Knowledge Of User Preferences Research Articles

Complementary Fashion Recommendation using Compatible Modelling

The last two years of COVID-19, repeated lockdowns, and plain fear of stepping out of the house have fast-paced the adoption of eCommerce by many decades. Total online spending in May 2020 hit $82.1 billion (about $210 per person in the US), up 77% year-over-year. Even though it is good news for retailers as they do not need to invest in numerous brick-and-mortar stores, they face the conundrum of serving an invisible customer. They cannot see the customer, do not have any insight into customer preferences, and might have a limited transaction history for the customers as well as the product sales. Walking into a store, salespeople see the customer and can recommend clothes based on the customer's looks. Our motivation is to create a cold start recommendation engine that does not need any knowledge of user preferences, user history, item propensity, or any other data to recommend products to the customer. In this paper we are trying to solve two problems, given a product, what are the similar products and what are the complementary products that will complete the outfit.

Knowledge-based recommender systems: overview and research directions.

Recommender systems are decision support systems that help users to identify items of relevance from a potentially large set of alternatives. In contrast to the mainstream recommendation approaches of collaborative filtering and content-based filtering, knowledge-based recommenders exploit semantic user preference knowledge, item knowledge, and recommendation knowledge, to identify user-relevant items which is of specific relevance when dealing with complex and high-involvement items. Such recommenders are primarily applied in scenarios where users specify (and revise) their preferences, and related recommendations are determined on the basis of constraints or attribute-level similarity metrics. In this article, we provide an overview of the existing state-of-the-art in knowledge-based recommender systems. Different related recommendation techniques are explained on the basis of a working example from the domain of survey software services. On the basis of our analysis, we outline different directions for future research.

KDRank: Knowledge-driven user-aware POI recommendation

Accurate user modeling is crucial for point-of-interest (POI) recommendation as it can significantly improve user satisfaction with recommended POIs and enrich user experience. However, existing methods typically rely on simple time-series models for user check-in sequences, which ignore similar information of global users and fail to capture the user-preference knowledge hidden in complex social networks. To address this issue, we propose a novel knowledge-driven and user-aware POI recommendation method called KDRank. First, we construct a knowledge graph containing users’ personal attributes for POI recommendation, which can reflect users’ historical check-in preferences. Second, we derive users’ knowledge representations using a cross-embedding method, which facilitates feature interaction by sharing information between segments of knowledge representations to achieve a more precise representation of low-dimensional embedding. Third, we propose a knowledge aggregation module to combine users’ knowledge and historical check-in features to achieve knowledge enhancement of check-in data. Furthermore, to enhance global user awareness of our model, we introduce an attention mechanism that focuses on the most similar and significant users in the global context. It allows KDRank to capture more personalized user preferences and increases the precision of POI recommendations. The effectiveness of the proposed method was evaluated on two real datasets, and the results indicated its ability to increase the POI recommendation accuracy. The code associated with this study is available at https://github.com/itshardtocode/KDRank.

A multi-graph neural group recommendation model with meta-learning and multi-teacher distillation

Group recommendation has garnered significant attention recently, aiming to suggest items of interest to groups. Most deep learning-based approaches for group recommendation focus on learning group representation through group–user interactions or by aggregating user preferences. However, these approaches face challenges due to the complexity of group representation learning and the limited availability of group–item interactions. To address these difficulties, we propose a multi-graph neural group recommendation model with meta-learning and multi-teacher distillation. Our model consists of three stages: multiple graphs representation learning (MGRL), meta-learning-based knowledge transfer (MLKT), and multi-teacher distillation (MTD). In the MGRL stage, we construct two bipartite graphs and an undirected weighted graph based on group–user interactions, group–item interactions, and similarity between groups, respectively. We then utilize a linear variational graph auto-encoder network to learn the semantic features of groups, users, and items. In the MLKT stage, we introduce a multilayer perceptron (MLP) network for the recommendation task and design a meta-learning-based knowledge transfer algorithm to train an initialization parameter that incorporates user preference knowledge. In the MTD stage, we use the trained parameters as initialization parameters for all teacher networks. We evenly divide the group–item interactions among the teachers and sequentially train the teacher networks. Finally, we employ a multi-teacher distillation approach to train student networks with superior performance. Furthermore, we conduct extensive experiments on six real-world datasets to evaluate the effectiveness of our proposed model. The experimental findings demonstrate that our model, in comparison to state-of-the-art approaches, achieves superior performance in terms of group recommendation.

