Under the background of the Judicial Reform of China, big data of judicial cases are widely used to solve the problem of judicial research. Similarity analysis of judicial cases is the basis of wisdom judicature. In v...Under the background of the Judicial Reform of China, big data of judicial cases are widely used to solve the problem of judicial research. Similarity analysis of judicial cases is the basis of wisdom judicature. In view of the necessity of getting rid of the ineffective information and extracting useful rules and conditions from the descriptive document, the analysis of Chinese judicial cases with a certain format is a big challenge. Hence, we propose a method that focuses on producing recommendations that are based on the content of judicial cases. Considering the particularity of Chinese language, we use “jieba” text segmentation to preprocess the cases. In view of the lack of labels of user interest and behavior, the proposed method considers the content information via adopting TF-IDF combined with LDA topic model, as opposed to the traditional methods such as CF (Collaborative Filtering Recommendations). Users are recommended to compute cosine similarity of cases in the same topic. In the experiments, we evaluate the performance of the proposed model on a given dataset of nearly 200,000 judicial cases. The experimental result reveals when the number of topics is around 80, the proposed method gets the best performance.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
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
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
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
The content-basedfiltering technique has been used effectively in a variety of Recommender Systems(RS).The user explicitly or implicitly provides data in the Content-Based Recommender System.The system collects this da...The content-basedfiltering technique has been used effectively in a variety of Recommender Systems(RS).The user explicitly or implicitly provides data in the Content-Based Recommender System.The system collects this data and creates a profile for all the users,and the recommendation is generated by the user profile.The recommendation generated via content-basedfiltering is provided by observing just a single user’s profile.The primary objective of this RS is to recommend a list of movies based on the user’s preferences.A con-tent-based movie recommendation model is proposed in this research,which recommends movies based on the user’s profile from the Facebook platform.The recommendation system is built with a hybrid model that combines the Mon-arch Butterfly Optimization(MBO)with the Deep Belief Network(DBN).For feature selection,the MBO is utilized,while DBN is used for classification.The datasets used in the experiment are collected from Facebook and MovieLens.The dataset features are evaluated for performance evaluation to validate if data with various attributes can solve the matching recommendations.Eachfile is com-pared with features that prove the features will support movie recommendations.The proposed model’s mean absolute error(MAE)and root-mean-square error(RMSE)values are 0.716 and 0.915,and its precision and recall are 97.35 and 96.60 percent,respectively.Extensive tests have demonstrated the advantages of the proposed method in terms of MAE,RMSE,Precision,and Recall compared to state-of-the-art algorithms such as Fuzzy C-means with Bat algorithm(FCM-BAT),Collaborativefiltering with k-NN and the normalized discounted cumulative gain method(CF-kNN+NDCG),User profile correlation-based similarity(UPCSim),and Deep Autoencoder.展开更多
Content-based filtering E-commerce recommender system was discussed fully in this paper. Users＇ unique features can be explored by means of vector space model firstly. Then based on the qualitative value of products ...Content-based filtering E-commerce recommender system was discussed fully in this paper. Users＇ unique features can be explored by means of vector space model firstly. Then based on the qualitative value of products informa tion, the recommender lists were obtained. Since the system can adapt to the users＇ feedback automatically, its performance were enhanced comprehensively. Finally the evaluation of the system and the experimental results were presented.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
Recommender systems are very useful for people to explore what they really need.Academic papers are important achievements for researchers and they often have a great deal of choice to submit their papers.In order to ...Recommender systems are very useful for people to explore what they really need.Academic papers are important achievements for researchers and they often have a great deal of choice to submit their papers.In order to improve the efficiency of selecting the most suitable journals for publishing their works,journal recommender systems(JRS)can automatically provide a small number of candidate journals based on key information such as the title and the abstract.However,users or journal owners may attack the system for their own purposes.In this paper,we discuss about the adversarial attacks against content-based filtering JRS.We propose both targeted attack method that makes some target journals appear more often in the system and non-targeted attack method that makes the system provide incorrect recommendations.We also conduct extensive experiments to validate the proposed methods.We hope this paper could help improve JRS by realizing the existence of such adversarial attacks.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
