Analyzing email threads

• US 10,298,531 B2
• Filed: 03/09/2016
• Issued: 05/21/2019
• Est. Priority Date: 03/27/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bai, Song, Chi, Ming Qun, Huang, Hui, Liu, Hui, Shi, Xiang Xing, Yi, Ang
• Primary Examiner(s): Nguyen, Nhat Huy T
• Application Number:
  US15/064,707
• Publication Number:
  US 20160285810A1
• Time in Patent Office: 1,168 Days
• Field of Search: None
• CPC Class Codes: H04L 51/216 Handling conversation history, e.g. grouping of messages in sessions or threads
• CPC Class Codes: H04L 51/42 Mailbox-related aspects, e.g. synchronisation of mailboxes

Abstract

An emotion function chain may be generated from an email thread and displayed to a user in one or more windows on a computer display of a computer system. An email content analyzer may extract emotion indicators from the body of an email within the email thread. Using the emotion indicators, an attitude factor for an email may be determined. The attitude factors determined from each email within the email thread may be pictorially depicted in the emotion function chain. In response to a user interaction with a pictorial depiction of an attitude factor within the emotion function chain, a second window may generated to display the body of the email used to calculate the attitude factor interacted with.

Claims

1. A computer implemented method for displaying an emotion function chain from an email thread, comprising:

identifying, by a processor, a first signature line in a first email and a second signature line in a second email, wherein the first signature line includes a first signature and the second signature line includes a second signature;

extracting the first signature from the first email and the second signature from the second email;

determining an identity of a first user using the first signature and an identity of a second user using the second signature;

accessing a signature repository to retrieve user data using the determined identities of the first and second users, wherein the user data includes sender class data of the first user and sender class data of the second user;

determining that the sender class data of the first user and the second user includes a same sender class;

grouping at least the first email and the second email together based on the first and second emails being sent by users having the same sender class;

transmitting, in response to grouping the first email and the second email together, an email thread to an email content analyzer on an email application hosted by a computer system;

receiving, at the email content analyzer, the email thread comprising at least the first email and the second email, wherein receiving the email thread includes receiving the first email and the second email based on the sender class;

extracting, in response to receiving the email thread, a first set of emotion indicators from a body of the first email and a second set of emotion indicators from a body of the second email, wherein the first set of emotion indicators from the first email include one or more words from the body of the first email and the second set of emotion indicators from the second email include one or more words from the body of the second email, the one or more words from the body of the first email and the one or more words from the body of the second email being one or more positive words, comparative words, and superlative words;

determining, based upon the first set and the second set of emotion indicators, a first attitude factor for the first email and a second attitude factor for the second email;

displaying in a first window on a computer display of the computer system, based upon the first attitude factor and the second attitude factor, an emotion function chain including a pictorial depiction of the attitude factor for the first email and a pictorial depiction of the attitude factor for the second email, wherein the respective pictorial depictions of the attitude factors for the first and second emails are based upon the sender class, and wherein a marker connects the respective pictorial depictions of the attitude factors for the first and second emails, the marker indicating that the second email is a reply to the first email; and

displaying, in a second window on the computer display, the body of the first email in response to a user interaction with the pictorial depiction of the attitude factor for the first email.

2. The method of claim 1, wherein displaying the body of the first email includes visually differentiating the first set of emotion indicators identified in the first email from the body of the first email.

3. The method of claim 1, wherein a first sender of the first email and a second sender of the second email are grouped into a category, wherein the category is based upon a common attribute of the first sender and the second sender.

4. The method of claim 1, wherein the first email and the second email are nonconsecutive in the email thread.

5. The method of claim 1, wherein determining the attitude factor includes:

receiving weight values for each emotion indicator using an emotion indicator dictionary;

calculating, using the emotion indicators and the weight values, a probability of an emotion value of the first email and a probability of an emotion value of the second email; and

determining, based upon the probability of the emotion value of the first email and the second email, the attitude factor for the first email and the attitude factor for the second email.

6. The method of claim 5, wherein the emotion indicator dictionary is a collection of emotion indicators with preassigned weight values.