Online Product Recommendation Research Articles

Context congruity effects of online product recommendations: an eye-tracking study

PurposeThe purpose of this paper is to understand how the context congruity effects of online product recommendations (PRs) by recommendation agents (RAs) influence consumers’ attention to and memory of recommended products in an online shopping environment.Design/methodology/approachThe study focuses on the context congruity effects of online PRs by examining consumers’ browsing patterns and attention characteristics (fixation counts and fixation duration) using an eye-tracking device and by measuring memory performance with an aided memory test. Three types of PRs (highly congruent, lowly congruent and incongruent PRs) and two degrees of involvement (high and low involvement) are considered.FindingsThe results of the gaze data show that context congruity effects can influence consumers’ PR attention, but this effect is not moderated by involvements. The results of the memory data show that PR recognition is influenced not only by context congruity effects but also by involvement. Another significant finding is that attention to a PR does not necessarily guarantee better memory performance.Practical implicationsThe study significantly contributes to deepening the understanding of how context congruity can influence consumers’ attention to and memory of PRs. The findings also have important managerial and practical implications, such that selecting and presenting PRs should be based on context congruity effects.Originality/valueFirst, introducing context congruity effects to investigate the effectiveness of online PRs by RAs not only provides an important theoretical contribution to research on recommendation effectiveness but also enriches its application. Second, the findings suggest that the relationship between visual attention and memory is not definitely positive. Third, to interpret the complex translation process from attention to memory, the authors propose a methodology that considers stimulus attributes, issue involvement, cognitive capacity and cognitive interference.

A Propound Hybrid Approach for Personalized Online Product Recommendations

ABSTRACT The main aim of e-commerce websites is to turn their visitors into customers. For this purpose, recommender system is used as a tool that helps in turning clicks into purchases. Obtaining explicit ratings often faces problems such as authenticity of the ratings given by customers and queries that leads to low accuracy of the recommendations. Implicit ratings play a vital role in providing refined ranking of products. Preference level of the customers are predicted based on collaborative filtering (CF) approach using implicit details and mining click stream paths of like-minded users. Extracting the similarity among products using sequential patterns improves the accuracy of ranking. Integrating these two approaches improves the recommendation quality. Based on the results of experiment carried out to compare the performance of CF, sequential path of products viewed and integration of the two, we conclude that integration of mentioned approaches is superior to the existing ones.

Stagewise learning for noisy k-ary preferences

The aggregation of k-ary preferences is a novel ranking problem that plays an important role in several aspects of daily life, such as ordinal peer grading, online image-rating, meta-search and online product recommendation. Meanwhile, crowdsourcing is increasingly emerging as a way to provide a plethora of k-ary preferences for these types of ranking problems, due to the convenience of the platforms and the lower costs. However, preferences from crowd workers are often noisy, which inevitably degenerates the reliability of conventional aggregation models. In addition, traditional inferences usually lead to massive computational costs, which limits the scalability of aggregation models. To address both of these challenges, we propose a reliable CrowdsOUrced Plackett–LucE (COUPLE) model combined with an efficient Bayesian learning technique. To ensure reliability, we introduce an uncertainty vector for each crowd worker in COUPLE, which recovers the ground truth of the noisy preferences with a certain probability. Furthermore, we propose an Online Generalized Bayesian Moment Matching (OnlineGBMM) algorithm, which ensures that COUPLE is scalable to large-scale datasets. Comprehensive experiments on four large-scale synthetic datasets and three real-world datasets show that, COUPLE with OnlineGBMM achieves substantial improvements in reliability and noisy worker detection over other well-known approaches.

