A good recommender system can infer customers’ preferences based on their historical purchase records, and recommend products that the customers may be interested in, saving them a lot of time and energy. For enterprises, it is difficult to recommend accurately to each customer, and the bad recommendation may be counterproductive. Customer loyalty is an indicator that measures the preference relationship between customers and products in the field of marketing. A hypothesis is proposed in this study: if companies can divide customers into different groups based on customer loyalty, the recommendation effect on certain groups is better than that on overall customers. In this study, customer loyalty is measured by four features of the RFML model. All customers are viewed as points on a four-dimensional space, which are clustered by the k-means model. Two recommendation algorithms based on complex networks are tested: recommendation algorithm based on bipartite graph and PersonalRank (BGPR), and recommendation algorithm based on a single vertex set network and DeepWalk (SVDW). The experimental results show that customer loyalty has improved the effectiveness of the two algorithms over 14%, and the recommendation effect is the best on customer groups with a loyalty level of 4 (the highest level is 5). The recommendation algorithms with customer loyalty are better than using them alone.

Recommender systems have been applied in a wide range of domains such as e-commerce, media, banking, and utilities. This kind of system provides personalized suggestions based on large amounts of data in order to increase user satisfaction. These suggestions help client select products, while organizations can increase the consumption of a product. In the case of social data, sentiment analysis can help gain better understanding of a user’s attitudes, opinions and emotions, which is beneficial to integrate in recommender systems for achieving higher recommendation reliability. On the one hand, this information can be used to complement explicit ratings given to products by users. On the other hand, sentiment analysis of items that can be derived from online news services, blogs, social media or even from the recommender systems themselves is seen as capable of providing better recommendations to users. In this study, we present and evaluate a recommendation approach that integrates sentiment analysis into collaborative filtering methods. The recommender system proposal is based on an adaptive architecture, which includes improved techniques for feature extraction and deep learning models based on sentiment analysis. The results of the empirical study performed with two popular datasets show that sentiment–based deep learning models and collaborative filtering methods can significantly improve the recommender system’s performance.

Online learning is gaining massive popularity with time. The e-learning platforms operate differently from traditional educational institutions and hence need different strategy for course recommendations. This survey aims to cover the major emerging research areas in e-learning recommender systems. Our study covers different areas of research including graph-based methodologies, ITS, query optimization, content-based, and collaborative filtering, big data, and association rules mining. This survey aimed to explore all major emerging directions of recommender systems in education. This study analyzed existing literature in all of the areas mentioned before and performed objective-based analysis. A brief performance analysis was also done for the researches where the values were comparable. Limitations of existing researches were also identified and studied to shed some light on future directions.

This paper focuses on the new users cold-start issue in the context of recommender systems. New users who do not receive pertinent recommendations may abandon the system. In order to cope with this issue, we use active learning techniques. These methods engage the new users to interact with the system by presenting them with a questionnaire that aim to understand their preferences to the related items. Example of questions may include "do you like this book?" and the users answer,"yes", "no", "I have not read it (unknown)", will reflect the degree of interest for the item by the users. As a consequence, the system can learn the users' preferences from these answers. The goal of active learning is to correctly choose the questions (items) for users. Thus it is necessary to personalize the questionnaires to retrieve the maximum information by avoiding "unknown" answers. In this paper, we propose an active learning technique that exploits past users' interests and past users' predictions in order to identify the best questions to ask.

This paper proposes the utilization of large language models as recommendations systems for museums. Since the aforementioned models lack the notion of context, they can’t work with temporal information that is often present in recommendations for cultural environments (e.g. special exhibitions or events). In this respect, the current work aims at enhancing the capabilities of large language models through a fine-tuning process that incorporates contextual information and user instructions. The resulting models are expected to be capable of providing personalized recommendations, aligned with user preferences and desires. More specifically, Generative Pre-trained Transformer 4, a knowledge-based large language model is fine-tuned and turned into a context-ware recommendation system, adapting its suggestions based on user input and specific contextual factors such as location, time of visit, and other relevant parameters. The effectiveness of the proposed approach is evaluated through certain user studies, which ensure an improved user experience and engagement within the museum environment.