This study presents a new approach that advances the algorithm of similarity measures between generalized fuzzy numbers. Following a brief introduction to some properties of the proposed method, a comparative analysis...This study presents a new approach that advances the algorithm of similarity measures between generalized fuzzy numbers. Following a brief introduction to some properties of the proposed method, a comparative analysis based on 36 sets of generalized fuzzy numbers was performed, in which the degree of similarity of the fuzzy numbers was calculated with the proposed method and seven methods established by previous studies in the literature. The results of the analytical comparison show that the proposed similarity outperforms the existing methods by overcoming their drawbacks and yielding accurate outcomes in all calculations of similarity measures under consideration. Finally, in a numerical example that involves recommending cars to customers based on a nine-member linguistic term set, the proposed similarity measure proves to be competent in addressing fuzzy number recommendation problems.展开更多
Session-based recommendation systems(SBR)are pivotal in suggesting items by analyzing anonymized sequences of user interactions.Traditional methods,while competent,often fall short in two critical areas:they fail to a...Session-based recommendation systems(SBR)are pivotal in suggesting items by analyzing anonymized sequences of user interactions.Traditional methods,while competent,often fall short in two critical areas:they fail to address potential inter-session item transitions,which are behavioral dependencies that extend beyond individual session boundaries,and they rely on monolithic item aggregation to construct session representations.This approach does not capture the multi-scale and heterogeneous nature of user intent,leading to a decrease in modeling accuracy.To overcome these limitations,a novel approach called HMGS has been introduced.This system incorporates dual graph architectures to enhance the recommendation process.A global transition graph captures latent cross-session item dependencies,while a heterogeneous intra-session graph encodesmulti-scale item embeddings through localized feature propagation.Additionally,amulti-tier graphmatchingmechanism aligns user preference signals across different granularities,significantly improving interest localization accuracy.Empirical validation on benchmark datasets(Tmall and Diginetica)confirms HMGS’s efficacy against state-of-the-art baselines.Quantitative analysis reveals performance gains of 20.54%and 12.63%in Precision@10 on Tmall and Diginetica,respectively.Consistent improvements are observed across auxiliary metrics,with MRR@10,Precision@20,and MRR@20 exhibiting enhancements between 4.00%and 21.36%,underscoring the framework’s robustness in multi-faceted recommendation scenarios.展开更多
In L2 content-based classrooms,code-switching or translanguaging seem to be a common practice adopted by teachers.There has been growing research discussing the potentials of L1 in these classrooms.Most of the current...In L2 content-based classrooms,code-switching or translanguaging seem to be a common practice adopted by teachers.There has been growing research discussing the potentials of L1 in these classrooms.Most of the current studies have focused on the analysis of lesson interactions and yet the perception of the content teachers has remained underexplored.This case study investigated the introspective views of a group of content teachers at a secondary school using questionnaires and written accounts.Data analyses showed that these teachers were generally aware of the interpersonal and ideational functions achieved by the use of L1 and they also seemed to have a positive view towards their practices of using L1 in English-medium classrooms.Based on the findings,practical implications for content teachers in relation to making medium of instruction decisions and suggestions for further research are discussed.展开更多
