Effects of Image Contents on Tourists' Review Helpfulness : An Analytical Approach with Big Data

다국어 초록 (Multilingual Abstract)

Online reviews have become integral in decision-making, particularly for experiential goods in tourism and hospitality, addressing uncertainties linked to intangible products. Existing studies underscore the vital role of User-Generated Content (UGC) in streamlining information search and managing risks across various tourism phenomena. In recent years, there has been a notable surge in interest in User-Generated Photo (UGP), marking a paradigm shift from a text-centered to an image-centered approach. Both academia and industry have been actively exploring the impact of this shift on photos. Despite the widespread use of UGPs in reviews, their informational relevance still needs to be more adequately understood, primarily due to ongoing technical challenges. This study delves into the influence of image characteristics on review helpfulness, employing machine learning methods and econometric analysis to address research questions: distinguishing between reviews with and without images and identifying specific image characteristics that impact review helpfulness. Recognizing the evolving paradigm towards image-centric content, the research adopts a dual coding theory framework to elucidate the implications of image analysis in tourism and hospitality. Through a thorough examination of image attributes, the study aims to provide novel insights into the factors shaping the effectiveness of online reviews, offering valuable implications for the tourism and hospitality field.

목차 (Table of Contents)

• Abstract vi
• 1. Introduction · 1
• 2. Literature Review · 6
• 2.1 Online reviews analytics in tourism and hospitality · 8
• 2.2 Online review helpfulness · 12
• 2.3 User-generated photos in online reviews · 17
• 3. Research design and hypotheses development · 19
• 3.1 Determinants on review helpfulness · 19
• 3.2 Determinants of user-generated photos on review helpfulness · 23
• 3.3 Control variables · 24
• 4. Methodology · 26
• 4.1 Data collection · 27
• 4.2 Data processing · 29
• 4.2.1 Textual data coding · 31
• 4.2.2 Image data coding · 32
• 4.3 Study 1 – Supervised machine learning · 34
• 4.3.1 Classification algorithms · 34
• 4.3.2 Result of decision tree · 37
• 4.4 Study 2 – Count model · 39
• 4.4.1 Zero-inflated negative binomial model · 39
• 4.4.2 Descriptive analysis · 40
• 4.4.3 Econometric model analysis · 43
• 5. Discussion and conclusion · 45
• 5.1 Theoretical implications · 48
• 5.2 Practical implications · 49
• 5.3 Limitations and future research · 50
• 6. References · 51