The practice of quantified-self sleep tracking is increasingly common nowadays among healthy individuals as well as patients with sleep problems. However, existing sleep-tracking technologies only support simple data collection and visualization, and are incapable of providing actionable recommendations that are tailored to users' physical, behavioral and environmental context. Here we coined the term context-aware sleep health recommender system (CASHRS) as an emerging multidisciplinary research field that bridges ubiquitous sleep computing and context-aware recommender systems. In this paper, we presented a narrative review to analyze the type of contextual information, the recommendation algorithms, the context filtering techniques, the behavior change techniques, the system evaluation, and the challenges in peer-reviewed publications that meet the characteristics of CASHRS. Analysis results identified current research trends, the knowledge gap, and future research opportunities in CASHRS.

The advent of the era of big data will bring more convenience to people and greater development to society. But at the same time, it will also bring people the problem of 'information overload', i.e., when people are faced with huge information data, there are many redundant and worthless data. The redundant and worthless data information seriously interferes with the accurate selection of information data. Even though people can use Internet search engines to access information data, they cannot meet the individual needs of specific users in specific contexts. The personalized needs of a particular user in a particular context. Therefore, how to find useful and valuable information quickly has become one of the key issues in the development of big data. With the advent of the era of big data, recommendation systems, as an important technology to alleviate information overload, have been widely used in the field of e-commerce. Recommender systems suffer from a key problem: data sparsity. The sparsity of user history rating data causes insufficient training of collaborative filtering recommendation models, which leads to a significant decrease in the accuracy of recommendations. In fact, traditional recommendation systems tend to focus on scoring information and ignore the contextual context in which users interact. There are various contextual modal information in people's real life, which also plays an important role in the recommendation process. In this paper we achieve data degradation and feature extraction, solving the problem of sparse data in the recommendation process. An interaction context-aware sub-model is constructed based on a tensor decomposition model with interaction context information to model the specific influence of interaction context in the recommendation process. Then an attribute context-aware sub-model is constructed based on the matrix decomposition model and using attribute context information to model the influence of user attribute contexts and item attribute contexts on recommendations. In the process of building the model, the method not only utilizes the explicit feedback rating information of users in the original dataset, but also utilizes the interaction context and attribute context information of the implicit feedback and the unlabeled rating data. We evaluate our model by extensive experiments. The results illustrate the effectiveness of our recommender model.

Recommender Systems (RecSys) are crucial in managing information overload and enhancing user satisfaction across various digital platforms, including e-commerce and entertainment. Evolving from traditional models to Deep Neural Networks (DNNs) and, more recently, Large Language Models (LLMs), these systems leverage sophisticated algorithms to analyze user behaviors and preferences. LLMs, such as GPT-4, are trained on extensive datasets to comprehend and generate natural language, significantly advancing their ability to deliver personalized recommendations. This tutorial explores the transformative impact of LLMs on RecSys, discussing their development, application in handling complex datasets, and the integration of contextual insights. Real-world examples illustrate how LLMs enhance recommendation accuracy and user experience, highlighting challenges and future directions in the field.

In this paper, we provide a study on the effects of applying classical clustering algorithms, such as k-Means to free text recommender systems. A typical recommender system may face problems when the number of items from a database goes from a few items to hundreds of items. Currently, one of the most prominent techniques to scale the database is applying clustering, however clustering may have a negative impact on the accuracy of the system when applied without taking into consideration the underlying items. In this work, we build a conceptual text recommender system and use k-Means to partition its search space into different groups. We study how the variation of the number of clusters affects its performance in the light of two performance measurements: recommendation time and precision. We also analyze if this clustering is affected by the representation of text we use. All the techniques used in this study uses word-embeddings to represent the document. One of the main findings of this work is that using clustering we can improve the recommendation time in up to almost 30 times without affecting much off its initial accuracy. Another interesting finding is that the increment of the number of clusters is not directly translated into linear performance.

For personalized recommender systems,matrix factorization and its variants have become mainstream in collaborative filtering.However,the dot product in matrix factorization does not satisfy the triangle inequality and therefore fails to capture fine-grained information. Metric learning-based models have been shown to be better at capturing fine-grained information than matrix factorization. Nevertheless,most of these models only focus on rating data and social information, which are not sufficient for dealing with the challenges of data sparsity. In this paper,we propose a metric learning-based social recommendation model called SRMC.SRMC exploits users' co-occurrence pattern to discover their potentially similar or dissimilar users with symmetric relationships and change their relative positions to achieve better recommendations.Experiments on three public datasets show that our model is more effective than the compared models.