Recently,many Sequential Recommendation methods adopt self-attention mechanisms to model user preferences.However,these methods tend to focus more on low-frequency information while neglecting highfrequency informatio...Recently,many Sequential Recommendation methods adopt self-attention mechanisms to model user preferences.However,these methods tend to focus more on low-frequency information while neglecting highfrequency information,which makes them ineffective in balancing users’long-and short-term preferences.At the same time,manymethods overlook the potential of frequency domainmethods,ignoring their efficiency in processing frequency information.To overcome this limitation,we shift the focus to the combination of time and frequency domains and propose a novel Hybrid Time-Frequency Dual-Branch Transformer for Sequential Recommendation,namely HyTiFRec.Specifically,we design two hybrid filter modules:the learnable hybrid filter(LHF)and the window hybrid filter(WHF).We combine these with the Efficient Attention(EA)module to form the dual-branch structure to replace the self-attention components in Transformers.The EAmodule is used to extract sequential and global information.The LHF andWHF modules balance the proportion of different frequency bands,with LHF globally modulating the spectrum in the frequency domain and WHF retaining frequency components within specific local frequency bands.Furthermore,we use a time domain residual information addition operation in the hybrid filter module,which reduces information loss and further facilitates the hybrid of time-frequency methods.Extensive experiments on five widely-used real-world datasets show that our proposed method surpasses state-of-the-art methods.展开更多
A survey conducted on the premature bolting of Huarong large leaf mustard from 2018 to 2024 revealed that Huarong large leaf mustard sown in middle August was associated with a higher propensity for premature bolting....A survey conducted on the premature bolting of Huarong large leaf mustard from 2018 to 2024 revealed that Huarong large leaf mustard sown in middle August was associated with a higher propensity for premature bolting. Furthermore, it was observed that the earlier being sown, the greater the rate of premature bolting when being sown prior to middle August. The rate of premature bolting observed in seedlings sown on August 8 was recorded at 35.6%. It was noted that as the age of the seedlings increased, the rate of premature bolting correspondingly increased. There were notable differences in the tolerance of various cultivars to elevated temperatures and prolonged sunlight exposure. For instance, cultivars such as Zhangjie 1 and Sichuan Shaguodi, which exhibit greater heat resistance, did not demonstrate premature bolting when sown in early August. The prolonged exposure to elevated temperatures, drought conditions, and extended periods of sunlight during the seedling stage of Huarong large leaf mustard, coupled with delayed irrigation and transplantation, contributed to the occurrence of premature bolting. The Huarong large leaf mustard, when been sown from late August to early September and transplanted at the appropriate time, exhibited normal growth and development, with no instances of premature bolting observed. It is advisable to select heat-resistant varieties, such as Zhangjie 1, prior to middle August. Huarong large leaf mustard should be sown in early to middle September. Additionally, it is essential to ensure centralized production and timely release of seeds, prompt transplantation and harvesting, and enhance the management of pests and diseases.展开更多
As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as...As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as a new challenge in the field of recommendation systems.This paper introduces a group-buying recommendation model based on multi-head attention mechanisms and multi-task learning,termed the Multi-head Attention Mechanisms and Multi-task Learning Group-Buying Recommendation(MAMGBR)model,specifically designed to optimize group-buying recommendations on e-commerce platforms.The core dataset of this study comes from the Chinese maternal and infant e-commerce platform“Beibei,”encompassing approximately 430,000 successful groupbuying actions and over 120,000 users.Themodel focuses on twomain tasks:recommending items for group organizers(Task Ⅰ)and recommending participants for a given group-buying event(Task Ⅱ).In model evaluation,MAMGBR achieves an MRR@10 of 0.7696 for Task I,marking a 20.23%improvement over baseline models.Furthermore,in Task II,where complex interaction patterns prevail,MAMGBR utilizes auxiliary loss functions to effectively model the multifaceted roles of users,items,and participants,leading to a 24.08%increase in MRR@100 under a 1:99 sample ratio.Experimental results show that compared to benchmark models,such as NGCF and EATNN,MAMGBR’s integration ofmulti-head attentionmechanisms,expert networks,and gating mechanisms enables more accurate modeling of user preferences and social associations within group-buying scenarios,significantly enhancing recommendation accuracy and platform group-buying success rates.展开更多
