these kinds of online media which help software engineers improve their performance in software development,maintenance,and test processes as software information sites.In this paper,we propose TagCombine,an automatic tag recommendation method which analyzes objects in software information sites.TagCombine has three different components:1)multi-label ranking component which considers tag recommendation as a multi-label learning problem;2)similarity-based ranking component which recommends tags from similar objects;3)tag-term based ranking component which considers the relationship between different terms and tags,and recommends tags after analyzing the terms in the objects.We evaluate TagCombine on four software information sites,Ask Different,Ask Ubuntu,Feecode,and Stack Overflow.On averaging across the four projects,TagCombine achieves recall@5 and recallS10 to 0.6198 and 0.7625 respectively,which improves TagRec proposed by Al-Kofahi et al.by 14.56%and 10.55%respectively,and the tag recommendation method proposed by Zangerle et al.by 12.08%and 8.16%respectively.
文摘Azoospermia,defined as the absence of sperm in the ejaculate,is a well-documented consequence of exogenous testosterone(ET)and anabolic–androgenic steroid(AAS)use.These agents suppress the hypothalamic–pituitary–gonadal(HPG)axis,leading to reduced intratesticular testosterone levels and impaired spermatogenesis.This review examines the pathophysiological mechanisms underlying azoospermia and outlines therapeutic strategies for recovery.Azoospermia is categorized into pretesticular,testicular,and post-testicular types,with a focus on personalized treatment approaches based on the degree of HPG axis suppression and baseline testicular function.Key strategies include discontinuing ET and monitoring for spontaneous recovery,particularly in patients with shorter durations of ET use.For cases of persistent azoospermia,gonadotropins(human chorionic gonadotropin[hCG]and follicle-stimulating hormone[FSH])and selective estrogen receptor modulators(SERMs),such as clomiphene citrate,are recommended,either alone or in combination.The global increase in exogenous testosterone use,including testosterone replacement therapy and AAS,underscores the need for improved management of associated azoospermia,which can be temporary or permanent depending on individual factors and the type of testosterone used.Additionally,the manuscript discusses preventive strategies,such as transitioning to short-acting testosterone formulations or incorporating low-dose hCG to preserve fertility during ET therapy.While guidelines for managing testosterone-related azoospermia remain limited,emerging research indicates the potential efficacy of hormonal stimulation therapies.However,there is a notable lack of well-structured,controlled,and long-term studies addressing the management of azoospermia related to exogenous testosterone use,highlighting the need for such studies to inform evidence-based recommendations.
基金Shanghai Frontier Science Research Center for Modern Textiles,Donghua University,ChinaOpen Project of Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment,Zhengzhou University of Light Industry,China(No.IM202303)National Key Research and Development Program of China(No.2019YFB1706300)。
文摘A personalized outfit recommendation has emerged as a hot research topic in the fashion domain.However,existing recommendations do not fully exploit user style preferences.Typically,users prefer particular styles such as casual and athletic styles,and consider attributes like color and texture when selecting outfits.To achieve personalized outfit recommendations in line with user style preferences,this paper proposes a personal style guided outfit recommendation with multi-modal fashion compatibility modeling,termed as PSGNet.Firstly,a style classifier is designed to categorize fashion images of various clothing types and attributes into distinct style categories.Secondly,a personal style prediction module extracts user style preferences by analyzing historical data.Then,to address the limitations of single-modal representations and enhance fashion compatibility,both fashion images and text data are leveraged to extract multi-modal features.Finally,PSGNet integrates these components through Bayesian personalized ranking(BPR)to unify the personal style and fashion compatibility,where the former is used as personal style features and guides the output of the personalized outfit recommendation tailored to the target user.Extensive experiments on large-scale datasets demonstrate that the proposed model is efficient on the personalized outfit recommendation.
基金supported by National Natural Science Foundation of China(62072416)Key Research and Development Special Project of Henan Province(221111210500)Key TechnologiesR&DProgram of Henan rovince(232102211053,242102211071).