In today’s Web 2.0 era, online social media has become an integral part of our lives. In the course of the information revolution, the form of information has undergone a radical change, from simple text information to today’s integrated video, image, text and audio, and there has also been a great change in the way of dissemination and access, as people nowadays do not just rely on traditional media to passively receive information, but more actively and selectively obtain information from social media. Therefore, it has become a great challenge for us to effectively utilize these massive and integrated multi-modal media information to form an effective system of retrieval, browsing, analysis and usage. Unlike movies and traditional long-form video content, micro-videos are usually short in length, between a few seconds and tens of seconds, which allows users to quickly browse different contents and make full use of the fragmented time in their lives, while users can also share their micro-videos to their friends or the public, forming a unique social way. Video contains rich multimodal information, and fusing information from multiple modalities in a video recommendation task can improve the accuracy of the video recommendation task.According to the micro-video recommendation task, a new combinatorial network model is proposed to combine the discrete features of each modality into the overall features of various modalities through the network, and then fuse the various modal features to obtain the overall video features, which will be used for recommendation. In order to verify the effectiveness of the algorithm proposed in this paper, experiments are conducted in the public dataset, and it is shown the effectiveness of our model.

The TeNDER project aims to improve the Quality of Life (QoL) of chronic patients through an integrated care ecosystem. This study evaluates the Health Recommender System (HRS) developed for the project, which offers personalized recommendations based on data collected from a set of monitoring devices. The list of notifications covered different areas of daily-life such as physical activity, nutrition, and sleep. We conducted this case of study to evaluate the effectiveness and usability of the HRS in providing accurate and relevant recommendations to users. Evaluation process consisted on survey administration for QoL assessment, and the satisfaction and usability of the HRS. The four-week pilot study involved several patients and caregivers, and demonstrated that the HRS was perceived as user-friendly, consistent, and helpful, with a positive impact on patients’ QoL. However, the study highlights the need for improvement in terms of personalization of recommendations.

Bundle recommendation is provided to users by combining related items into a bundle, enhancing the user's shopping experience and increase the merchant's sales revenue. Despite the success of existing solutions based on Graph Neural Networks(GNN), there are still some significant challenges: (1)It is demanding to explicitly model multiple complex associations using standard graph neural networks. (2)Numerous additional nodes and edges are introduced to approximate higher-order associations. (3)The user-bundle historical interaction data is highly sparse. In this work, we propose a Global Structural Hypergraph Convolution model for Bundle Recommendation (SHCBR) to address the above problems. Specifically, we jointly incorporate multiple complex interactions between users, items, and bundles(three entities) into a relational hypergraph without introducing additional nodes and edges. Higher-order associations are already involved in the hypergraph structure which alleviate the training burden of neural networks and the dilemma of scarce data. In addition, we design a special matrix propagation rule that captures non-pairwise complex relationships between entities. Using item nodes as links, hypergraph convolution propagation between user and bundle nodes generates the learned representations. Experimental results on two real-world datasets demonstrate that SHCBR outperforms the state-of-the-art baselines by 11.07%-25.66\% on Recall and 16.81%-33.53% on NDCG. We publicly release the codes and datasets at https://github.com/Lxtdzh/SHCBR-master/.

In recent years, the scientific and technological developments led to an explosion of available videos on the web, increasing the necessity of fast and effective video analysis and summarization. Video summarization methods aim to generate a synopsis by selecting the most informative parts of the video content. The user’s personal preferences, often involved in the expected results, should be taken into account in the video summaries. In this paper, we provide the first comprehensive survey on personalized video summarization relevant to the techniques and datasets used. In this context, we classify and review personalized video summary techniques based on the type of personalized summary, on criteria, on the video domain, on the source of information, on the time of summarization, and on the machine learning technique. Depending on the type of methodology used by the personalized video summarization techniques for the summary production process, we classify the techniques into five major categories, which are feature-based video summarization, key frame selection, shot selection-based approach, video summarization using trajectory analysis, and personalized video summarization using clustering. We also compare personalized video summarization methods and present 37 datasets used to evaluate personalized video summarization methods. Finally, we analyze opportunities and challenges in the field and suggest innovative research lines.

Continuous discovery and update of applications or their boundaries running in cloud environments in an automatic way is a highly required function of modern data center ops solutions. Prior attempts to address this problem within various products or projects were/are applying rule-driven approaches or machine learning techniques on specific types of data − network traffic as well as property/configuration data of infrastructure objects, which all have their drawbacks in effectively identifying roles of those resources. The current proposal (ADLog) leverages log data of sources, which contain incomparably richer contextual information, and native constructs of VMware Aria Operations for Logs in terms of Event Types and their distributions to group those entities, potentially automatically enrich with indicative tags, and recommend for further management tasks and policies. Our methods discriminate not only diverse kinds of applications, but also their specific deployments thus providing hierarchical representation of the applications in time and topology. For several applications under Aria Ops management in our experimental test bed, we discover those in terms of similarity behavior of their components with a high accuracy. The validation of the proposal paves the path for an AI-driven solution in the cloud management scenarios.