This paper examines the impact of algorithmic recommendations and data-driven marketing on consumer engagement and business performance.By leveraging large volumes of user data,businesses can deliver personalized cont...This paper examines the impact of algorithmic recommendations and data-driven marketing on consumer engagement and business performance.By leveraging large volumes of user data,businesses can deliver personalized content that enhances user experiences and increases conversion rates.However,the growing reliance on these technologies introduces significant risks,including privacy violations,algorithmic bias,and ethical concerns.This paper explores these challenges and provides recommendations for businesses to mitigate associated risks while optimizing marketing strategies.It highlights the importance of transparency,fairness,and user control in ensuring responsible and effective data-driven marketing.展开更多
This special issue of the Asian Journal of Andrology is fully dedicated to the thematic area of non-obstructive azoospermia(NOA),one of the most complex and challenging conditions in the realm of andrology,urology,and...This special issue of the Asian Journal of Andrology is fully dedicated to the thematic area of non-obstructive azoospermia(NOA),one of the most complex and challenging conditions in the realm of andrology,urology,and reproductive medicine.展开更多
In the era of exponential growth of digital information,recommender algorithms are vital for helping users navigate vast data to find relevant items.Traditional approaches such as collaborative filtering and contentba...In the era of exponential growth of digital information,recommender algorithms are vital for helping users navigate vast data to find relevant items.Traditional approaches such as collaborative filtering and contentbasedmethods have limitations in capturing complex,multi-faceted relationships in large-scale,sparse datasets.Recent advances in Graph Neural Networks(GNNs)have significantly improved recommendation performance by modeling high-order connection patterns within user-item interaction networks.However,existing GNN-based models like LightGCN and NGCF focus primarily on single-type interactions and often overlook diverse semantic relationships,leading to reduced recommendation diversity and limited generalization.To address these challenges,this paper proposes a dual multi-relational graph neural network recommendation algorithm based on relational interactions.Our approach constructs two complementary graph structures:a User-Item Interaction Graph(UIIG),which explicitly models direct user behaviors such as clicks and purchases,and a Relational Association Graph(RAG),which uncovers latent associations based on user similarities and item attributes.The proposed Dual Multi-relational Graph Neural Network(DMGNN)features two parallel branches that perform multi-layer graph convolutional operations,followed by an adaptive fusion mechanism to effectively integrate information from both graphs.This design enhances the model’s capacity to capture diverse relationship types and complex relational patterns.Extensive experiments conducted on benchmark datasets—including MovieLens-1M,Amazon-Electronics,and Yelp—demonstrate thatDMGNN outperforms state-of-the-art baselines,achieving improvements of up to 12.3%in Precision,9.7%in Recall,and 11.5%in F1 score.Moreover,DMGNN significantly boosts recommendation diversity by 15.2%,balancing accuracy with exploration.These results highlight the effectiveness of leveraging hierarchical multi-relational information,offering a promising solution to the challenges of data sparsity and relation heterogeneity in recommendation systems.Our work advances the theoretical understanding of multi-relational graph modeling and presents practical insights for developing more personalized,diverse,and robust recommender systems.展开更多
基金the National Key Research and Development Program of China (2016YFC0800805)the National Natural Science Foundation of China (61772014).
文摘Under the background of the Judicial Reform of China, big data of judicial cases are widely used to solve the problem of judicial research. Similarity analysis of judicial cases is the basis of wisdom judicature. In view of the necessity of getting rid of the ineffective information and extracting useful rules and conditions from the descriptive document, the analysis of Chinese judicial cases with a certain format is a big challenge. Hence, we propose a method that focuses on producing recommendations that are based on the content of judicial cases. Considering the particularity of Chinese language, we use “jieba” text segmentation to preprocess the cases. In view of the lack of labels of user interest and behavior, the proposed method considers the content information via adopting TF-IDF combined with LDA topic model, as opposed to the traditional methods such as CF (Collaborative Filtering Recommendations). Users are recommended to compute cosine similarity of cases in the same topic. In the experiments, we evaluate the performance of the proposed model on a given dataset of nearly 200,000 judicial cases. The experimental result reveals when the number of topics is around 80, the proposed method gets the best performance.
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.
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.
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.
