No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Advances in Collaborative Filtering
Abstract
The collaborative filtering (CF) approach to recommenders has recently enjoyed much interest and progress. The fact that it played a central role within the recently completed Netflix competition has contributed to its popularity. This chapter surveys the recent progress in the field. Matrix factorization techniques, which became a first choice for implementing CF, are described together with recent innovations. We also describe several extensions that bring competitive accuracy into neighborhood methods, which used to dominate the field. The chapter demonstrates how to utilize temporal models and implicit feedback to extend models accuracy. In passing, we include detailed descriptions of some the central methods developed for tackling the challenge of the Netflix Prize competition.
... Recommender systems (RS) could deal with the problem of information overload [1] in the big data era, which has been widely studied [2,3,4]. By analyzing the previous user-item interactions (e.g. ...
... The basic assumption of matrix factorization (MF) [1] is that a set of k-dimensional features can represent user preferences and item attributes. MF extracts users' and items' latent factor vectors from the rating matrix. ...
In recommender systems, the rating matrix is usually not a global low-rank but local low-rank. Constructing low-rank submatrices for matrix factorization can improve the accuracy of rating prediction. This paper proposes a novel network embedding-based local matrix factorization model, which can built more meaningful sub-matrices. To alleviate the sparsity of the rating matrix, the social data and the rating data are integrated into a heterogeneous information network, which contains multiple types of objects and relations. The network embedding algorithm extracts the node representations of users and items from the heterogeneous information network. According to the correlation of the node representations, the rating matrix is divided into different sub-matrices. Finally, the matrix factorization is performed on the sub-matrices for rating prediction. We test our network embedding-based method on two real-world public data sets (Yelp and Douban). Experimental results show that our method can obtain more accurate prediction ratings.
... The collaborative filtering stream of recommendation studies commonly specifies consumer interest in a product as a whole, models interest as a probabilistic distribution over the product-level set, and learns the latent interest directly from the visible behaviors (Linden et al. 2003, He et al. 2018, Koren and Bell 2021. In this regard, consumer interest can be referred to as the attitude/ tendency to respond to a product with some degree of favorableness (Ajzen 2008) and is an important determinant of consumers' interactive behaviors with products (e.g., click, reclick, add-to-favorites, add-to-cart, and purchase), as indicated by well-established behavioral theories, such as theory of reasoned action (Shepherd et al. 1988) and theory of planned behavior (Ajzen 1991). ...
... As an important determinant of behavior, interest essentially motivates consumers' observed interactions with products in online shopping (Howard andSheth 1969, Jansen andSchuster 2011), often in a latent and hybrid manner because of subjectivity, uncertainty, and cognitive limitations (Edwards andFasolo 2001, Ajzen 2008). Thus, a common treatment in recommendation studies is to model consumer interest as a probabilistic distribution over the product set (He et al. 2018, Koren andBell 2021), whose shifting patterns can be further captured by the dynamic methods (Rabiu et al. 2020). In practical shopping journeys, the interest shifts of consumers are very complex and varying (Lam and Mostafa 2001), that is, showing variability in interest shifts, and the distinct psychological stages can act as underlying drivers that bring about different patterns of interest shifts (Court et al. 2009, Mulpuru 2011, Goldstein and Hajaj 2022. ...
Recommender systems are widely used by platforms/merchants to find the products that are likely to interest consumers. However, existing dynamic methods still face challenges with regard to diverse behaviors, variability in interest shifts, and the identification of psychological dynamics. Premised on the marketing funnel perspective to analyze consumer shopping journeys, this study proposes a novel and effective machine learning approach for product recommendation, namely, multi-stage dynamic Bayesian network (MS-DBN), which models the generative processes of consumers’ interactive behaviors with products in light of their stage transitions and interest shifts. In this way, consumers’ stage-interest-behavior dynamics can be learnt, especially the variability in interest shifts. This provides managerial implications for practice. MS-DBN demonstrates significant performance advantage with general applicability by extracting the generalizable regularity during shopping journeys, which compensates the diversity and sparsity frequently observed in consumer behaviors. In addition, aided by the identification strategies integrated into the learning process, the latent variables in the model can be detected such that consumers’ invisible psychological stages and interests in products can be identified from their observed behaviors, shedding light on the targeted marketing of platforms/merchants and thus enriching the practical value of the approach.
... The recommendation systems literature offers a wide range of methods that represent users, and items in R d . These methods then propose a compatibility score function between the user and item, ϕ : R d × R d → R. A common and effective choice for ϕ is the dot product, which underpins matrix factorization (Rendle et al., 2020;Koren & Bell, 2015). To capture more complex interactions among users, items, and attributes, (He et al., 2017) extend matrix factorization by replacing the dot product with a neural networkbased similarity function. ...
Personalized item recommendation typically suffers from data sparsity, which is most often addressed by learning vector representations of users and items via low-rank matrix factorization. While this effectively densifies the matrix by assuming users and movies can be represented by linearly dependent latent features, it does not capture more complicated interactions. For example, vector representations struggle with set-theoretic relationships, such as negation and intersection, e.g. recommending a movie that is "comedy and action, but not romance". In this work, we formulate the problem of personalized item recommendation as matrix completion where rows are set-theoretically dependent. To capture this set-theoretic dependence we represent each user and attribute by a hyper-rectangle or box (i.e. a Cartesian product of intervals). Box embeddings can intuitively be understood as trainable Venn diagrams, and thus not only inherently represent similarity (via the Jaccard index), but also naturally and faithfully support arbitrary set-theoretic relationships. Queries involving set-theoretic constraints can be efficiently computed directly on the embedding space by performing geometric operations on the representations. We empirically demonstrate the superiority of box embeddings over vector-based neural methods on both simple and complex item recommendation queries by up to 30 \% overall.
... Content-based filtering, while leveraging course attributes, often fails when course descriptions are inconsistent or incomplete. These challenges highlight the need for approaches that balance efficiency, scalability, and effectiveness [18]. While DORIS, DECOR, and DRL-based methods demonstrate significant advancements, they often depend on computationally intensive deep learning models that are impractical for small-scale or resourcelimited settings. ...