Scalable online product recommendation engine based on implicit feature extraction domain

Scalable online product recommendation engine based on implicit feature extraction domain

Trust aware recommender system with distrust in different views of trusted users

A recommender system aims to provide users with personalized online product or service recommendations to handle the online information overload problem that keep rapidly increasing. The main problems in the CF recommender system are sparsity and cold start. In order to resolve the problems, one of the current researches has been directed to the CF with a trust aware mechanism that includes trust as additional information in order to predict the rating for sparse data. This paper provides a review of the existing recommender system implementing the CF and trust aware. Furthermore, based on an empirical experiment, the performances of two recommender system approaches with trust aware and distrust in different views of trusted users are also reported in this paper. The results have shown that the different views have an effect on the accuracy and rating coverage of the two algorithms.

Inducing Exploration in Service Platforms

Crowd-sourced content in the form of online product reviews or recommendations is an integral feature of most Internet-based service platforms and marketplaces, including Yelp, TripAdvisor, Netflix, and Amazon. Customers may find such information useful when deciding between potential alternatives; at the same time, the process of generating such content is mainly driven by the customers' decisions themselves. In other words, the service platform or marketplace the set of available options through its customers' decisions, while they exploit the information they obtain from the platform about past experiences to determine whether and what to purchase. Unlike the extensive work on the trade-off between exploration and exploitation in the context of multi-armed bandits, the canonical framework we discuss in this chapter involves a principal that explores a set of options through the actions of self-interested agents. In this framework, the incentives of the principal and the agents towards exploration are misaligned, but the former can potentially incentivize the actions of the latter by appropriately designing a payment scheme or an information provision policy.

Reputation based online product recommendations

Reputation based online product recommendations

Robust Plackett–Luce model for k-ary crowdsourced preferences

The aggregation of k-ary preferences is an emerging ranking problem, which plays an important role in several aspects of our daily life, such as ordinal peer grading and online product recommendation. At the same time, crowdsourcing has become a trendy way to provide a plethora of k-ary preferences for this ranking problem, due to convenient platforms and low costs. However, k-ary preferences from crowdsourced workers are often noisy, which inevitably degenerates the performance of traditional aggregation models. To address this challenge, in this paper, we present a RObust PlAckett–Luce (ROPAL) model. Specifically, to ensure the robustness, ROPAL integrates the Plackett–Luce model with a denoising vector. Based on the Kendall-tau distance, this vector corrects k-ary crowdsourced preferences with a certain probability. In addition, we propose an online Bayesian inference to make ROPAL scalable to large-scale preferences. We conduct comprehensive experiments on simulated and real-world datasets. Empirical results on “massive synthetic” and “real-world” datasets show that ROPAL with online Bayesian inference achieves substantial improvements in robustness and noisy worker detection over current approaches.

A cross-domain recommender system with consistent information transfer

Recommender systems provide users with personalized online product and service recommendations and are a ubiquitous part of today's online entertainment smorgasbord. However, many suffer from cold-start problems due to a lack of sufficient preference data, and this is hindering their development. Cross-domain recommender systems have been proposed as one possible solution. These systems transfer knowledge from one domain that has adequate preference information to another domain that does not. The outlook for cross-domain recommendation is promising, but existing methods cannot ensure the knowledge extracted from the source domain is consistent with the target domain, which may impact the accuracy of the recommendations. To address this challenging issue, we propose a cross-domain recommender system with consistent information transfer (CIT). Knowledge consistency is based on user and item latent groups, and domain adaptation techniques are used to map and adjust these groups in both domains to maintain consistency during the transfer learning process. Experiments were conducted on five real-world datasets in three categories: movies, books, and music. The results for nine cross-domain recommendation tasks show that CIT outperforms five benchmarks and increases the accuracy of recommendations in the target domain, especially with sparse data. Practically, our proposed method is applied into a telecom product recommender system and a business partner recommender system (Smart BizSeeker) to enhance personalized decision making for both businesses and individual customers.