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The content-basedfiltering technique has been used effectively in a variety of Recommender Systems(RS).The user explicitly or implicitly provides data in the Content-Based Recommender System.The system collects this data and creates a profile for all the users,and the recommendation is generated by the user profile.The recommendation generated via content-basedfiltering is provided by observing just a single user’s profile.The primary objective of this RS is to recommend a list of movies based on the user’s preferences.A con-tent-based movie recommendation model is proposed in this research,which recommends movies based on the user’s profile from the Facebook platform.The recommendation system is built with a hybrid model that combines the Mon-arch Butterfly Optimization(MBO)with the Deep Belief Network(DBN).For feature selection,the MBO is utilized,while DBN is used for classification.The datasets used in the experiment are collected from Facebook and MovieLens.The dataset features are evaluated for performance evaluation to validate if data with various attributes can solve the matching recommendations.Eachfile is com-pared with features that prove the features will support movie recommendations.The proposed model’s mean absolute error(MAE)and root-mean-square error(RMSE)values are 0.716 and 0.915,and its precision and recall are 97.35 and 96.60 percent,respectively.Extensive tests have demonstrated the advantages of the proposed method in terms of MAE,RMSE,Precision,and Recall compared to state-of-the-art algorithms such as Fuzzy C-means with Bat algorithm(FCM-BAT),Collaborativefiltering with k-NN and the normalized discounted cumulative gain method(CF-kNN+NDCG),User profile correlation-based similarity(UPCSim),and Deep Autoencoder.
基金Supported bythe Hunan Teaching Reformand Re-search Project of Colleges and Universities (2003-B72) the HunanBoard of Review on Philosophic and Social Scientific Pay-off Project(0406035) the Hunan Soft Science Research Project(04ZH6005)
文摘Content-based filtering E-commerce recommender system was discussed fully in this paper. Users＇ unique features can be explored by means of vector space model firstly. Then based on the qualitative value of products informa tion, the recommender lists were obtained. Since the system can adapt to the users＇ feedback automatically, its performance were enhanced comprehensively. Finally the evaluation of the system and the experimental results were presented.
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This work is supported by the National Natural Science Foundation of China under Grant Nos.U1636215,61902082the Guangdong Key R&D Program of China 2019B010136003Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme(2019).
文摘Recommender systems are very useful for people to explore what they really need.Academic papers are important achievements for researchers and they often have a great deal of choice to submit their papers.In order to improve the efficiency of selecting the most suitable journals for publishing their works,journal recommender systems(JRS)can automatically provide a small number of candidate journals based on key information such as the title and the abstract.However,users or journal owners may attack the system for their own purposes.In this paper,we discuss about the adversarial attacks against content-based filtering JRS.We propose both targeted attack method that makes some target journals appear more often in the system and non-targeted attack method that makes the system provide incorrect recommendations.We also conduct extensive experiments to validate the proposed methods.We hope this paper could help improve JRS by realizing the existence of such adversarial attacks.
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.
文摘This study presents a new approach that advances the algorithm of similarity measures between generalized fuzzy numbers. Following a brief introduction to some properties of the proposed method, a comparative analysis based on 36 sets of generalized fuzzy numbers was performed, in which the degree of similarity of the fuzzy numbers was calculated with the proposed method and seven methods established by previous studies in the literature. The results of the analytical comparison show that the proposed similarity outperforms the existing methods by overcoming their drawbacks and yielding accurate outcomes in all calculations of similarity measures under consideration. Finally, in a numerical example that involves recommending cars to customers based on a nine-member linguistic term set, the proposed similarity measure proves to be competent in addressing fuzzy number recommendation problems.
基金funded by the State Grid Hebei Electric Power Company(Project Number:KJ2023-093).
文摘Session-based recommendation systems(SBR)are pivotal in suggesting items by analyzing anonymized sequences of user interactions.Traditional methods,while competent,often fall short in two critical areas:they fail to address potential inter-session item transitions,which are behavioral dependencies that extend beyond individual session boundaries,and they rely on monolithic item aggregation to construct session representations.This approach does not capture the multi-scale and heterogeneous nature of user intent,leading to a decrease in modeling accuracy.To overcome these limitations,a novel approach called HMGS has been introduced.This system incorporates dual graph architectures to enhance the recommendation process.A global transition graph captures latent cross-session item dependencies,while a heterogeneous intra-session graph encodesmulti-scale item embeddings through localized feature propagation.Additionally,amulti-tier graphmatchingmechanism aligns user preference signals across different granularities,significantly improving interest localization accuracy.Empirical validation on benchmark datasets(Tmall and Diginetica)confirms HMGS’s efficacy against state-of-the-art baselines.Quantitative analysis reveals performance gains of 20.54%and 12.63%in Precision@10 on Tmall and Diginetica,respectively.Consistent improvements are observed across auxiliary metrics,with MRR@10,Precision@20,and MRR@20 exhibiting enhancements between 4.00%and 21.36%,underscoring the framework’s robustness in multi-faceted recommendation scenarios.
