Data-driven Machine Learning and Neural Network Algorithms in the Retailing Environment: Consumer Engagement, Experience, and Purchase Behaviors.

• Author: Kliestik, Tomas

1. Introduction

   The purpose of our systematic review is to examine the recently published literature on data-driven machine learning and neural network algorithms in the retailing environment and integrate the insights it configures on consumer engagement, experience, and purchase behaviors. By analyzing the most recent (2022) and significant (Web of Science, Scopus, and ProQuest) sources, our paper has attempted to prove that customer brand perception and satisfaction can be carried out according to machine learning algorithms and big data. The actuality and novelty of this study are articulated by addressing algorithmic shopping decision journeys, that is an emerging topic involving much interest. Our research problem is whether artificial intelligence payment tools may increase willingness to buy quickly or over-purchase, optimizing consumption of goods and services in terms of utility, shaping online consumption pattern behaviors and consumer purchase decisions.

   In this review, prior findings have been cumulated indicating that the impact of artificial intelligence-enabled checkouts develops on users' convenience perception. The identified gaps advance user interaction, purchase intention, and consumer behavior across retail business operations. Our main objective is to indicate that user decision-making algorithms throughout the online environment can be pivotal in artificial intelligence technologies to more thoroughly grasp the consumer journey. This systematic review contributes to the literature on predictive analytics, automated decisionmaking processes, and machine and deep learning techniques (Andronie et al, 2021a, b; Birtus and Lazaroiu, 2021; Lazaroiu, 2013; Musova et al, 2021) in retail e-commerce business by clarifying that collaborative datadriven tools can assist in developing employee-customer interaction. This research endeavors to elucidate whether artificial intelligence service is prompt, accommodating, unbiased, considerate, and stress relieving. Our contribution is by integrating research findings indicating that artificial intelligence assistant upsides constitute determinants influencing perceived utilitarian/hedonic value, thus shaping user willingness.

2. Theoretical Overview of the Main Concepts

   By use of data gathering and analysis, data-driven retail innovations (Alimamy and Gnoth, 2022; Huang and Rust, 2022; Liao et al, 2022; Rausch et al, 2022; Zhao et al, 2022) can articulate content production, shopping, and interaction. Chatbots are pivotal in digital customer service in terms of purchase intentions through machine learning technologies. Knowledgeable power over artificial intelligence assistants can decrease consumer risk perceptions. Perceived confidence and risk are related to increased customers' intentions to create together across both web and augmented reality-based shopping. Digital technology and artificial intelligence tools can optimize user engagement, sales performance, shared value creation, distinct purchasing decisions, and customized shopping experiences. Online purchasing intentions, choices, behaviors, patterns, and habits can be configured by machine learning and neural networks. The manuscript is organized as following: theoretical overview (section 2), methodology (section 3), collaborative data-driven tools in web and augmented reality-based shopping (section 4), machine learning and neural networks in retail e-commerce business (section 5), data-driven retail innovations and machine learning technologies in digital customer service (section 6), discussion (section 7), synopsis of the main research outcomes (section 8), conclusions (section 9), limitations, implications, and further directions of research (section 10).

3. Methodology

   Throughout January 2022, a quantitative literature review of the Web of Science, Scopus, and ProQuest databases was carried out, with search terms comprising "retail" + "data-driven machine learning," "neural network algorithm," "consumer engagement," "consumer experience," and "purchase behavior." The search terms were determined as being the most employed words or phrases across the analyzed literature. As research published in 2022 was analyzed, only 148 articles met the suitability criteria. By removing questionable or indeterminate findings (insubstantial/inconsequent data), results unconfirmed by replication, too imprecise content, or having quite similar titles, 22, chiefly empirical, sources were selected (Tables 1 and 2). Extracting and inspecting publicly accessible files (scholarly sources) as evidence, before the research began no institutional ethics approval was required. (Figures 1 and 2)

4. Collaborative Data-driven Tools in Web and Augmented Reality-based Shopping

   Customer interactions with virtual and augmented reality (Alimamy and Gnoth, 2022; Ameen et al., 2022; Rausch et al., 2022) can result in shopping mall loyalty. The accurate records of previous usage behaviors encompassed by log documents and subsequent clickstream information can be inspected by retailers to gain relevant insights. Perceived confidence and risk are related to increased customers' intentions to create together across both web and augmented reality-based shopping.

   Artificial intelligence customers and service providers (Huang and Rust, 2022; Liao et al., 2022; Zhao et al., 2022) can augment or take over human ones as regards purchase decisions. While human customer service is more adjustable and sympathetic than artificial intelligence service, users may be gratified with the latter as its feedback is prompt, accommodating, unbiased, considerate, and stress relieving. Customized recommendation systems can harness consumers' interests and online purchasing intentions, choices, behaviors, patterns, and habits to suggest tailored information and merchandise by use of data gathering and analysis, articulating content production, shopping, and interaction.

   ...