Inbound Email Feature Analysis for Automated ITSM Ticket Routing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing ITSM systems face challenges in efficiently routing and categorizing unstructured inbound service requests, such as emails, to the appropriate help desk, leading to difficulties in determining urgency and providing timely self-help options.

Innovation Solution

An inbound request handler service analyzes email content using feature extraction and categorization metrics to automatically generate issue objects in an ITSM system, identifying the appropriate portal endpoint based on predefined ticket types and sub-types, and requesting additional information if needed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional issue tracking systems are used to handle unstructured email requests, then manual processing is required, but routing efficiency and categorization accuracy deteriorate

Engineering Contradiction:

Improverequest processing efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

An NLP-based intermediary service is introduced between the email inbox and the issue tracking system. This service extracts features from unstructured email content, generates feature vectors, compares them with predetermined categorization vectors, and automatically determines the appropriate portal endpoint and ticket type, thereby resolving the contradiction by automating the routing process without significantly increasing system complexity

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The manual mechanical process of reading and categorizing emails is replaced with an automated NLP-based system that uses feature extraction, vector comparison, and machine learning algorithms to automatically route requests, dramatically improving productivity while keeping the overall system architecture relatively simple

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If manual analysis of email content is performed, then categorization accuracy can be maintained, but processing time increases

Engineering Contradiction:

Improvecategorization accuracyVSAvoidrequest processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

Predetermined categorization vectors are pre-computed and stored in the system, representing different ticket types and portal endpoints. When an email arrives, its feature vector is quickly compared against these pre-prepared vectors using efficient similarity metrics, enabling fast and accurate categorization without manual intervention

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system transforms unstructured email text into structured feature vectors by extracting key parameters such as sentiment, urgency indicators, and topic keywords. This parameter transformation enables rapid comparison with predetermined vectors while maintaining high categorization accuracy, resolving the time-accuracy tradeoff

Inventive Principle:

Principle #35Parameter changes

3Speed

If automated feature extraction is implemented, then processing speed improves, but handling of diverse unstructured formats becomes more difficult

Engineering Contradiction:

Improveemail processing speedVSAvoidhandling diverse request formats

Core Design Contradiction:

SpeedVSAdaptability or versatility

Solution Approach 1:

The NLP-based feature extraction system is designed with universal capabilities to handle multiple unstructured formats including emails, chat messages, and social media posts. It uses general-purpose language modeling and feature extraction techniques that can process various text formats simultaneously, maintaining both speed and adaptability

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The feature extraction process is dynamic and adaptive, automatically adjusting to different input formats and content types. The system can identify and extract relevant features from diverse unstructured data while maintaining consistent output formats for comparison with predetermined vectors, resolving the contradiction between speed and versatility

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12367436B2System including automated content analysis of inbound email content for creating an issue object for an issue tracking system

Publication Date: 2025.07.22 ATLASSIAN PTY LTD

AI summary

A computer-implemented method for handling inbound service requests analyzes content of the message to extract a set of features, and generating a feature vector based on the set of extracted features. A categorization metric for the feature vector is then determined, using a set of predetermined categorization vectors, wherein each predetermined categorization vector corresponds to a respective portal endpoint associated with one or more predefined ticket types and sub-types of an information technology service management system. In response to the categorization metric satisfying a comparison criteria with respect to one or more predetermined categorization vectors, a particular portal endpoint is selected. Content from the email message is then extracted. A new issue object request is then automatically generated in accordance with a particular object requirement set associated with the particular portal endpoint for submission to an issue tracking system to create a new issue object.