文摘In L2 content-based classrooms,code-switching or translanguaging seem to be a common practice adopted by teachers.There has been growing research discussing the potentials of L1 in these classrooms.Most of the current studies have focused on the analysis of lesson interactions and yet the perception of the content teachers has remained underexplored.This case study investigated the introspective views of a group of content teachers at a secondary school using questionnaires and written accounts.Data analyses showed that these teachers were generally aware of the interpersonal and ideational functions achieved by the use of L1 and they also seemed to have a positive view towards their practices of using L1 in English-medium classrooms.Based on the findings,practical implications for content teachers in relation to making medium of instruction decisions and suggestions for further research are discussed.
基金supported by a grant from the Natural Science Foundation of Zhejiang Province under Grant LY21F010016.
文摘Recently,many Sequential Recommendation methods adopt self-attention mechanisms to model user preferences.However,these methods tend to focus more on low-frequency information while neglecting highfrequency information,which makes them ineffective in balancing users’long-and short-term preferences.At the same time,manymethods overlook the potential of frequency domainmethods,ignoring their efficiency in processing frequency information.To overcome this limitation,we shift the focus to the combination of time and frequency domains and propose a novel Hybrid Time-Frequency Dual-Branch Transformer for Sequential Recommendation,namely HyTiFRec.Specifically,we design two hybrid filter modules:the learnable hybrid filter(LHF)and the window hybrid filter(WHF).We combine these with the Efficient Attention(EA)module to form the dual-branch structure to replace the self-attention components in Transformers.The EAmodule is used to extract sequential and global information.The LHF andWHF modules balance the proportion of different frequency bands,with LHF globally modulating the spectrum in the frequency domain and WHF retaining frequency components within specific local frequency bands.Furthermore,we use a time domain residual information addition operation in the hybrid filter module,which reduces information loss and further facilitates the hybrid of time-frequency methods.Extensive experiments on five widely-used real-world datasets show that our proposed method surpasses state-of-the-art methods.
基金Supported by Key R&D Projects of Hunan Provincial Department of Science and Technology"Study on Key Modern Processing Techniques and Product Development of Huarong Mustard"(2023NK2039).
文摘A survey conducted on the premature bolting of Huarong large leaf mustard from 2018 to 2024 revealed that Huarong large leaf mustard sown in middle August was associated with a higher propensity for premature bolting. Furthermore, it was observed that the earlier being sown, the greater the rate of premature bolting when being sown prior to middle August. The rate of premature bolting observed in seedlings sown on August 8 was recorded at 35.6%. It was noted that as the age of the seedlings increased, the rate of premature bolting correspondingly increased. There were notable differences in the tolerance of various cultivars to elevated temperatures and prolonged sunlight exposure. For instance, cultivars such as Zhangjie 1 and Sichuan Shaguodi, which exhibit greater heat resistance, did not demonstrate premature bolting when sown in early August. The prolonged exposure to elevated temperatures, drought conditions, and extended periods of sunlight during the seedling stage of Huarong large leaf mustard, coupled with delayed irrigation and transplantation, contributed to the occurrence of premature bolting. The Huarong large leaf mustard, when been sown from late August to early September and transplanted at the appropriate time, exhibited normal growth and development, with no instances of premature bolting observed. It is advisable to select heat-resistant varieties, such as Zhangjie 1, prior to middle August. Huarong large leaf mustard should be sown in early to middle September. Additionally, it is essential to ensure centralized production and timely release of seeds, prompt transplantation and harvesting, and enhance the management of pests and diseases.
