Mining Interaction Behaviors for Email Reply Order Prediction

Abstract

In email networks, user behaviors affect the way emails are sent and replied. While knowing these user behaviors can help to create more intelligent email services, there has not been much research into mining these behaviors. In this paper, we investigate user engagingness and responsiveness as two interaction behaviors that give us useful insights into how users email one another. Engaging users are those who can effectively solicit responses from other users. Responsive users are those who are willing to respond to other users. By modeling such behaviors, we are able to mine them and to identify engaging or responsive users. This paper proposes four types of models to quantify engagingness and responsiveness of users. These behaviors can be used as features in the email reply order prediction task which predicts the email reply order given an email pair. Our experiments show that engagingness and responsiveness behavior features are more useful than other nonbehavior features in building a classifier for the email reply order prediction task. When combining behavior and non-behavior features, our classifier is also shown to predict the email reply order with good accuracy.

Building a Multimodal Classifier of Email Behavior: Towards a Social Network Understanding of Organizational Communication

“…In recent years, email data have emerged as a pivotal resource for understanding organizational dynamics, offering insights into information flow, stylistic intricacies, and broader implications for communication and organizational culture [14][15][16][17][18][19][20].…”

“…A significant portion of prior research has been dedicated to predicting email responses. Classical machine learning models have been the mainstay, often augmented with features capturing social interactions [14,16,21]. For instance, a study [22] proposed a model to gauge the likelihood and time frame within which a recipient might respond to an email, emphasizing the relevance of features such as the number of attachments and the length of the email body.…”

“…Extracting tasks and intents from email has also been a focal point of much NLP research [22,30,31]. Deep learning, while a powerful tool, has been primarily applied to problems such as email sequencing [14,16,32] and predicting the sender's gender [33], while more recently, the inclusion of natural language generation capabilities in email servers and software has spurred studies of email reply suggestions [34,35].…”

Effective email network visualization techniques by means of user behaviors

“…Recently, On et al conducted preliminary study of behavior models for the email reply order prediction [9] and on mobile social networks [10]. -Email interaction prediction: To predict whether emails need replies, Dredze et al present a logistic regression model with a variety of features e.g., dates and times, salutations, questions, and header fields of emails [2].…”