文摘The rapid development of short video platforms poses new challenges for traditional recommendation systems.Recommender systems typically depend on two types of user behavior feedback to construct user interest profiles:explicit feedback(interactive behavior),which significantly influences users’short-term interests,and implicit feedback(viewing time),which substantially affects their long-term interests.However,the previous model fails to distinguish between these two feedback methods,leading it to predict only the overall preferences of users based on extensive historical behavior sequences.Consequently,it cannot differentiate between users’long-term and shortterm interests,resulting in low accuracy in describing users’interest states and predicting the evolution of their interests.This paper introduces a video recommendationmodel calledCAT-MFRec(CrossAttention Transformer-Mixed Feedback Recommendation)designed to differentiate between explicit and implicit user feedback within the DIEN(Deep Interest Evolution Network)framework.This study emphasizes the separate learning of the two types of behavioral feedback,effectively integrating them through the cross-attention mechanism.Additionally,it leverages the long sequence dependence capabilities of Transformer technology to accurately construct user interest profiles and predict the evolution of user interests.Experimental results indicate that CAT-MF Rec significantly outperforms existing recommendation methods across various performance indicators.This advancement offers new theoretical and practical insights for the development of video recommendations,particularly in addressing complex and dynamic user behavior patterns.
基金supported by the National Key Research and Development Program of China(Grant numbers:2021YFF0901705,2021YFF0901700)the State Key Laboratory of Media Convergence and Communication,Communication University of China+1 种基金the Fundamental Research Funds for the Central Universitiesthe High-Quality and Cutting-Edge Disciplines Construction Project for Universities in Beijing(Internet Information,Communication University of China).
文摘In the Internet era,recommendation systems play a crucial role in helping users find relevant information from large datasets.Class imbalance is known to severely affect data quality,and therefore reduce the performance of recommendation systems.Due to the imbalance,machine learning algorithms tend to classify inputs into the positive(majority)class every time to achieve high prediction accuracy.Imbalance can be categorized such as by features and classes,but most studies consider only class imbalance.In this paper,we propose a recommendation system that can integrate multiple networks to adapt to a large number of imbalanced features and can deal with highly skewed and imbalanced datasets through a loss function.We propose a loss aware feature attention mechanism(LAFAM)to solve the issue of feature imbalance.The network incorporates an attention mechanism and uses multiple sub-networks to classify and learn features.For better results,the network can learn the weights of sub-networks and assign higher weights to important features.We propose suppression loss to address class imbalance,which favors negative loss by penalizing positive loss,and pays more attention to sample points near the decision boundary.Experiments on two large-scale datasets verify that the performance of the proposed system is greatly improved compared to baseline methods.
基金supported by the National Natural Science Foundation of China(NSFC)(No.72304007)the Huatong Guokang Medical Research Fund(No.2023HT010)。
文摘Mycophenolic acid(MPA),the active moiety of both mycophenolate mofetil(MMF)and enteric-coated mycophenolate sodium(EC-MPS),serves as a primary immunosuppressant for maintaining solid organ transplants.Therapeutic drug monitoring(TDM)enhances treatment outcomes through tailored approaches.This study aimed to develop an evidence-based guideline for MPA TDM,facilitating its rational application in clinical settings.The guideline plan was drawn from the Institute of Medicine and World Health Organization(WHO)guidelines.Using the Delphi method,clinical questions and outcome indicators were generated.Systematic reviews,Grading of Recommendations Assessment,Development,and Evaluation(GRADE)evidence quality evaluations,expert opinions,and patient values guided evidence-based suggestions for the guideline.External reviews further refined the recommendations.The guideline for the TDM of MPA(IPGRP-2020CN099)consists of four sections and 16 recommendations encompassing target populations,monitoring strategies,dosage regimens,and influencing factors.High-risk populations,timing of TDM,area under the curve(AUC)versus trough concentration(C0),target concentration ranges,monitoring frequency,and analytical methods are addressed.Formulation-specific recommendations,initial dosage regimens,populations with unique considerations,pharmacokinetic-informed dosing,body weight factors,pharmacogenetics,and drug–drug interactions are covered.The evidence-based guideline offers a comprehensive recommendation for solid organ transplant recipients undergoing MPA therapy,promoting standardization of MPA TDM,and enhancing treatment efficacy and safety.
