Predicting personality from patterns of behavior collected with smartphones
Predicting personality from patterns of behavior collected with smartphones
复制标题
DOI：
10.1073/pnas.1920484117
复制
发表时间：
2020-07-28
影响因子：
11.1
通讯作者：
Buehner, Markus
中科院分区：
文献类型：
--
作者：
Stachl, Clemens;Au, Quay;Buehner, Markus
智能手机在全球享有很高的采用率。这些富含传感器的设备几乎不超过一臂距离，可以轻松地改变用途，收集丰富而广泛的用户行为记录（例如位置、通信、媒体消费），对个人隐私构成严重威胁。这里研究了根据通过传感器收集的六种不同类别的行为信息和从智能手机收集的日志数据可以在多大程度上预测个人的“大五人格”维度。采用机器学习方法，我们根据连续 30 天从 624 名志愿者收集的行为数据（25,347,089 个记录事件）预测宽域（r（中值）= 0.37）和窄行方面水平（r（中值）= 0.40）的性格。我们的交叉验证结果表明，1) 沟通和社交行为、2) 音乐消费、3) 应用程序使用、4) 移动性、5) 总体电话活动和 6) 白天和夜间活动等领域的特定行为模式可以明显预测大五人格特征。这些预测的准确性与基于社交媒体平台的数字足迹的预测相似，并证明了从智能手机被动收集的行为模式中获取有关个人私人特征的信息的可能性。总的来说，我们的结果指出了从智能手机获得的行为数据的广泛收集和建模所带来的好处（例如，在研究环境中）和危险（例如，隐私影响、心理目标）。
Smartphones enjoy high adoption rates around the globe. Rarely more than an arm's length away, these sensor-rich devices can easily be repurposed to collect rich and extensive records their users' behaviors (e.g., location, communication, media con-sumption), posing serious threats to individual privacy. Here examine the extent to which individuals' Big Five personality dimensions can be predicted on the basis of six different classes of behavioral information collected via sensor and log data har-vested from smartphones. Taking a machine-learning approach, we predict personality at broad domain (r(median) = 0.37) and nar-row facet levels (r(median) = 0.40) based on behavioral data collected from 624 volunteers over 30 consecutive days (25,347,089 logging events). Our cross-validated results reveal that specific patterns behaviors in the domains of 1) communication and social behav-ior, 2) music consumption, 3) app usage, 4) mobility, 5) overall phone activity, and 6) day-and night-time activity are distinc-tively predictive of the Big Five personality traits. The accuracy of these predictions is similar to that found for predictions based on digital footprints from social media platforms and demon-strates the possibility of obtaining information about individuals' private traits from behavioral patterns passively collected from their smartphones. Overall, our results point to both the bene-fits (e.g., in research settings) and dangers (e.g., privacy impli-cations, psychological targeting) presented by the widespread collection and modeling of behavioral data obtained from smartphones.