Information explosion creates dilemma in finding preferred products from the digital marketplaces. Thus, it is challenging for online companies to develop an efficient recommender system for large portfolio of products. The aim of this research is to develop an integrated recommender system model for online companies, with the ability of providing personalized services to their customers. The K-nearest neighbors (KNN) algorithm uses similarity matrices for performing the recommendation system; however, multiple drawbacks associated with the conventional KNN algorithm have been identified. Thus, an algorithm considering weight metric is used to select only significant nearest neighbors (SNN). Using secondary dataset on MovieLens and combining four types of prediction models, the study develops an integrated recommender system model to identify SNN and predict accurate personalized recommendations at lower computation cost. A timestamp used in the integrated model improves the performance of the personalized recommender system. The research contributes to behavioral analytics and recommender system literature by providing an integrated decision-making model for improved accuracy and aggregate diversity. The proposed prediction model helps to improve the profitability of online companies by selling diverse and preferred portfolio of products to their customers.
Recommender systems solve the problem of information overload, by helping to find the most suitable items from a large set. Evaluating recommender system and made recommendations are equally important in an efficient recommender system.
PurposeAn uncertain product demand in online retailing leads to loss of opportunity cost and customer dissatisfaction due to instances of product unavailability. On the other hand, when e-retailers store excessive inventory of durable goods to fulfill uncertain demand, it results in significant inventory holding and obsolescence cost. In view of such overstocking/understocking situations, this study attempts to mitigate online demand risk by exploring novel e-retailing approaches considering the trade-offs between opportunity cost/customer dissatisfaction and inventory holding/obsolescence cost.Design/methodology/approachFour e-retailing approaches are introduced to mitigate uncertain demand and minimize the economic losses to e-retailer. Using three months of purchased history data of online consumers for durable goods, four proposed approaches are tested by developing product attribute based algorithm to calculate the economic loss to the e-retailer.FindingsMixed e-retailing method of selling unavailable products from collaborative e-retail partner and alternative product's suggestion from own e-retailing method is found to be best for mitigating uncertain demand as well as limiting customer dissatisfaction.Research limitations/implicationsLimited numbers of risk factor have been considered in this study. In the future, others risk factors like fraudulent order of high demand products, long delivery time window risk, damage and return risk of popular products can be incorporated and handled to reduce the economic loss.Practical implicationsThe analysis can minimize the economic losses to an e-retailer and also can maximize the profit of collaborative e-retailing partner.Originality/valueThe study proposes a retailer to retailer collaboration approach without sharing the forecasted products' demand information.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.