Cognitive Email Analytics for Automated Task Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current email systems are inefficient and prone to errors due to the manual parsing and categorization of emails, leading to wasted time and missed deadlines as employees struggle to identify action items and manage reminders within the high volume of incoming messages.
Innovation Solution
A cognitive-enabled message analytics system that automatically classifies emails as informational or action-oriented, generates rules based on user actions, and applies these rules to new messages, reducing manual reminders and follow-ups by using cognitive analysis and big data analytics to automate task creation and alert systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual parsing and categorization of emails is used, then employees can identify action items and manage reminders, but time is wasted and productivity decreases due to high volume of incoming messages
Solution Approach 1:
The system enables emails to automatically classify themselves and generate their own action items without human intervention. The cognitive system analyzes email content, determines classifications (informational vs. action-oriented), and creates task cards automatically, allowing the email system to serve itself rather than requiring manual processing by employees
Solution Approach 2:
The patent replaces the mechanical manual process of reading and categorizing emails with a cognitive system that uses natural language understanding and machine learning. The cognitive system processes email content semantically to identify action items, deadlines, and task priorities, substituting human cognitive effort with automated intelligent processing
2Extent of automation
If automated systems parse email messages based on specified rules, then email categorization is improved, but the system lacks the ability to understand contextual meaning and generate intelligent insights
Solution Approach 1:
The system transitions from simple rule-based parameters (sender, subject line keywords) to cognitive parameters that include semantic understanding, contextual analysis, and intent recognition. The cognitive system analyzes the meaning and context of email content to determine whether emails are informational or action-oriented, and to extract task-specific parameters like deadlines and priorities
Solution Approach 2:
The system implements feedback loops where the cognitive system continuously learns from user interactions with generated task cards. User actions on task cards provide feedback that refines the cognitive model's understanding of what constitutes actionable content, improving the system's ability to capture contextual meaning and generate relevant insights over time
Data Source
AI summary
A computer-implemented method includes: determining, by a computer device, classifications of plural messages sent amongst plural users; detecting, by the computer device, actions performed by users in response to receiving ones of the plural messages; determining, by the computer device, insights based on the determined classifications and the detected actions; automatically generating, by the computer device, at least one new rule based on at least one of the insights; and automatically applying, by the computer device, the at least one new rule to new messages sent amongst the plural users.