Course recommendation aims to find suitable and attractive courses for students based on their needs, playing a significant role in the curricula-variable system. However, with the abundant available courses, students often face cognitive overload when selecting the most appropriate ones. This research proposes a course recommendation system called the Enhanced Hybrid Course Recommender to address this challenge. This system uses an ensemble learning approach to combine and leverage the power of multiple machine learning classifiers, including Random Forest, Naive Bayes, and Support Vector Machine. By utilizing TF-IDF vectorization for text data transformation and label encoding for target label compatibility, this experiment significantly enhances recommendation precision and relevance, easing students' decision-making process and improving the overall quality of course recommendations. A hybrid approach is applied to improve the recommendation quality by combining predictions from all three classifiers through weighted voting. This ensemble method improves overall robustness and accuracy. This approach not only mitigates the cognitive overload faced by students but also significantly improves the quality of recommendations. Our hybrid model represents a substantial advancement in personalized course recommendation technology by demonstrating superior performance across key evaluation metrics such as accuracy, precision, recall, F1-score, ARHR, and NDCG.
... Especially, the current approaches of explicit matrix factorization which often consider only the positive values, and thus the methodology developed in this work cannot be immediately applied in this setting. Indeed, Koren and Bell in [63] have analyzed the relationship between neighborhood and factorization models under explicit settings. It remains to be seen whether the insights gained here can be applied to the explicit setting. ...
Recommender systems have become increasingly important with the rise of the web as a medium for electronic and business transactions. One of the key drivers of this technology is the ease with which users can provide feedback about their likes and dislikes through simple clicks of a mouse. This feedback is commonly collected in the form of ratings, but can also be inferred from a user's browsing and purchasing history. Recommender systems utilize users' historical data to infer customer interests and provide personalized recommendations. The basic principle of recommendations is that significant dependencies exist between user- and item-centric activity, which can be learned in a data-driven manner to make accurate predictions. Collaborative filtering is one family of recommendation algorithms that uses ratings from multiple users to predict missing ratings or uses binary click information to predict potential clicks. However, recommender systems can be more complex and incorporate auxiliary data such as content-based attributes, user interactions, and contextual information.
... Based on the way recommendations are generated, RS can be classified as given in Figure 1. (1) Collaborative filtering: Collaborative filtering [4] recommender systems leverage the preferences and behaviours of other users to suggest items or content to a particular user. By analysing the choices and interactions of a diverse user base, these systems identify patterns and correlations, allowing them to generate personalized recommendations. ...
... Still, methods based on matrix factorization are most broadly used due to their ability to address the efficient way they handle scalability and sparsity issues [52], [53]. One of the most popular techniques applied to solve the matrix factorization problem in this regard is singular value decomposition (SVD) [54]. ...
Recommender systems (RS) are substantial for online shopping or digital content services. However, due to some data characteristics or insufficient historical data, may encounter considerable difficulties impacting the quality of their recommendations. This study introduces the clustering-based frequent pattern mining framework for recommender systems (Clustering-based FPRS) - a novel RS constituting several recommendation strategies leveraging agglomerative clustering and FP-growth algorithms. The developed strategies combine the generated frequent itemsets with collaborative- and content-filtering methods to address the cold-start problem, which occurs whenever a new user or item enters the system. In such cases, the RS has limited information about the new user or object. Thus, the recommendations may be inaccurate. The experimental evaluation on several benchmark datasets showed that Clustering-based FPRS is superior to state-of-the-art and could effectively alleviate the cold-start problem.
... Collaborative recommender systems are tools that provide recommendations based on the preferences and behavior of multiple users [1] . In recent years, these systems have grown in popularity because to their capacity to tailor recommendations for users and enhance the overall user experience. ...
p>Collaborative recommender systems are information filtering systems that seek to predict a user’s rating or preference for an item. They play a vital role in various business use cases, such as personalized recommendations, item ranking and sorting, targeted marketing and promotions, content curation and catalog organization, and feedback analysis and quality control. When evaluating these systems, rating prediction metrics are commonly employed. Efficiency, including the prediction time, is another crucial aspect to consider. In this study, the performance of different algorithms was investigated. The study employed a dataset consisting of e-commerce product ratings and assessed the algorithms based on rating prediction metrics and efficiency. The results demonstrated that each algorithm had its own set of strengths and weaknesses. For the metric of Root Mean Squared Error (RMSE), the BaselineOnly algorithm achieved the lowest mean value. Regarding Mean Absolute Error (MAE), the Singular Value Decomposition with Positive Perturbations Singular Value Decomposition with Positive Perturbations (SVDPP) algorithm exhibited the lowest mean value; Mean Squared Error (MSE) also achieved the lowest mean value. Moreover, the BaselineOnly algorithm showcased superior performance with the lowest mean test times when considering efficiency. Researchers and practitioners can use the findings of this study to select the best algorithm for a particular application. Researchers can develop new algorithms that combine the strengths of different algorithms. Practitioners can also use the findings of this study to tune the parameters of existing algorithms.</p
... It important to note that λi will be zero, if all customers rated product i. This measure has shown improved prediction accuracy in user-based recommendation (Koren and R. Bell, 2015). ...
Purpose: Purpose of this research is to carry out survey on Recommendation systems techniques in Big Data Analytics. This article presents designing of recommender systems and evaluates it with help of various performance metrics in IoT applications. Theoretical framework: With fast development and applications of Internet of Things, large amount of user data is generated and accumulated every day. Growth of media consumption in online social networks is exponential which requires an efficient and effective recommendation system to enhance excellence in experience for users. Recommender systems help users to overcome Information Overload problem by providing them relevant contents. Method/design/approach: The main aspect of recommender system is how to take complete advantage of this ubiquitous data. Recommender system is mainly used to guess or predict users’ interests and make relevant recommendations. Collaborative filtering is the technique that uses the relationships between users and between items in order to build a prediction. Collaborative filtering algorithms are mainly categorized as model-based methods and memory-based methods. In this article, various methods to build recommender system are described. Similarly, Collaborative filtering uses Pearson cosine, cosine vector, Jaccard similarity to identify same users or items. Recommender system has various applications in domain such as healthcare, transportation, agriculture, e-media etc. Findings: Evaluation of recommender system with help of metrics such as Precision and Recall is presented. Comparison of experimental results is presented with help of MAE and RMSE. Recommendation system helps to discover relevant insights and can be one of the vital technologies in future IoT solutions. Research, Practical & social implications: The research makes significant contribution by providing survey of existing recommender systems along with challenges faced while designing effective and accurate recommender. Various similarity measures to find similar users or items are investigated with future pointer direction. Recommender system help in decision making process. Originality/value: The results and conclusion obtained in this research are helpful in development of novel Recommender systems which definitely assist users to overcome Information Overload issue. It helps user to save network load as well.