基金supported by the Key Research and Development Program of Heilongjiang Province(No.2022ZX01A35).
文摘As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as a new challenge in the field of recommendation systems.This paper introduces a group-buying recommendation model based on multi-head attention mechanisms and multi-task learning,termed the Multi-head Attention Mechanisms and Multi-task Learning Group-Buying Recommendation(MAMGBR)model,specifically designed to optimize group-buying recommendations on e-commerce platforms.The core dataset of this study comes from the Chinese maternal and infant e-commerce platform“Beibei,”encompassing approximately 430,000 successful groupbuying actions and over 120,000 users.Themodel focuses on twomain tasks:recommending items for group organizers(Task Ⅰ)and recommending participants for a given group-buying event(Task Ⅱ).In model evaluation,MAMGBR achieves an MRR@10 of 0.7696 for Task I,marking a 20.23%improvement over baseline models.Furthermore,in Task II,where complex interaction patterns prevail,MAMGBR utilizes auxiliary loss functions to effectively model the multifaceted roles of users,items,and participants,leading to a 24.08%increase in MRR@100 under a 1:99 sample ratio.Experimental results show that compared to benchmark models,such as NGCF and EATNN,MAMGBR’s integration ofmulti-head attentionmechanisms,expert networks,and gating mechanisms enables more accurate modeling of user preferences and social associations within group-buying scenarios,significantly enhancing recommendation accuracy and platform group-buying success rates.
文摘This paper examines the impact of algorithmic recommendations and data-driven marketing on consumer engagement and business performance.By leveraging large volumes of user data,businesses can deliver personalized content that enhances user experiences and increases conversion rates.However,the growing reliance on these technologies introduces significant risks,including privacy violations,algorithmic bias,and ethical concerns.This paper explores these challenges and provides recommendations for businesses to mitigate associated risks while optimizing marketing strategies.It highlights the importance of transparency,fairness,and user control in ensuring responsible and effective data-driven marketing.
文摘This special issue of the Asian Journal of Andrology is fully dedicated to the thematic area of non-obstructive azoospermia(NOA),one of the most complex and challenging conditions in the realm of andrology,urology,and reproductive medicine.
文摘In the era of exponential growth of digital information,recommender algorithms are vital for helping users navigate vast data to find relevant items.Traditional approaches such as collaborative filtering and contentbasedmethods have limitations in capturing complex,multi-faceted relationships in large-scale,sparse datasets.Recent advances in Graph Neural Networks(GNNs)have significantly improved recommendation performance by modeling high-order connection patterns within user-item interaction networks.However,existing GNN-based models like LightGCN and NGCF focus primarily on single-type interactions and often overlook diverse semantic relationships,leading to reduced recommendation diversity and limited generalization.To address these challenges,this paper proposes a dual multi-relational graph neural network recommendation algorithm based on relational interactions.Our approach constructs two complementary graph structures:a User-Item Interaction Graph(UIIG),which explicitly models direct user behaviors such as clicks and purchases,and a Relational Association Graph(RAG),which uncovers latent associations based on user similarities and item attributes.The proposed Dual Multi-relational Graph Neural Network(DMGNN)features two parallel branches that perform multi-layer graph convolutional operations,followed by an adaptive fusion mechanism to effectively integrate information from both graphs.This design enhances the model’s capacity to capture diverse relationship types and complex relational patterns.Extensive experiments conducted on benchmark datasets—including MovieLens-1M,Amazon-Electronics,and Yelp—demonstrate thatDMGNN outperforms state-of-the-art baselines,achieving improvements of up to 12.3%in Precision,9.7%in Recall,and 11.5%in F1 score.Moreover,DMGNN significantly boosts recommendation diversity by 15.2%,balancing accuracy with exploration.These results highlight the effectiveness of leveraging hierarchical multi-relational information,offering a promising solution to the challenges of data sparsity and relation heterogeneity in recommendation systems.Our work advances the theoretical understanding of multi-relational graph modeling and presents practical insights for developing more personalized,diverse,and robust recommender systems.
