Data Preprocessing for Evaluation of Recommendation Models in E-Commerce

Abstract

E-commerce businesses employ recommender models to assist in identifying a personalized set of products for each visitor. To accurately assess the recommendations' influence on customer clicks and buys, three target areas-customer behavior, data collection, user-interface-will be explored for possible sources of erroneous data. Varied customer behavior misrepresents the recommendations' true influence on a customer due to the presence of B2B interactions and outlier customers. Non-parametric statistical procedures for outlier removal are delineated and other strategies are investigated to account for the effect of a large percentage of new customers or high bounce rates. Subsequently, in data collection we identify probable misleading interactions in the raw data, propose a robust method of tracking unique visitors, and accurately attributing the buy influence for combo products. Lastly, user-interface issues discuss the possible problems caused due to the recommendation widget's positioning on the e-commerce website and the stringent conditions that should be imposed when utilizing data from the product listing page. This collective methodology results in an exact and valid estimation of the customer's interactions influenced by the recommendation model in the context of standard industry metrics, such as Click-through rates, Buy-through rates, and Conversion revenue.

One Size Does Not Fit All: Multivariant User Interface Personalization in E-Commerce

“…by sampling and using noise filtering techniques [25]. In addition, solving this problem requires the right approach to preprocessing and preparing data for further analysis [26]. Given the above considerations, in the case of e-commerce systems, personalization will apply to groups (segments) of customers, and for systems used by a smaller number of users, individual or group personalization may be considered.…”

Exploration on Scientific Research Data-Targeted Intelligent Recommendation System Using Machine Learning Under the Background of Sustainable Development

“…Accordingly, some scholars propose to mix various recommendation technologies to make up for the shortcomings of a single algorithm to improve the information recommendation quality. In particular, some attempt to combine CF technology with other technologies for user recommendation services ( Chaudhary and Roy Chowdhury, 2019 ). Given the scholars’ user Psychology in pursuit of highly relevant SR information, the present work will design the SRD-targeted IRS through the optimization algorithm combined with the user’s psychological model to pursue personalized recommendation.…”

Beauty matters: reducing bounce rate by aesthetics of experience product portal page

“…To attract and convert customers, sellers invest efforts to improve customer click-through, such as installing ways for customers to subscribe to promotional services on online platforms. However, sellers often find it difficult to improve sales in an equivalent base once IMDS 121,8 click-through has been enhanced due to the high bounce rate which is the percentage of users who enter and immediately leave the site (Chaudhary and Roy Chowdhury, 2019). One industrial report shows that bounce rate is a major issue in online marketplaces with an average of 58.18% of users entering and immediately leaving retailers' sites (Baker, 2017).…”