... The RecSys problem has been approached in a variety of ways in the literature. These techniques are divided into contentbased filtering [31], [32], CF [33], [34], and hybrid [35] algorithms. Popular buckets incorporate heuristic approaches, matrix factorization-based CF approaches [36], [37], neighborhood-based CF approaches, and ML approaches. ...
Smart strategies and intelligent technologies are enabling the designing of a smart learning environment that successfully supports the development of personalized learning and adaptive learning. This trend towards integration is in line with the growing prevalence of Internet of Things (IoT)-enabled smart education systems, which can leverage Machine Learning (ML) techniques to provide Personalized Course Recommendations (PCR) to students. Furthermore, the existing recommendation techniques are based on either explicit or implicit feedback and fail to capture the changes in learners’ preferences while integrating implicit or explicit feedback. To this end, this paper proposes a new model for personalized learning and PCR that is enabled by a smart E-Learning (EL) platform. The model aims to gather data on students’ academic performance, interests, and learning preferences and utilize this data to recommend the courses that will be most beneficial to each student. The proposed approach makes suggestions based on the learner’s interactions with the system and the cosine similarity in related contents by combining explicit (user ratings) and implicit (views and behavior) methodologies. The suggested method makes use of ML algorithms and an EL Recommender System (RecSys) based on Collaborative Filtering (CF).This includes Random Forest Regressor (RFR), Decision Tree Regressor (DTR), K-Nearest Neighbors (KNN), Singular Value Decomposition (SVD), eXtreme Gradient Boosting Regressor (XGBR), and Linear Regression (LR). The proposed solution is benchmarked against existing approaches on both predictive accuracy and running time. Experimental results are conducted based on two benchmark datasets (Coursera and Udemy). The proposed model outperforms existing top-K recommendations techniques in terms of accuracy metrics such as precision@k, Mean Average Precision (MAP)@k, recall@k, Normalized Discounted Cumulative Gain (NDCG)@k, Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) for PCR. From the experiments, it can be shown that SVD can perform well in terms of higher accuracy and MAP and NDCG and lower MAE, RMSE, and MSE values when contrasted to other proposed algorithms because it is better suited to capture complex student-course interactions. The proposed solutions are promising on two different datasets and can be applied to various RecSys domains.
... Then CF recommends items that other users are interested in but the target user has not seen or purchased to the target user. Matrix factorization (MF) [6] is a typical traditional recommendation approach. It can embed users and items in the same vector space. ...
The session-based recommendation system aims to predict the user’s next click based on their previous session sequence. The current studies generally learn user preferences according to the transitions of items in the user’s session sequence. However, other effective information in the session sequence, such as user profiles, is largely ignored which may lead to the model unable to learn the user’s specific preferences. In this paper, we propose SR-HetGNN, a novel session recommendation method that uses a heterogeneous graph neural network (HetGNN) to learn session embeddings and capture the specific preferences of anonymous users. Specifically, SR-HetGNN first constructs heterogeneous graphs containing various types of nodes according to the session sequence, which can capture the dependencies among items, users, and sessions. Second, HetGNN captures the complex transitions between items and learns the item embeddings containing user information. Finally, local and global session embeddings are combined with the attentional network to obtain the final session embedding, considering the influence of users’ long and short-term preferences. SR-HetGNN is shown to be superior to the existing state-of-the-art session-based recommendation methods through extensive experiments over two real large datasets Diginetica and Tmall. The code and datasets can be found in our GitHub repository https://github.com/LeeHY1996/SR-HetGNN.
Artificial intelligence (AI) has played an important role for data-driven decision making in complex engineering problems. However, there has been a huge waste of efforts to configure AI methods (e.g., to select preprocessing and modeling methods, etc.), catering to different contexts (e.g., data analytics objectives, data distributions, etc.). In current practice, data scientists need to manually configure the AI methods in trial-and-errors according to a specific context, including determining the different options of the pipeline components and evaluating the advantages and limitations of an AI method. In this article, we propose a local low-rank response imputation (Lori) method, which will automatically configure AI methods to specific contexts by completing a sparse context-pipeline response matrix. Different from the traditional recommendation systems, Lori performs multivariate partition of the entire context-pipeline response matrix based on the principal Hessian directions of the low-rank imputed response matrix. Thus, the partitioned local low-rank response matrices can be closely modeled to automatically match the AI methods with the datasets. A small-scale and a large-scale case studies in three manufacturing processes demonstrated the merits of the proposed Lori method.
This book covers the Proceedings of the 13th International Conference on Information Systems and Advanced Technologies “ICISAT’2023.” One of the evocative and valuable dimensions of this conference is the way it brings together researchers, scientists, academics, and engineers in the field from different countries and enables discussions and debate of relevant issues, challenges, opportunities, and research findings.
The ICISAT’2023 conference provided a forum for research and developments in the field of information systems and advanced technologies and new trends in developing information systems organizational aspects of their development and intelligent aspects of the final product.
The aim of the ICISAT’2023 is to report progress and development of methodologies, technologies, planning and implementation, tools, and standards in information systems, technologies, and sciences.
ICISAT’2023 aims at addressing issues related to the intelligent information, data science, and decision support system, from multidisciplinary perspectives and to discuss the research, teaching, and professional practice in the field.
The book of ICISAT’2023 includes selected papers from the 13th International Conference on Information Systems and Advanced Technologies “ICISAT’2023,” organized during December 29–30, 2023.
In this book, researchers, professional software, and systems engineers from around the world addressed intelligent information, data science, and decision support system for the conference. The ideas and practical solutions described in the book are the outcome of dedicated research by academics and practitioners aiming to advance theory and practice in this research domain.
The list of topics is in all the areas of modern intelligent information systems and technologies such as neural networks, evolutionary computing, adaptive systems, pervasive system, ubiquitous system, E-learning and teaching, knowledge-based paradigms, learning paradigms, intelligent data analysis, intelligent decision making and support system, intelligent network security, web intelligence, deep learning, natural language processing, image processing, general machine learning, and unsupervised learning.
Providing system-generated explanations for recommendations represents an important step towards transparent and trustworthy recommender systems. Explainable recommender systems provide a human-understandable rationale for their outputs. Over the past two decades, explainable recommendation has attracted much attention in the recommender systems research community. This paper aims to provide a comprehensive review of research efforts on visual explanation in recommender systems. More concretely, we systematically review the literature on explanations in recommender systems based on four dimensions, namely explanation aim, explanation scope, explanation method, and explanation format. Recognizing the importance of visualization, we approach the recommender system literature from the angle of explanatory visualizations, that is using visualizations as a display style of explanation. As a result, we derive a set of guidelines that might be constructive for designing explanatory visualizations in recommender systems and identify perspectives for future work in this field. The aim of this review is to help recommendation researchers and practitioners better understand the potential of visually explainable recommendation research and to support them in the systematic design of visual explanations in current and future recommender systems.