One of the main challenges for autonomous learning in virtual environments is finding the right material that fits students’ needs and supports their learning process. Personalized recommender systems partially solve this problem by suggesting online educational resources to students based on their preferences. However, in educational environments (which need a proper characterization of both users and educational resources), most existing recommendation algorithms either fail to include all the available information or use hybrid processes that do not exploit possible relationships between users and item features. This article presents a personalized recommender system for educational resources aimed at combining user and item information into a single mathematical model based on matrix factorization. As a result, estimated latent factors can provide insight into possible interactions between users and item features, improving the quality of the information retrieval process. We validated the proposed model on a real dataset that contains the ratings assigned by students from Universidad Nacional de Colombia and Universidade Feevale to educational resources in the Colombian Federation of Learning Object Repositories (FROAC in Spanish). User characterization included learning style and educational level, whereas item characterization (obtained from the objects’ metadata), included interactivity level, aggregation level and type, and resource format. These results, compared to those obtained when not all the available information is included, show that our method can improve the recommendation process.

Session-based recommendations which aim to predict subsequent user-item interaction based on historical user behavior during anonymous sessions can be challenging tasks. Two main challenges need to be addressed and improved: (1) How to analyze these sessions to accurately and completely capture users’ preferences, and (2) how to identify and eliminate any interference caused by noisy behavior. Existing methods have not adequately addressed these issues since they either neglected the valuable insight that can be gained from analyzing consecutive groups of items or failed to take these noisy data in sessions seriously and handled them properly, which can jointly impede recommendation systems from capturing users’ real intentions. To address these two problems, we designed a multi-order semantic denoising (MSD) model for session-based recommendations. Specifically, we grouped items of different lengths as varying multi-order semantic units to mine the user’s primary intention from multiple dimensions. Meanwhile, a novel denoising network was designed to alleviate the interference of noisy behavior and provide a more precise session representation. The result of extensive experiments on three real-world datasets demonstrated that the proposed MSD model exhibited improved performance compared with existing state-of-the-art methods in session-based recommendation.

Recently, e-commerce companies have been adopting and continuously enhancing personalized product recommendation systems, and there is active research in this field within academia. However, personalized product recommendations in offline retail stores have not yielded significant outcomes thus far. Especially for small-format offline retail businesses like convenience stores, where customer information might be limited, providing personalized product recommendations poses even greater challenges. To address this issue, this study aimed to find a solution by shifting the perspective on recommendation methods and altering the target of recommendations in existing personalized recommendation models. In this study, recommending products that customers need in an offline store was defined as suggesting products that should be introduced and displayed within the store. In other words, the recommendation focus shifted from individuals to the stores themselves. This recommendation system proposes products that individual stores have not yet introduced but are anticipated to be purchased by customers among the products managed by the headquarters. Building upon this, we devised a store-based product recommendation system. The widely used user-based collaborative filtering model, a common technique in existing recommendation systems, was adapted into a store-based collaborative filtering model. Furthermore, various rules and logic pertinent to store operations and business considerations for convenience stores were integrated to implement this store product recommendation system. The accuracy and effectiveness of the system were demonstrated through its application in actual convenience stores. Results from the pilot implementation of the system showed that 88% of the newly recommended products in individual stores were sold based on one week of sales data, and the sales revenue was 1.75 times higher than the average sales revenue of those products across the entire company. Survey results on business owners' satisfaction yielded a score of 4.2 out of 5, indicating a high level of contentment. Additional observed effects included the expansion of product range and reduction in order lead times. However, 12% of the total recommended SKUs did not sell within one week, and 34% of the SKUs did not meet the overall average sales quantity. This research holds significance in extending the scope of personalized recommendation studies from primarily online platforms to offline retail businesses like convenience stores. It provides tangible methodologies and outcomes that can be implemented in real-world settings through the construction and validation of an actual system. The study also suggests avenues for future research to address some of the identified limitations.