Modeling User Reviews through Bayesian Graph Attention Networks for Recommendation

Recommender systems relieve users from cognitive overloading by predicting preferred items for users. Due to the complexity of interactions between users and items, graph neural networks (GNN) use graph structures to effectively model user–item interactions. However, existing GNN approaches have the following limitations: (1) User reviews are not adequately modeled in graphs. Therefore, user preferences and item properties that are described in user reviews are lost for modeling users and items; and (2) GNNs assume deterministic relations between users and items, which lack the stochastic modeling to estimate the uncertainties in neighbor relations. To mitigate the limitations, we build tripartite graphs to model user reviews as nodes that connect with users and items. We estimate neighbor relations with stochastic variables and propose a Bayesian graph attention network (i.e., ContGraph) to accurately predict user ratings. ContGraph incorporates the prior knowledge of user preferences to regularize the posterior inference of attention weights. Our experimental results show that ContGraph significantly outperforms 13 state-of-the-art models and improves the best performing baseline (i.e., ANR) by 5.23% on 25 datasets in the five-core version. Moreover, we show that correctly modeling the semantics of user reviews in graphs can help express the semantics of users and items.

RESUS: Warm-up Cold Users via Meta-learning Residual User Preferences in CTR Prediction

Click-through Rate (CTR) prediction on cold users is a challenging task in recommender systems. Recent researches have resorted to meta-learning to tackle the cold-user challenge, which either perform few-shot user representation learning or adopt optimization-based meta-learning. However, existing methods suffer from information loss or inefficient optimization process, and they fail to explicitly model global user preference knowledge, which is crucial to complement the sparse and insufficient preference information of cold users. In this article, we propose a novel and efficient approach named RESUS, which decouples the learning of global preference knowledge contributed by collective users from the learning of residual preferences for individual users. Specifically, we employ a shared predictor to infer basis user preferences, which acquires global preference knowledge from the interactions of different users. Meanwhile, we develop two efficient algorithms based on the nearest neighbor and ridge regression predictors, which infer residual user preferences via learning quickly from a few user-specific interactions. Extensive experiments on three public datasets demonstrate that our RESUS approach is efficient and effective in improving CTR prediction accuracy on cold users, compared with various state-of-the-art methods.

A Novel Group Recommendation Model With Two-Stage Deep Learning

Group recommendation has recently drawn a lot of attention to the recommender system community. Currently, several deep learning-based approaches are leveraged to learn preferences of groups for items and predict next items in which groups may be interested. Yet, their recommendation performance is still unsatisfactory due to sparse group–item interactions. To address this challenge, this study presents a novel model, called group recommendation model with two-stage deep learning (GRMTDL), which encompasses two sequential stages: 1) group representation learning (GRL) and 2) group preference learning (GPL). In GRL, we first construct an undirected tripartite graph over group–user–item interactions, and then employ it to accurately learn group semantic features through a spatial-based variational graph autoencoder network. While in GPL, we first introduce a dual PL-network that contains two structure-sharing subnetworks: 1) group PL-network employed for GPL and 2) user PL-network utilized for user preference learning. Then, we design a novel layered transfer learning (LTL) method to learn group preferences by alternately optimizing these two subnetworks. In particular, it can effectively absorb knowledge of user preferences into the process of GPL. Furthermore, extensive experiments on four real-world datasets demonstrate that the proposed GRMTDL model outperforms the state-of-the-art baselines for group recommendation.