The increase in web data leads to abundant information in the data wealth through different methodologies used in sparse of time. The evaluation is based on achieving profit based on the purchase of the products. The recommendation system (RS) comprises automated strategies to provide appropriate suggestions to the customers with appropriate interaction for the vast range of data. On the other hand, the e-commerce business is a significant player in different scenarios such as photos, magazines, films, and songs for the suggestions such as weddings, financial services, and so on. This paper presents a review of the recommender system for e-commerce applications. The article focuses on the review of customer reviews on e-commerce websites. The e-commerce website provides significant contributions and advancement in product purchases. The purchase intention of the product in e-commerce websites relies on the review of customers. Hence, it is necessary to adopt an effective recommender system based on the trust level of customers. The analysis is based on an examination of the recommendation system (RS) revolution, taxonomy, classification, and dataset used. The extensive review presents existing literature under different classification models for recommendation system (RS).
This paper presents our recent initiatives to foster the discoverability of new releases on the music streaming service Deezer. After introducing our search and recommendation features dedicated to new releases, we outline our shift from editorial to personalized release suggestions using cold start embeddings and contextual bandits. Backed by online experiments, we discuss the advantages of this shift in terms of recommendation quality and exposure of new releases on the service.
Recommender systems play a crucial role in the e-commerce stores, enabling customers to explore products and facilitating the discovery of relevant items. Typical recommender systems are built using n most recent user interactions, where value of n is chosen based on trade-off between incremental gains in performance and compute/memory costs associated with processing long sequences. State-of-the-art recommendation models like Transformers, based on attention mechanism, have quadratic computation complexity with respect to sequence length, thus limiting the length of past customer interactions to be considered for recommendations. Even with the availability of compute resources, it is crucial to design an algorithm that strikes delicate balance between long term and short term information in identifying relevant products for personalised recommendation. Towards this, we propose a novel extension of Memory Networks, a neural network architecture that harnesses external memory to encapsulate information present in lengthy sequential data. The use of memory networks in recommendation use-cases remains limited in practice owing to their high memory cost, large compute requirements and relatively large inference latency, which makes them prohibitively expensive for online stores with millions of users and products. To address these limitations, we propose a novel transformer-based sequential recommendation model GLAD, with external graph-based memory that dynamically scales user memory by adjusting the memory size according to the user’s history, while facilitating the flow of information between users with similar interactions. We establish the efficacy of the proposed model by benchmarking on multiple public datasets as well as an industry dataset against state-of-the-art sequential recommendation baselines.
Purpose: Purpose of this research is to carry out survey on Recommendation systems techniques in Big Data Analytics. This article presents designing of recommender systems and evaluates it with help of various performance metrics in IoT applications. Theoretical framework: With fast development and applications of Internet of Things, large amount of user data is generated and accumulated every day. Growth of media consumption in online social networks is exponential which requires an efficient and effective recommendation system to enhance excellence in experience for users. Recommender systems help users to overcome Information Overload problem by providing them relevant contents. Method/design/approach: The main aspect of recommender system is how to take complete advantage of this ubiquitous data. Recommender system is mainly used to guess or predict users’ interests and make relevant recommendations. Collaborative filtering is the technique that uses the relationships between users and between items in order to build a prediction. Collaborative filtering algorithms are mainly categorized as model-based methods and memory-based methods. In this article, various methods to build recommender system are described. Similarly, Collaborative filtering uses Pearson cosine, cosine vector, Jaccard similarity to identify same users or items. Recommender system has various applications in domain such as healthcare, transportation, agriculture, e-media etc. Findings: Evaluation of recommender system with help of metrics such as Precision and Recall is presented. Comparison of experimental results is presented with help of MAE and RMSE. Recommendation system helps to discover relevant insights and can be one of the vital technologies in future IoT solutions. Research, Practical & social implications: The research makes significant contribution by providing survey of existing recommender systems along with challenges faced while designing effective and accurate recommender. Various similarity measures to find similar users or items are investigated with future pointer direction. Recommender system help in decision making process. Originality/value: The results and conclusion obtained in this research are helpful in development of novel Recommender systems which definitely assist users to overcome Information Overload issue. It helps user to save network load as well.
This chapter explores the importance of personalization and recommendation algorithms in OTT era, emphasizing their role in enhancing content discovery and customizing user experiences. Crucial techniques like collaborative filtering and content-based filtering underpin these algorithms, which ensues personalized user experiences. Recommendation algorithms shape media consumption patterns, content discovery, influencing user behavior, cross-platform consumption and binge-watching habits. This chapter also paid attention to acknowledging ethical considerations like privacy concerns and algorithmic bias. Additionally, it also explores the challenges and opportunities for content creators in catering to algorithmic preferences, along with significance of balancing effective ad targeting and user privacy in personalized advertising. Improving and assessing recommendation algorithms using different metrics and feedback loops is important, however future trends concentrate on contextual personalization and adaptive experiences, enhancing user's entertainment journey.
Graph convolutional networks (GCNs)GCN have been widely used to learn high-quality representations (a.k.a. embeddings) from multiorder neighbors in recommendation tasks. However, many existing graph convolutional network (GCN)-based methods learn user and item embeddings in the user–item interaction bipartite graph indistinguishably, ignoring the inherent heterogeneity of the bipartite graph, i.e., users and items are two distinct types of entities. This article explores in depth the high-order connections for user and his (her) neighbor items. We propose an innovative model, neighbor item embedding-aware graph convolutional network (NIE-GCN). As opposed to previous GCN-based approaches, NIE-GCN employs a novel dual user aggregationdual user aggregation (DUA) scheme and a neighbor-aware attention mechanism to construct user embeddings and distinguish the contribution of different neighbor nodes. In addition, we propagate information in an alternating manner to eliminate the effects of heterogeneity of user–item interaction bipartite graph. According to detailed experiments on three large-scale datasets, the proposed NIE-GCN significantly outperforms state-of-the-art approaches on the Top-
N
recommendation task while reducing model parameters by about half. Further analyses show the effectiveness and rationality of dual user aggregationDUA and neighbor-aware attention mechanism.
