The Effect of Ad Image's Sentiment Scores and Mobile Device Attributes on Mobile Ad Response Behavior
研究了移动广告图片的情感得分（效价和唤醒度）以及移动设备的显示尺寸和像素密度如何影响消费者点击广告的概率，发现情感得分提升增加响应概率，而设备属性提升反而降低响应概率，且设备属性会调节情感得分的效果。
Using a state-of-the-art image processing algorithm applied to mobile ad creative and a unique dataset on technology features for mobile platforms, we examine how visual sentiment of mobile ad images, mobile device attributes that aid visual communication, and their interactions influence consumers’ mobile ad responses. Our results show that an increase in mobile ads’ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">valence</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">arousal</i> levels by one-unit enhances the instantaneous probability of consumers’ mobile ad response by 2.4% and 1.6%, respectively. An increase in mobile devices’ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">display size</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">pixel density</i> by 1 in and 1%, respectively, decreases the probability of mobile ad response by 42.6% and 0.7%. Interestingly, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">display size</i> positively moderates the effect of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">arousal</i> , while <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">pixel density</i> positively moderates the effect of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">valence</i> on consumers’ mobile ad responses. This study contributes to the IS and marketing literature by examining the role of ad images’ visual sentiment scores such as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">valence</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">arousal</i> , device-specific attributes on mobile display, such as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">screen display size</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">pixel density</i> , and their interaction effects on consumers’ mobile ad response behaviors. Importantly, the application of computationally efficient image sentiment processing sheds new light on the extant mobile advertising and marketing literature both from a methodological and findings perspective.