Generative Content Service Intent Analysis Multi-Path Pipeline

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In collaborative work environments, locating and synthesizing relevant user-generated content across multiple platforms is inefficient and inaccurate, making it difficult to provide timely and accurate responses to user queries.

Innovation Solution

A generative content service that performs intent analysis on natural language inputs to determine the user's intent, selectively searches appropriate content stores (issue tracking, knowledge base, user contact), and generates tailored responses, including pre-populating forms and providing follow-up prompts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If content is searched across multiple platforms, then the completeness of content retrieval is improved, but the time required for retrieval increases

Engineering Contradiction:

Improvecompleteness of content retrievalVSAvoidretrieval time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the search process by dividing content across multiple specialized platforms (issue tracking, knowledge base, contacts) and searching them in parallel rather than sequentially. The system performs intent analysis to determine which platforms are relevant, then simultaneously queries those platforms, reducing total retrieval time while maintaining completeness.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary intent analysis on user queries before executing searches. By analyzing the query intent first, the system can pre-determine which content stores are relevant and prepare search strategies in advance, reducing the actual search execution time while ensuring all relevant platforms are covered.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If intent analysis is performed to determine user intent, then the accuracy of content retrieval is improved, but the complexity of the system increases

Engineering Contradiction:

Improveaccuracy of content retrievalVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intent analysis component as an intermediary between the user query and the search execution. This intermediary analyzes the query to determine user intent and selects appropriate content stores, improving accuracy by ensuring the right platforms are searched. The complexity is managed by keeping the intent analysis focused on specific dimensions (information seeking, action request, contact inquiry) rather than attempting comprehensive analysis.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250217371A1Generative content service with multi-path content pipeline

Publication Date: 2025.07.03 ATLASSIAN PTY LTD

AI summary

Embodiments described herein relate to systems and methods for automatically generating content for a generative content interface of a collaboration platform. The system may perform an intent analysis on a natural language user input to the generative content interface to determine an intent confidence score with respect to each of a set of request classifiers, the set of request classifiers comprising a first request classifier associated with a request for an action, a second request classifier associated with a request for information, and a third request classifier associated with a request for a contact. Based on the intent confidence scores of the request classifiers, the system may select a content store in which to search for content to satisfy a user's query.