By providing personalized suggestions to users, recommender systems have become essential to numerous online platforms. Collaborative filtering, particularly graph-based approaches using Graph Neural Networks (GNNs), have demonstrated great results in terms of recommendation accuracy. However, accuracy may not always be the most important criterion for evaluating recommender systems' performance, since beyond-accuracy aspects such as recommendation diversity, serendipity, and fairness can strongly influence user engagement and satisfaction. This review paper focuses on addressing these dimensions in GNN-based recommender systems, going beyond the conventional accuracy-centric perspective. We begin by reviewing recent developments in approaches that improve not only the accuracy-diversity trade-off but also promote serendipity, and fairness in GNN-based recommender systems. We discuss different stages of model development including data preprocessing, graph construction, embedding initialization, propagation layers, embedding fusion, score computation, and training methodologies. Furthermore, we present a look into the practical difficulties encountered in assuring diversity, serendipity, and fairness, while retaining high accuracy. Finally, we discuss potential future research directions for developing more robust GNN-based recommender systems that go beyond the unidimensional perspective of focusing solely on accuracy. This review aims to provide researchers and practitioners with an in-depth understanding of the multifaceted issues that arise when designing GNN-based recommender systems, setting our work apart by offering a comprehensive exploration of beyond-accuracy dimensions.
In the present paper we propose a hybrid recommender system dealing with implicit feedbacks in the domain of fashion retail. The proposed architecture is based on a collaborative-filtering module taking into account the fact that users feedbacks are not explicit scores about the items, but are obtained through user interactions with the products in terms of number of purchases; moreover, a second module provides a knowledge-based contextual post-filtering, based on both customer-oriented and business-oriented objectives. We finally present a case study where “look-oriented” recommendations have been implemented for a specific fashion retail brand.
Sequential recommendation has drawn a lot of attention due to its good performance in recent years. The temporal order of user interactions cannot be ignored in sequential recommendation, and user preferences are constantly changing over time. The application of deep neural network in sequential recommendation has achieved many remarkable results, especially self-attention based methods. However, previous works mainly focused on item-item temporal information of the sequence while ignoring the latent collaborative relations in user-item interactions. Therefore, we propose a new method named Collaborative-Temporal modeling for Sequential Recommendation (CTSR) to learn both collaborative relations and temporal information. The proposed CTSR method introduces a graph convolutional network to learn the user-item collaborative relations while using self-attention to capture item-item temporal information. We apply focal loss to reduce the loss contribution of the easy samples and increase the contribution of the hard samples, to train the model more effectively. We extract a portion of item-item pairs that are most valuable, and then feed these pairs as augmented information into adjacency matrix of the graph neural network. More importantly, it is the first work to encode the relative positions of items into their embeddings in sequential recommendation. The experimental results show that CTSR surpasses previous works on three real-world datasets.
The recommendation system can alleviate the problem of ?information overload?, tap the potential value of data, push personalized information to users in need, and improve information utilization. Sequence recommendation has become a hot research direction because of its practicality and high precision. Deep Neural Networks (DNN) have the natural advantage of capturing comprehensive relations among different entities, thus almost occupying a dominant position in sequence recommendation in the past few years. However, as Deep Learning (DL)-based methods are widely used to model local preferences under user behavior sequences, the global preference modeling of users is often underestimated, and usually, only some simple and crude user latent representations are introduced. Therefore, this paper proposes a sequential recommendation based on Fusing Session-Aware models and Self-Attention networks (FSASA). Specifically, we use the Self-Attentive Sequential Recommendation (SASRec) model as a global representation learning module to capture long-term preferences under user behavior sequences and further propose an improved session-aware sequential recommendation model as a local learning representation module from user model the user?s dynamic preferences in the historical behavior, and finally use the Gated Recurrent Unit (GRU) module to calculate their weights. Experiments on three widely used recommendation atasets show that FSASA outperforms state-of-the-art baselines on two commonly used metrics.
Recommender system (RS) aims to predict user preferences based on automatic data acquisition, and those collected data assist in achieving the final decision. However, RS suffers from data sparsity issues over the newly launched system, and the lack of time to deal with the massive data is also a challenging factor. To acquire proper outcomes, cross-domain RS intends to transfer knowledge from the specific domain with quality enriched data to help recommendations to the target domains. The entities may or may not be overlapped, and it is common for the entities of two domains to be overlapped. These overlapping entities may show variations in their target domain, and avoiding these issues leads to distorted prediction outcomes over the cross-domain RS. To address these issues, this research concentrates on modeling and efficient cross-domain RS using the generative and discriminative adversarial network (CRS-GDAN) model for kernel-based transfer modeling. Domain specific is considered to handle the feature space of overlapped entities, and transfer computation is adopted to handle the overlapping and non-overlapping entity correlation among the domains. Based on the anticipated concept, knowledge transfer is achieved rigorously even in the case of overlapping entities, thus diminishing the data sparsity issues. The experimentation is performed using an available online dataset, and the model attains a 20% better outcome than other approaches. The outcomes specify that the knowledge transfer from source to destination target is advantageous even in overlapping issues.
Fashion recommendation is a key research field in computational fashion research and has attracted considerable interest in the computer vision, multimedia, and information retrieval communities in recent years. Due to the great demand for applications, various fashion recommendation tasks, such as personalized fashion product recommendation, complementary (mix-and-match) recommendation, and outfit recommendation, have been posed and explored in the literature. The continuing research attention and advances impel us to look back and in-depth into the field for a better understanding. In this paper, we comprehensively review recent research efforts on fashion recommendation from a technological perspective. We first introduce fashion recommendation at a macro level and analyse its characteristics and differences with general recommendation tasks. We then clearly categorize different fashion recommendation efforts into several sub-tasks and focus on each sub-task in terms of its problem formulation, research focus, state-of-the-art methods, and limitations. We also summarize the datasets proposed in the literature for use in fashion recommendation studies to give readers a brief illustration. Finally, we discuss several promising directions for future research in this field. Overall, this survey systematically reviews the development of fashion recommendation research. It also discusses the current limitations and gaps between academic research and the real needs of the fashion industry. In the process, we offer a deep insight into how the fashion industry could benefit from the computational technologies of fashion recommendation.
Neighborhood-based algorithms are frequently used mod-ules of recommender systems. Usually, the choice of the similarity measure used for evaluation of neighborhood re-lationships is crucial for the success of such approaches. In this article we propose a way to calculate similarities by for-mulating a regression problem which enables us to extract the similarities from the data in a problem-specific way. An-other popular approach for recommender systems is regular-ized matrix factorization (RMF). We present an algorithm – neighborhood-aware matrix factorization – which efficiently includes neighborhood information in a RMF model. This leads to increased prediction accuracy. The proposed meth-ods are tested on the Netflix dataset.