The role of online product recommendations on customer decision making and loyalty in social shopping communities

Online product recommendation (OPR) provided by product recommender systems as well as consumers is a crucial service in social shopping communities (SSCs) for improving consumers’ shopping experience and fostering long-term relationships. However, little is understood about how these two sources of recommendations influence consumers' decision-making and their implications for customer loyalty. This study develops a model to examine how the positive (enablers) and negative (inhibitors) factors of OPR’s quality influence consumer decision process, and how the decision process ultimately influences customer loyalty. The results indicate that consumer product screening cost and decision-making quality significantly influence customer loyalty. Consumer product screening cost is negatively associated with self reference and positively associated with deceptiveness and information overload. Product evaluation cost is positively affected by self reference, deceptiveness and information overload and negatively affected by vividness. And consumer decision-making quality is positively associated with self reference and negatively associated with deceptiveness and information overload. Consumer product screening cost and product evaluation cost are comparatively more negatively influenced by inhibitors than enablers. Moreover, UGC level plays a negative moderating role on the relationships between product screening cost and customer loyalty, and between product evaluation cost and customer loyalty.

Decision Support System for Online Product Recommendation Service based on Consumer Behavior

Decision Support System for Online Product Recommendation Service based on Consumer Behavior

Semi-supervised multi-graph classification using optimal feature selection and extreme learning machine

Abstract A multi-graph is represented by a bag of graphs. Semi-supervised multi-graph classification is a partly supervised learning problem, which has a wide range of applications, such as bio-pharmaceutical activity tests, scientific publication categorization and online product recommendation. However, to the best of our knowledge, few research works have be reported. In this paper, we propose a semi-supervised multi-graph classification algorithm to handle the semi-supervised multi-graph classification problem. Our algorithm consists of three main steps, including the optimal subgraph feature selection, the subgraph feature representation of multi-graph and the semi-supervised classifier building. We first propose an evaluation criterion of the optimal subgraph features, which not only considers unlabeled multi-graphs but also considers the constraints between the multi-graph level and the graph level. Then, the optimal subgraph feature selection problem is equivalently converted into the problem of mining m most informative subgraph features. Based on those derived m subgraph features, every multi-graph is represented by an m -dimensional vector, where the i th dimension equals to 1 if at least one graph involved in the multi-graph contains the i th subgraph feature. At last, based on these vectors, semi-supervised extreme learning machine(semi-supervised ELM) is adopted to build the prediction model for predicting the labels of unseen multi-graphs. Extensive experiments on real-world and synthetic graph datasets show that the proposed algorithm is effective and efficient.

Personalized and Location aware Online Product Recommendation System

Personalized and Location aware Online Product Recommendation System

Parallel multi-graph classification using extreme learning machine and MapReduce

A multi-graph is represented by a bag of graphs and modeled as a generalization of a multi-instance. Multi-graph classification is a supervised learning problem, which has a wide range of applications, such as scientific publication categorization, bio-pharmaceutical activity tests and online product recommendation. However, existing algorithms are limited to process small datasets due to high computation complexity of multi-graph classification. Specially, the precision is not high enough for a large dataset. In this paper, we propose a scalable and high-precision parallel algorithm to handle the multi-graph classification problem on massive datasets using MapReduce and extreme learning machine. Extensive experiments on real-world and synthetic graph datasets show that the proposed algorithm is effective and efficient.

Enterprise Online Product Recommendation Service Model based on Big Data Environment

Enterprise Online Product Recommendation Service Model based on Big Data Environment

The Mediation Effect of Trusting Beliefs on The Relationship between Expectation-Confirmation and Satisfaction with The Usage of Online Product Recommendation

Online Product Recommendations (OPRs) are increasingly available to online customers as a value-added self-service in evaluating and choosing a product. Research has highlighted several advantages that customers can gain from using OPRs. However, the realization of these advantages depends on whether and to what extent customers embrace and fully utilise them. The relatively low OPR usage rate indicates that customers have not yet developed trust in OPRs’ performance. Past studies also have established that satisfaction is a valid measure of system performance and a consistent significant determinant of users’ continuous system usage. Therefore, this study aimed to examine the mediation effect of trusting beliefs on the relationship between expectation-confirmation and satisfaction. The proposed research model is tested using data collected via an online survey from 626 existing users of OPRs. The empirical results revealed that social psychological beliefs (perceived confirmation and trust) are significant contributors to customer satisfaction with OPRs. Additionally, trusting beliefs partially mediate the impact of perceived confirmation on customer satisfaction. Moreover, this study validates the extensions of the interpersonal trust construct to trust in OPRs and examines the nomological validity of trust in terms of competence, benevolence, and integrity. The findings provide a number of theoretical and practical implications.