Entity knowledge transfer-oriented dual-target cross-domain recommendations

Cross-domain recommendations can assist users in selecting suitable items in the target domain by aggregating or transferring the abundant available data from the auxiliary domain, which has gradually become a promising research area. However, most existing recommendations are only single-target cross-domain recommendations, and ignore the analysis of entity-side (user side or item side) knowledge, resulting in recommender systems suffering from low accuracy of rating predictions. To address these concerns, we propose a novel Entity Knowledge Transfer-oriented Dual-Target Cross-Domain Recommendations (EKTDCR) to improve the prediction performance of these two domains simultaneously in our paper. Specifically, the latent factor and denoising autoencoder techniques are first utilized to extract and learn entity interaction embedding and entity-side embedding respectively, in separate domains, which are the basis of the proposed EKTDCR approach for dual-domain knowledge transfer and model training. Then, based on the previous entity embeddings, a dual-attention mechanism consisting of intra-domain attention and inter-domain attention modules is designed to transfer user preference knowledge to achieve entity feature fusion across domains. In addition, a deep belief network model is adopted to better train the proposed EKTDCR method, and the learnable regularization constraints are further considered to achieve dual-target cross-domain recommendations. Extensive experiments on four real-world datasets demonstrate that our EKTDCR approach can improve the recommendation performance of both domains simultaneously and significantly outperform the state-of-the-art single-domain and cross-domain baseline methods.

A General Cooperative Optimization Approach for Distributing Service Points in Mobility Applications

This article presents a cooperative optimization approach (COA) for distributing service points for mobility applications, which generalizes and refines a previously proposed method. COA is an iterative framework for optimizing service point locations, combining an optimization component with user interaction on a large scale and a machine learning component that learns user needs and provides the objective function for the optimization. The previously proposed COA was designed for mobility applications in which single service points are sufficient for satisfying individual user demand. This framework is generalized here for applications in which the satisfaction of demand relies on the existence of two or more suitably located service stations, such as in the case of bike/car sharing systems. A new matrix factorization model is used as surrogate objective function for the optimization, allowing us to learn and exploit similar preferences among users w.r.t. service point locations. Based on this surrogate objective function, a mixed integer linear program is solved to generate an optimized solution to the problem w.r.t. the currently known user information. User interaction, refinement of the matrix factorization, and optimization are iterated. An experimental evaluation analyzes the performance of COA with special consideration of the number of user interactions required to find near optimal solutions. The algorithm is tested on artificial instances, as well as instances derived from real-world taxi data from Manhattan. Results show that the approach can effectively solve instances with hundreds of potential service point locations and thousands of users, while keeping the user interactions reasonably low. A bound on the number of user interactions required to obtain full knowledge of user preferences is derived, and results show that with 50% of performed user interactions the solutions generated by COA feature optimality gaps of only 1.45% on average.

Research on Intelligent Recommendation Model Based on Knowledge Map

The integration of knowledge mapping tech and recommendation tech helps to better mine the knowledge of user preferences, better coordinate the similarity of recommended products, and enhance the user experience. Based on this, this paper first analyzes the concept and connotation of knowledge map and recommendation system, then studies the understanding tech based on knowledge map, and finally gives the utilization of intelligent recommendation data based on knowledge picture.

Intelligent information recommendation algorithm under background of big data land cultivation

In order to solve the problem of serious information overload in the era of big data and improve the informatization of intelligent recommendation result, an intelligent information recommendation algorithm based on user preference mining was put forward. According to the background of big data, user behavior data is unified. The advantage of spark relative to compared Hadoop Map Reduce was analyzed through the operating architecture and upper ecosystem of spark, so that the data processing ability was improved. The user preference mining technology was integrated with the intelligent information recommendation algorithm. Moreover, the explicit user preference knowledge and implicit user preference knowledge were analyzed to obtain the user preference knowledge and the nearest neighbor community, and thus to complete the intelligent information recommendation. Experimental results show that the proposed algorithm can exactly reflect the user preferences in different groups, with good recommendation accuracy. In addition, the desired effect is achieved.