We describe the TiVo television show collaborative recommendation system which has been fielded in over one million TiVo clients for four years. Over this install base, TiVo currently has approximately 100 million ratings by users over approximately 30,000 distinct TV shows and movies. TiVo uses an item-item (show to show) form of collaborative filtering which obviates the need to keep any persistent memory of each user's viewing preferences at the TiVo server. Taking advantage of TiVo's client-server architecture has produced a novel collaborative filtering system in which the server does a minimum of work and most work is delegated to the numerous clients. Nevertheless, the server-side processing is also highly scalable and parallelizable. Although we have not performed formal empirical evaluations of its accuracy, internal studies have shown its recommendations to be useful even for multiple user households. TiVo's architecture also allows for throttling of the server so if more server-side resources become available, more correlations can be computed on the server allowing TiVo to make recommendations for niche audiences.
Memory-based methods for collaborative filtering predict new ratings by averaging (weighted) ratings between, respectively, pairs of similar users or items. In practice, a large number of ratings from similar users or similar items are not available, due to the sparsity inherent to rating data. Consequently, prediction quality can be poor. This paper re-formulates the memory-based collaborative filtering problem in a generative probabilistic framework, treating individual user-item ratings as predictors of missing ratings. The final rating is estimated by fusing predictions from three sources: predictions based on ratings of the same item by other users, predictions based on different item ratings made by the same user, and, third, ratings predicted based on data from other but similar users rating other but similar items. Existing user-based and item-based approaches correspond to the two simple cases of our framework. The complete model is however more robust to data sparsity, because the different types of ratings are used in concert, while additional ratings from similar users towards similar items are employed as a background model to smooth the predictions. Experiments demonstrate that the proposed methods are indeed more robust against data sparsity and give better recommendations.
Collaborative filtering (CF) approaches proved to be effec- tive for recommender systems in predicting user preferences in item selection using known user ratings of items. This subfield of machine learning has gained a lot of popularity with the Netflix Prize competition started in October 2006. Two major approaches for this problem are matrix factor- ization (MF) and the neighbor based approach (NB). In this work, we propose various variants of MF and NB that can boost the performance of the usual ensemble based scheme. First, we investigate various regularization scenarios for MF. Second, we introduce two NB methods: one is based on cor- relation coefficients and the other on linear least squares. At the experimentation part, we show that the proposed approaches compare favorably with existing ones in terms of prediction accuracy and/or required training time. We present results of blending the proposed methods.
The Netflix Prize is a collaborative filtering problem. This subfield of machine learning became popular in the late 1990s with the spread of online services that used recom- mendation systems (e.g. Amazon, Yahoo! Music, and of course Netflix). The aim of such a system is to predict what items a user might like based on his/her and other users' previous ratings. The Netflix Prize dataset is much larger than former benchmark datasets, therefore the scala- bility of the algorithms is a must. This paper describes the major components of our blending based solution, called the Gravity Recommendation System (GRS). In the Net- flix Prize contest, it attained RMSE 0.8743 as of November 2007. We now compare the effectiveness of some selected individual and combined approaches on a particular subset of the Prize dataset, and discuss their important features and drawbacks.
Collaborative filtering aims at learning predictive models of user preferences, interests or behavior from community data, that is, a database of available user preferences. In this article, we describe a new family of model-based algorithms designed for this task. These algorithms rely on a statistical modelling technique that introduces latent class variables in a mixture model setting to discover user communities and prototypical interest profiles. We investigate several variations to deal with discrete and continuous response variables as well as with different objective functions. The main advantages of this technique over standard memory-based methods are higher accuracy, constant time prediction, and an explicit and compact model representation. The latter can also be used to mine for user communitites. The experimental evaluation shows that substantial improvements in accucracy over existing methods and published results can be obtained.
This paper provides a comprehensive review of explanations in recommender systems. We highlight seven possible advantages of an explanation facility, and describe how existing measures can be used to evaluate the quality of explanations. Since explanations are not independent of the recommendation process, we consider how the ways recommendations are presented may affect explanations. Next, we look at different ways of interacting with explanations. The paper is illustrated with examples of explanations throughout, where possible from existing applications.
We investigate the use of dimensionality reduction to improve performance for a new class of data analysis software called "recommender systems". Recommender systems apply knowledge discovery techniques to the problem of making product recommendations during a live customer interaction. These systems are achieving widespread success in E-commerce nowadays, especially with the advent of the Internet. The tremendous growth of customers and products poses three key challenges for recommender systems in the E-commerce domain. These are: producing high quality recommendations, performing many recommendations per second for millions of customers and products, and achieving high coverage in the face of data sparsity. One successful recommender system technology is collaborative filtering , which works by matching customer preferences to other customers in making recommendations. Collaborative filtering has been shown to produce high quality recommendations, but the performance degrades with ...
Collaborative filtering (CF) is valuable in e-commerce, and for direct recommendations for music, movies, news etc. But today's systems have several disadvantages, including privacy risks. As we move toward ubiquitous computing, there is a great potential for individuals to share all kinds of information about places and things to do, see and buy, but the privacy risks are severe. In this paper we describe a new method for collaborative filtering which protects the privacy of individual data. The method is based on a probabilistic factor analysis model. Privacy protection is provided by a peer-to-peer protocol which is described elsewhere, but outlined in this paper. The factor analysis approach handles missing data without requiring default values for them. We give several experiments that suggest that this is most accurate method for CF to date. The new algorithm has other advantages in speed and storage over previous algorithms. Finally, we suggest applications of the approach to other kinds of statistical analyses of survey or questionaire data.
Recommender systems apply knowledge discovery techniques to the problem of making personalized recommendations for information, products or services during a live interaction. These systems, especially the k-nearest neighbor collaborative filtering based ones, are achieving widespread success on the Web. The tremendous growth in the amount of available information and the number of visitors to Web sites in recent years poses some key challenges for recommender systems. These are: producing high quality recommendations, performing many recommendations per second for millions of users and items and achieving high coverage in the face of data sparsity. In traditional collaborative filtering systems the amount of work increases with the number of participants in the system. New recommender system technologies are needed that can quickly produce high quality recommendations, even for very large-scale problems. To address these issues we have explored item-based collaborative filtering techniques. Itembased techniques first analyze the user-item matrix to identify relationships between different items, and then use these relationships to indirectly compute recommendations for users. In this paper we analyze different item-based recommendation generation algorithms. We look into different techniques for computing item-item similarities (e.g., item-item correlation vs. cosine similarities between item vectors) and different techniques for obtaining recommendations from them (e.g., weighted sum vs. regression model). Finally, we experimentally evaluate our results and compare them to the basic k-nearest neighbor approach. Our experiments suggest that item-based algorithms provide dramatically better performance than user-based algorithms, while at the same time p...
