A Hyper Personalization Model for Determining End User Computer Devices Specifications

Abstract

Abstract This research addresses the prodigality of device specifications, where end users both individuals and institutions often purchase computer devices that do not match their needs due to a lack of technical understanding and reliance on biased, company-centric information from online agents. This often results in substantial resource wastage. The proposed solution is a hyper-personalization model, implemented as a chatbot, that prioritizes the end user rather than the product. The methodology involved developing a robust model that integrates machine learning algorithms specifically Bidirectional Encoder Representations from Transformers (BERT)-based models, hybrid recommendation engine filtering collaborative, content-based and Natural Language Processing (NLP). The study utilized a mixed-method design, gathering quantitative and qualitative data from end-users in corporate institutions within Nakuru town, Kenya. Findings indicate that consumers prioritize Features and Price but face significant structural challenges, including feeling overwhelmed by technical specifications and encountering too many options. A key insight is the substantial majority of respondents expressing a high likelihood of adopting a personalized recommendation system. The implemented hyper-personalization chatbot model was validated to effectively address these issues by aligning device specifications with complex user needs. The study concludes that this model significantly enhances user satisfaction, optimizes resource utilization, and establishes a clear, non-prejudiced path for informed device selection, offering a tangible contribution to consumer empowerment and organizational efficiency.