Factors Affecting Perception of Artificial Intelligence-based Intelligent Services : Using Big Data Analytics in Mixed-method Approach

• 영문명: Factors Affecting Perception of Artificial Intelligence-based Intelligent Services : Using Big Data Analytics in Mixed-method Approach
• 발행기관: 경희대학교 학술단체협의회
• 저자명: 손권상
• 간행물 정보: 『고황논집』제71집, 3~37쪽, 전체 35쪽
• 주제분류: 복합학 > 학제간연구
• 파일형태: PDF
• 발행일자: 2022.12.30

영문 초록

This study aims to propose an accommodation model from an integrated perspective, which breaks away from the existing approaches that are centered on acceptance theory to AI-based intelligent service acceptance phenomena, which combines user recognition structures and theories. Therefore, the study first qualitatively identifies the critical factors of consumer value for AI-based intelligent service products using big data analysis techniques. Second, the explanatory power of major AI-based intelligent service intention models is quantitatively verified. Third, a new acceptance model is proposed and the factors that have a higher effect on behavioral intention are verified. To achieve the purpose of this study, review data on AI-based intelligent service products were collected from Twitter and Amazon. User-aware structures were conceptualized using BERT-based topic modeling analysis. Furthermore, among the existing acceptance models, the four most widely used acceptance models, such as TAM, TPB, UTAUT2, and VAM were selected to construct key concepts, and their explanatory power of AI-based intelligent service products acceptance was verified through quantitative analysis. Next, a new acceptance model was proposed focusing on the main influencing factors derived from the qualitative and quantitative analysis results, and then, a major influencing factor was derived to better explain the acceptance of AI-based intelligent service products through comparison with representative acceptance models.

목차

1. Introduction

2. Conceptual Backgrounds

3. Research Design

4. Results

5. Discussion

References

키워드

Artificial Intelligence Technology Acceptance Consumer Perception Mixed-method Topic Modeling Hybrid SEM-ANN