Automated collaborative filtering (ACF) systems predict a person's affinity for items or information by connecting that person's recorded interests with the recorded interests of a community of people and sharing ratings between likeminded persons. However, current recommender systems are black boxes, providing no transparency into the working of the recommendation. Explanations provide that transparency, exposing the reasoning and data behind a recommendation. In this paper, we address explanation interfaces for ACF systems -- how they should be implemented and why they should be implemented. To explore how, we present a model for explanations based on the user's conceptual model of the recommendation process. We then present experimental results demonstrating what components of an explanation are the most compelling. To address why, we present experimental evidence that shows that providing explanations can improve the acceptance of ACF systems. We also describe some initial explor...
Recommender systems provide users with personalized suggestions for products or services. These systems often rely on collaborating filtering (CF), where past transactions are analyzed in order to establish connections between users and products. The most common approach to CF is based on neighborhood models, which originate from similarities between products or users. In this work we introduce a new neighborhood model with an improved prediction accuracy. Unlike previous approaches that are based on heuristic similarities, we model neighborhood relations by minimizing a global cost function. Further accuracy improvements are achieved by extending the model to exploit both explicit and implicit feedback by the users. Past models were limited by the need to compute all pairwise similarities between items or users, which grow quadratically with input size. In particular, this limitation vastly complicates adopting user similarity models, due to the typical large number of users. Our new model solves these limitations by factoring the neighborhood model, thus making both item-item and user-user implementations scale linearly with the size of the data. The methods are tested on the Netflix data, with encouraging results.
We describe latent Dirichlet allocation (LDA), a generative probabilistic model for collections of discrete data such as text corpora. LDA is a three-level hierarchical Bayesian model, in which each item of a collection is modeled as a finite mixture over an underlying set of topics. Each topic is, in turn, modeled as an infinite mixture over an underlying set of topic probabilities. In the context of text modeling, the topic probabilities provide an explicit representation of a document. We present efficient approximate inference techniques based on variational methods and an EM algorithm for empirical Bayes parameter estimation. We report results in document modeling, text classification, and collaborative filtering, comparing to a mixture of unigrams model and the probabilistic LSI model.
A key part of a recommender system is a collaborative filter-ing algorithm predicting users' preferences for items. In this paper we describe different efficient collaborative filtering techniques and a framework for combining them to obtain a good prediction. The methods described in this paper are the most im-portant parts of a solution predicting users' preferences for movies with error rate 7.04% better on the Netflix Prize dataset than the reference algorithm Netflix Cinematch. The set of predictors used includes algorithms suggested by Netflix Prize contestants: regularized singular value de-composition of data with missing values, K-means, postpro-cessing SVD with KNN. We propose extending the set of predictors with the following methods: addition of biases to the regularized SVD, postprocessing SVD with kernel ridge regression, using a separate linear model for each movie, and using methods similar to the regularized SVD, but with fewer parameters. All predictors and selected 2-way interactions between them are combined using linear regression on a holdout set.
Rating prediction is an important application, and a popular research topic
in collaborative filtering. However, both the validity of learning algorithms,
and the validity of standard testing procedures rest on the assumption that
missing ratings are missing at random (MAR). In this paper we present the
results of a user study in which we collect a random sample of ratings from
current users of an online radio service. An analysis of the rating data
collected in the study shows that the sample of random ratings has markedly
different properties than ratings of user-selected songs. When asked to report
on their own rating behaviour, a large number of users indicate they believe
their opinion of a song does affect whether they choose to rate that song, a
violation of the MAR condition. Finally, we present experimental results
showing that incorporating an explicit model of the missing data mechanism can
lead to significant improvements in prediction performance on the random sample
of ratings.
A new method for automatic indexing and retrieval is described. The approach is to take advantage of implicit higher-order structure in the association of terms with documents (“semantic structure”) in order to improve the detection of relevant documents on the basis of terms found in queries. The particular technique used is singular-value decomposition, in which a large term by document matrix is decomposed into a set of ca. 100 orthogonal factors from which the original matrix can be approximated by linear combination. Documents are represented by ca. 100 item vectors of factor weights. Queries are represented as pseudo-document vectors formed from weighted combinations of terms, and documents with supra-threshold cosine values are returned. initial tests find this completely automatic method for retrieval to be promising.
Collaborative filtering (CF) is one of the most popular recommender system technologies, and utilizes the known preferences of a group of users to predict the unknown preference of a new user. However, the existing CF techniques has the drawback that it requires the entire existing data be maintained and analyzed repeatedly whenever new user ratings are added. To avoid such a problem, Eigentaste, a CF approach based on the principal component analysis (PCA), has been proposed. However, Eigentaste requires that each user rate every item in the so called gauge set for executing PCA, which may not be always feasible in practice. Developed in this article is an iterative PCA approach in which no gauge set is required, and singular value decomposition is employed for estimating missing ratings and dimensionality reduction. Principal component values for users in reduced dimension are used for clustering users. Then, the proposed approach is compared to Eigentaste in terms of the mean absolute error of prediction using the Jester, MovieLens, and EachMovie data sets. Experimental results show that the proposed approach, even without a gauge set, performs slightly better than Eigentaste regardless of the data set and clustering method employed, implying that it can be used as a useful alternative when defining a gauge set is neither possible nor practical.
Recommender systems provide users with personalized suggestions for products or services. These systems often rely on Collaborating Filtering (CF), where past transactions are analyzed in order to establish connections between users and products. The two more successful approaches to CF are latent factor models, which directly profile both users and products, and neighborhood models, which analyze similarities between products or users. In this work we introduce some innovations to both approaches. The factor and neighborhood models can now be smoothly merged, thereby building a more accurate combined model. Further accuracy improvements are achieved by extending the models to exploit both explicit and implicit feedback by the users. The methods are tested on the Netflix data. Results are better than those previously published on that dataset. In addition, we suggest a new evaluation metric, which highlights the differences among methods, based on their performance at a top-K recommendation task.