Towards Emotion-aware Recommender Systems: an Affective Coherence Model based on Emotion-driven Behaviors

Decision making is the cognitive process of identifying and choosing alternatives based on preferences, beliefs, and degree of importance given by the decision maker to objects or actions. For instance, choosing which movie to watch is a simple, small-sized decision-making process. Recommender systems help people to make this kind of choices, usually by computing a short list of suggestions that reduces the space of possible options. These systems are strongly based on the knowledge of user preferences but, in order to fully support people, they should be grounded on a holistic view of the user behavior, that includes also how emotions, mood, and personality traits influence her choosing patterns.In this work, we investigate how to include emotional aspects in the recommendation process. We suggest that the affective state of the user, defined by a set of emotions (e.g., joy, surprise), constitutes part of choosing situation that should be taken into account when modeling user preferences.The main contribution of the paper is a general emotion-aware computational model based on affective user profiles in which each preference, such as a 5-star rating on a movie, is associated with the affective state felt by the user at the time when that preference was collected.The model estimates whether an unseen item is suitable for the current affective state of the user, by computing an affective coherence score that takes into account both the affective user profile and not-affective item features. The approach has been implemented into an Emotion-aware Music Recommender System, whose effectiveness has been assessed by performing in-vitro experiments on two benchmark datasets. The main outcome is that our system showed improved accuracy of recommendations compared to baselines which include no affective information in the recommendation model.

Social Group Recommendation With TrAdaBoost

In recent years, group recommendation has become a research hotspot and focus in online social network community. Currently, several deep-learning-based approaches are leveraged to learn preferences of groups for items and predict the next items in which groups may be interested. Yet, their recommendation performance is still unsatisfactory due to the sparse group-item interactions. In order to address this problem, in this article, we introduce an effective model, namely Social Group Recommendation model with TrAdaBoost (SGRTAB), to raise the performance of group recommendation in online social networks. The SGRTAB model includes two stages: data preprocessing (DP) and model optimization (MO). In DP, SGRTAB produces inputs for MO and implements three related tasks: extracting individual features, handling group data via GloVe, and utilizing user contribute ratings to their own groups, whereas in MO, SGRTAB implements group preference learning with the assistance of user preference learning based on the TrAdaBoost algorithm. Specifically, SGRTAB can effectively absorb the knowledge of user preferences into the process of group preference learning through the idea of transferring-ensemble learning. Moreover, extensive experiments on four real-world data sets indicate that the proposed SGRTAB model significantly outperforms the state-of-the-art baselines for social group recommendation.

Mobile Pervasive Augmented Reality Systems—MPARS

After briefly introducing aspects concerning Mobile Augmented Reality Systems, this article delves into the evolution of these systems as pervasive technology. The work debates also on acceptance of this technology in the context of outdoor applications. The need to develop context-aware, close-to-real-time feedback mechanisms that take into consideration a continuous measurement of Quality of Experience is also discussed. For this purpose, the work goes over how to integrate user preferences into context-aware feedback systems, proposing a theoretical model for measuring Quality of Experience. The model is derived from an analysis of previous technology adoption models and incorporates the knowledge of user preferences. This knowledge has been gathered via a public questionnaire.

Kernel method based human model for enhancing interactive evolutionary optimization.

A fitness landscape presents the relationship between individual and its reproductive success in evolutionary computation (EC). However, discrete and approximate landscape in an original search space may not support enough and accurate information for EC search, especially in interactive EC (IEC). The fitness landscape of human subjective evaluation in IEC is very difficult and impossible to model, even with a hypothesis of what its definition might be. In this paper, we propose a method to establish a human model in projected high dimensional search space by kernel classification for enhancing IEC search. Because bivalent logic is a simplest perceptual paradigm, the human model is established by considering this paradigm principle. In feature space, we design a linear classifier as a human model to obtain user preference knowledge, which cannot be supported linearly in original discrete search space. The human model is established by this method for predicting potential perceptual knowledge of human. With the human model, we design an evolution control method to enhance IEC search. From experimental evaluation results with a pseudo-IEC user, our proposed model and method can enhance IEC search significantly.

