Uncovering patterns of multichannel mobile communications

Abstract

Human communication patterns can be discovered or examined by analyzing mobile phone data. However, previous studies have mainly concentrated on analyzing one type of data, such as calling or Short Messaging Service (SMS). This limited approach hinders our comprehensive understanding of human communication behaviors as a whole. In this paper, we combine large-scale data and entropy evaluation methods to examine the patterns of multichannel mobile communications from various perspectives. First, we showcase users’ preferences across various channels and reveal that there is a U-shaped interactive relationship between calling and SMS in dual-channel communication as the number of contacts increases. Second, we identify an upper limit for communication vitality in multichannel communications, with the corresponding number of contacts being 10, 100, and 20 in calling, SMS, and hybrid channel. Furthermore, the patterns of strength allocation and communication vitality for both SMS and calling by users are very comparable. Third, compared to calling, users tend to send text messages more regularly and at shorter intervals. And they usually have a shorter delay between texting and calling than from calling to texting. Finally, we propose a method to quantify the complexity of multichannel behavior sequences based on multiscale entropy and two null models. These findings can facilitate a deeper understanding of human communication behaviors, while our novel method offers a universal framework for quantifying the complexity of multichannel behaviors.