Recently, the application of deep reinforcement learning in recommender system is flourishing and stands out by overcoming drawbacks of traditional methods and achieving high recommendation quality. The dynamics, long-term returns and sparse data issues in recommender system have been effectively solved. But the application of deep reinforcement learning brings problems of interpretability, overfitting, complex reward function design, and user cold start. This paper proposed a tag-aware recommender system based on deep reinforcement learning without complex function design, taking advantage of tags to make up for the interpretability problems existing in recommender system. Our experiment is carried out on MovieLens dataset. The result shows that, DRL based recommender system is superior than traditional algorithms in minimum error and the application of tags has little effect on accuracy when making up for interpretability. In addition, DRL based recommender system has excellent performance on user cold start problems.

Information overload is the biggest challenges nowadays for any website, especially the e-commerce website. However, this challenge arises for the fast growth of information on the web (WWW) with easy access to the internet. Collaborative filtering based recommender system is the most useful application to solve the information overload problem by filtering relevant information for the users according to their interests. But, the existing system faces some significant limitations like as data sparsity, low accuracy, cold-start and malicious attacks. To alleviate the mentioned issues, the relationship of trust incorporates in the system where it can be between the users or items, and such system is known as the trust-based recommender system (TBRS). From the user perspective, the motive of the TBRS is to utilize the reliability between the users to generate more accurate and trusted recommendations. However, the aim of the paper is to present a comparative analysis of different trust metrics in the context of the type of trust definition of TBRS. Also, the study accomplishes on twenty-four trust metrics in terms of the methodology, trust properties & measurement, validation approaches, and the experimented dataset.

This study introduces session-aware recommendation models, leveraging GRU (Gated Recurrent Unit) and attention mechanisms for advanced latent interaction data integration. A primary ad-vancement is enhancing latent context, a critical factor for boosting recommendation accuracy. We address existing models' rigidity by dynamically blending short-term (most recent) and long-term (historical) preferences, moving beyond static period definitions. Our approaches, pre-combination (LCII-Pre) and post-combination (LCII-Post) with fixed (Fix) and flexible learning (LP) weight configurations, are thoroughly evaluated. We conducted extensive experiments to as-sess our models' performance on public datasets such as Amazon and MovieLens 1M. Notably, on the MovieLens 1M dataset, LCII-PreFix achieved a 1.85% and 2.54% higher Recall@20 than II-RNN and BERT4Rec+st+TSA, respectively. On the Steam dataset, LCII-PostLP outperformed these models by 18.66% and 5.5%. Furthermore, on the Amazon dataset, LCII showed a 2.59% and 1.89% improvement in Recall@20 over II-RNN and CAII. These results affirm the significant enhancement our models bring to session-aware recommendation systems, showcasing their po-tential for both academic and practical applications in the field.

Multiple mRNA isoforms of the same gene are produced via alternative splicing, a biological mechanism that regulates protein diversity while maintaining genome size. Alternatively spliced mRNA isoforms of the same gene may sometimes have very similar sequence, but they can have significantly diverse effects on cellular function and regulation. The products of alternative splicing have important and diverse functional roles, such as response to environmental stress, regulation of gene expression, human heritable and plant diseases. The mRNA isoforms of the same gene, such as the apoptosis associated CASP3 gene, can have dramatically different functions. The shorter mRNA isoform product CASP3-S inhibits apoptosis, while the longer CASP3-L mRNA isoform promotes apoptosis. Despite the functional importance of mRNA isoforms, very little has been done to annotate their functions. The recent years have however seen the development of several computational methods aimed at predicting mRNA isoform level biological functions. These methods use a wide array of proteo-genomic data to develop machine learning-based mRNA isoform function prediction tools. In this review, we discuss the computational methods developed for predicting the biological function at the individual mRNA isoform level.

Here, we propose a prescriptive simulation model of a process operator’s decision-making assisted by artificial intelligence (AI) algorithm in a technical system control loop. We analyze situations fraught with a catastrophic threat that may cause unacceptable damage. Operators’ decision-making is interpreted in terms of a subjectively admissible probability of disaster and subjectively necessary reliability of its assessment. We distinguish four extreme decision-making strategies corresponding to different ratios between the above variables. An experiment simulating a process facility, an AI algorithm and operator's decision-making strategy was held. It showed that depending on the properties of a controlled process (the speed of hazard onset) and the AI algorithm characteristics (Type I and II error rate), each of such strategies or some intermediate strategy may prove to be more beneficial than others. The same approach is applicable to the identification and analysis of sustainability of strategies applied in real-life operating conditions, as well as to the development of a computer simulator to train operators to control hazardous technological processes using AI-generated advice.