A Deeper Understanding of Sequence in Narrative Visualization

Conveying a narrative with visualizations often requires choosing an order in which to present visualizations. While evidence exists that narrative sequencing in traditional stories can affect comprehension and memory, little is known about how sequencing choices affect narrative visualization. We consider the forms and reactions to sequencing in narrative visualization presentations to provide a deeper understanding with a focus on linear, 'slideshow-style' presentations. We conduct a qualitative analysis of 42 professional narrative visualizations to gain empirical knowledge on the forms that structure and sequence take. Based on the results of this study we propose a graph-driven approach for automatically identifying effective sequences in a set of visualizations to be presented linearly. Our approach identifies possible transitions in a visualization set and prioritizes local (visualization-to-visualization) transitions based on an objective function that minimizes the cost of transitions from the audience perspective. We conduct two studies to validate this function. We also expand the approach with additional knowledge of user preferences for different types of local transitions and the effects of global sequencing strategies on memory, preference, and comprehension. Our results include a relative ranking of types of visualization transitions by the audience perspective and support for memory and subjective rating benefits of visualization sequences that use parallelism as a structural device. We discuss how these insights can guide the design of narrative visualization and systems that support optimization of visualization sequence.

Ranking and Clustering Web Services Using Multicriteria Dominance Relationships

As the web is increasingly used not only to find answers to specific information needs but also to carry out various tasks, enhancing the capabilities of current web search engines with effective and efficient techniques for web service retrieval and selection becomes an important issue. Existing service matchmakers typically determine the relevance between a web service advertisement and a service request by computing an overall score that aggregates individual matching scores among the various parameters in their descriptions. Two main drawbacks characterize such approaches. First, there is no single matching criterion that is optimal for determining the similarity between parameters. Instead, there are numerous approaches ranging from Information Retrieval similarity measures up to semantic logic-based inference rules. Second, the reduction of individual scores to an overall similarity leads to significant information loss. Determining appropriate weights for these intermediate scores requires knowledge of user preferences, which is often not possible or easy to acquire. Instead, using a typical aggregation function, such as the average or the minimum of the degrees of match across the service parameters, introduces undesired bias, which often reduces the accuracy of the retrieval process. Consequently, several services, e.g., those having a single unmatched parameter, may be excluded from the result set, while being potentially good candidates. In this work, we present two complementary approaches that overcome the aforementioned deficiencies. First, we propose a methodology for ranking the relevant services for a given request, introducing objective measures based on dominance relationships defined among the services. Second, we investigate methods for clustering the relevant services in a way that reveals and reflects the different trade-offs between the matched parameters. We demonstrate the effectiveness and the efficiency of our proposed techniques and algorithms through extensive experimental evaluation on both real requests and relevance sets, as well as on synthetic scenarios.

A personal agent for Chinese financial news on the Web

AbstractAs the Web has become a major channel of information dissemination, many newspapers expand their services by providing electronic versions of news information on the Web. However, most investors find it difficult to search for the financial information of interest from the huge Web information space–information overloading problem. In this article, we present a personal agent that utilizes user profiles and user relevance feedback to search for the Chinese Web financial news articles on behalf of users. A Chinese indexing component is developed to index the continuously fetched Chinese financial news articles. User profiles capture the basic knowledge of user preferences based on the sources of news articles, the regions of the news reported, categories of industries related, the listed companies, and user‐specified keywords. User feedback captures the semantics of the user rated news articles. The search engine ranks the top 20 news articles that users are most interested in and report to the user daily or on demand. Experiments are conducted to measure the performance of the agents based on the inputs from user profiles and user feedback. It shows that simply using the user profiles does not increase the precision of the retrieval. However, user relevance feedback helps to increase the performance of the retrieval as the user interact with the system until it reaches the optimal performance. Combining both user profiles and user relevance feedback produces the best performance.

Experience with rule induction and k-nearest neighbor methods for interface agents that learn

Interface agents are being developed to assist users with a variety of tasks. To perform effectively, such agents need knowledge of user preferences. An agent architecture has been developed which observes a user performing tasks, and identifies features which can be used as training data by a learning algorithm. Using the learned profile, an agent can give advice to the user on dealing with new situations. The architecture has been applied to two different information filtering domains: classifying incoming mail messages (Magi) and identifying interesting USENet news articles (UNA). This paper describes the architecture and examines the results of experimentation with different learning algorithms and different feature extraction strategies within these domains.