Visual cues and consumer's booking intention in P2P accommodation: Exploring the role of social and emotional signals from hosts' profile photos

While previous studies have examined social and emotional value through textual cues, the precise impact of these values conveyed by hosts' facial images, as depicted in their profile photos, on consumer booking intentions is unclear. This study aims to fill this gap by exploring how these values from hosts' facial images affect consumer's booking intentions. To this end, we proposed a deep learning framework to evaluate the social and emotional signals conveyed through hosts' facial images. Then, using data from 44,583 listings in six major US cities, we find a positive correlation between the social signal and consumer's booking intention, while the emotional signal shows a U-shaped relationship. Additionally, the impact of social and emotional signals differs across room type, with non-entire rental consumers more likely to be influenced by social signals and entire room consumers by emotional signals. Our findings emphasize the significance of non-functional values in P2P accommodation.