The collaborative filtering approach to recommender system s pre- dicts user preferences for products or services by learning past user- item relationships. In this work, we propose novel algorithms for predicting user ratings of items by integrating complementary mod- els that focus on patterns at different scales. At a local sca le, we use a neighborhood-based technique that infers ratings from observed ratings by similar users or of similar items. Unlike previou s local approaches, our method is based on a formal model that accounts for interactions within the neighborhood, leading to improved esti- mation quality. At a higher, regional, scale, we use SVD-like ma- trix factorization for recovering the major structural pat terns in the user-item rating matrix. Unlike previous approaches that require imputations in order to fill in the unknown matrix entries, ou r new iterative algorithm avoids imputation. Because the models involve estimation of millions, or even billions, of parameters, sh rinkage of estimated values to account for sampling variability proves crucial to prevent overfitting. Both the local and the regional appro aches, and in particular their combination through a unifying model, com- pare favorably with other approaches and deliver substantially bet- ter results than the commercial Netflix Cinematch recommend er system on a large publicly available data set.
Most of the existing approaches to collab- orative ltering cannot handle very large data sets. In this paper we show how a class of two-layer undirected graphical mod- els, called Restricted Boltzmann Machines (RBM's), can be used to model tabular data, such as user's ratings of movies. We present ecien t learning and inference procedures for this class of models and demonstrate that RBM's can be successfully applied to the Netix data set, containing over 100 mil- lion user/movie ratings. We also show that RBM's slightly outperform carefully-tuned SVD models. When the predictions of mul- tiple RBM models and multiple SVD models are linearly combined, we achieve an error rate that is well over 6% better than the score of Netix's own system.
Several approaches to collaborative filtering have been stud- ied but seldom have studies been reported for large (several million users and items) and dynamic (the underlying item set is continually changing) settings. In this paper we de- scribe our approach to collaborative filtering for generating personalized recommendations for users of Google News. We generate recommendations using three approaches: collabo- rative filtering using MinHash clustering, Probabilistic La- tent Semantic Indexing (PLSI), and covisitation counts. We combine recommendations from different algorithms using a linear model. Our approach is content agnostic and con- sequently domain independent, making it easily adaptable for other applications and languages with minimal effort. This paper will describe our algorithms and system setup in detail, and report results of running the recommendations engine on Google News.
This article outlines the overall strategy and summarizes a few key innovations of the team that won the first Netflix progress prize.
Recommender systems provide users with personalized suggestions for products or services. These systems often rely on collaborating filtering (CF), where past transactions are analyzed in order to establish connections between users and products. The most common approach to CF is based on neighborhood models, which originate from similarities between products or users. In this work we introduce a new neighborhood model with an improved prediction accuracy. Unlike previous approaches that are based on heuristic similarities, we model neighborhood relations by minimizing a global cost function. Further accuracy improvements are achieved by extending the model to exploit both explicit and implicit feedback by the users. Past models were limited by the need to compute all pairwise similarities between items or users, which grow quadratically with input size. In particular, this limitation vastly complicates adopting user similarity models, due to the typical large number of users. Our new model solves these limitations by factoring the neighborhood model, thus making both item-item and user-user implementations scale linearly with the size of the data. The methods are tested on the Netflix data, with encouraging results.
We describe latent Dirichlet allocation (LDA), a generative probabilistic model for collections of discrete data such as text corpora. LDA is a three-level hierarchical Bayesian model, in which each item of a collection is modeled as a finite mixture over an underlying set of topics. Each topic is, in turn, modeled as an infinite mixture over an underlying set of topic probabilities. In the context of text modeling, the topic probabilities provide an explicit representation of a document. We present efficient approximate inference techniques based on variational methods and an EM algorithm for empirical Bayes parameter estimation. We report results in document modeling, text classification, and collaborative filtering, comparing to a mixture of unigrams model and the probabilistic LSI model.
Recommender systems based on collaborative filtering predict user preferences for products or services by learning past user-item relationships. A predominant approach to collaborative filtering is neighborhood based ("k-nearest neighbors"), where a user-item preference rating is interpolated from ratings of similar items and/or users. We enhance the neighborhood-based approach leading to substantial improvement of prediction accuracy, without a meaningful increase in running time. First, we remove certain so-called "global effects" from the data to make the ratings more comparable, thereby improving interpolation accuracy. Second, we show how to simultaneously derive interpolation weights for all nearest neighbors, unlike previous approaches where each weight is computed separately. By globally solving a suitable optimization problem, this simultaneous interpolation accounts for the many interactions between neighbors leading to improved accuracy. Our method is very fast in practice, generating a prediction in about 0.2 milliseconds. Importantly, it does not require training many parameters or a lengthy preprocessing, making it very practical for large scale applications. Finally, we show how to apply these methods to the perceivably much slower user-oriented approach. To this end, we suggest a novel scheme for low dimensional embedding of the users. We evaluate these methods on the netflix dataset, where they deliver significantly better results than the commercial netflix cinematch recommender system.
Recommendation algorithms are best known for their use on e-commerce Web sites, where they use input about a customer's interests to generate a list of recommended items. Many applications use only the items that customers purchase and explicitly rate to represent their interests, but they can also use other attributes, including items viewed, demographic data, subject interests, and favorite artists. At Amazon.com, we use recommendation algorithms to personalize the online store for each customer. The store radically changes based on customer interests, showing programming titles to a software engineer and baby toys to a new mother. There are three common approaches to solving the recommendation problem: traditional collaborative filtering, cluster models, and search-based methods. Here, we compare these methods with our algorithm, which we call item-to-item collaborative filtering. Unlike traditional collaborative filtering, our algorithm's online computation scales independently of the number of customers and number of items in the product catalog. Our algorithm produces recommendations in real-time, scales to massive data sets, and generates high quality recommendations.
Can implicit feedback substitute for explicit ratings in recommender systems? If so, we could avoid the difficulties associated with gathering explicit ratings from users. How, then, can we capture useful information unobtrusively, and how might we use that information to make recommendations ? In this paper we identify three types of implicit feedback and suggest two strategies for using implicit feedback to make recommendations. Introduction Recommender systems exploit ratings provided by an entire user population to reshape an information space for the benefit of one or more individuals (Oard, 1997b). In research systems, these ratings are often provided explicitly by each user using one or more ordinal or qualitative scales. The cognitive load effort to assign accurate ratings acts as disincentive, making it difficult to assemble large user populations and contributing to data sparsity within existing populations. Implicit feedback techniques seek to avoid this bottlene...
Netflix Update: Try This At Home
- S Funk
The Netflix Prize”, KDD Cup and Workshop
- J Bennet
- S Lanning