Impact of Online Product Recommendations on Consumers Purchasing Decisions

Impact of Online Product Recommendations on Consumers Purchasing Decisions

Effects of rational and social appeals of online recommendation agents on cognition- and affect-based trust

Online product recommendation agents (RAs) utilize rational and social appeals to enhance their persuasive power. However, the underlying mechanisms of how these various appeals work are still not clear. In this study, a laboratory experiment was conducted to examine the differential effects of explanation facilities and avatar interfaces as instances of rational and social appeals, respectively. We investigated (1) the effects of these appeals on cognition- and affect-based trust, and (2) the distinct persuasive outcomes of cognition- and affect-based trust.The results show that explanation facilities significantly enhance users' cognition-based trust. Meanwhile, avatar interfaces have a significant positive impact on users' affect-based trust, but only for those who rate the avatar as highly professional. Besides, our findings attest to the fact that cognition-based trust mainly delivers the utilitarian value (perceived usefulness), while affect-based trust mainly contribute to the hedonic value (perceived enjoyment) of using a RA.

Designing Warning Messages for Detecting Biased Online Product Recommendations: An Empirical Investigation

The increasing adoption of product recommendation agents (PRAs) by e-commerce merchants makes it an important area of study for information systems researchers. PRAs are a type of Web personalization technology that provides individual consumers with product recommendations based on their product-related needs and preferences expressed explicitly or implicitly. Whereas extant research mainly assumes that such recommendation technologies are designed to benefit consumers and focuses on the positive impact of PRAs on consumers’ decision quality and decision effort, this study represents an early effort to examine PRAs that are designed to produce their recommendations on the basis of benefiting e-commerce merchants (rather than benefiting consumers) and to investigate how the availability and the design of warning messages (a potential detection support mechanism) can enhance consumers’ performance in detecting such biased PRAs. Drawing on signal detection theory, the literature on warning messages, and the literature on message framing, we identified two content design characteristics of warning messages—the inclusion of risk-handling advice and the framing of risk-handling advice—and investigated how they influence consumers’ detection performance. The results of an online experiment reveal that a simple warning message without accompanying advice on how to detect bias is a double-edged sword, because it increases correct detection of biased PRAs (hits) at the cost of increased incorrect detection (false alarms). By contrast, including in warning messages risk-handling advice about how to check for bias (particularly when the advice is framed to emphasize the loss from not following the advice) increases correct detection and, more importantly, also decreases incorrect detection. The patterns of findings are in line with the predictions of signal detection theory. With an enriched understanding of how the availability and the content design of warning messages can assist consumers in the context of PRA-assisted online shopping, the results of this study serve as a basis for future theoretical development and yield valuable insights that can guide practice and the design of effective warning messages.

Redesigning promotion strategy for e-commerce competitiveness through pricing and recommendation

Internet sales have transformed the retail business. Today almost all products sold in the marketplace can be found online, and online retailers constantly engage in promotional activities. In this paper, we demonstrate that online price promotion and product recommendations should be jointly considered and optimally determined. Through attractive price discounts, e-tailers can motivate customers to purchase the promoted product; and by way of online recommendation systems, e-tailers can encourage customers to buy non-discounted items. The best promotional effect can be achieved by concurrently optimizing price promotion and product recommendation, because the loss from discount can be compensated for by the gains from the regular items. To maximize the profit, we propose an analytical model to help e-tailers exploit the potential of online promotion, and to maximize the influence of product recommendation on e-tailer’s sales and profits. Through numerical experiments, we find the proposed approach outperforms alternative methods in profit maximization. Finally, we offer managerial insights and useful guidelines for online promotion activities.