Automatic Conversation Creator for News Content

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users face a burden in interpreting one-way information delivery from websites with homogenous content presentation, leading to a diminished user experience due to passive information reception.

Innovation Solution

An automatic conversation generation system that uses a computing device to receive and respond to content items by identifying start comments and predicting replies based on a ranking model, enhancing user interaction and experience.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If one-way information delivery with homogenous content presentation is used, then information delivery is simple and direct, but user experience is diminished and user burden increases

Engineering Contradiction:
Improveuser experienceVSAvoidinformation presentation system
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent transforms static, homogenous content presentation into dynamic conversational interactions. The system automatically generates questions based on content and enables two-way communication between users and the system, making the information delivery process adaptive and interactive rather than fixed and passive.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements feedback mechanisms where user responses to generated questions are processed and incorporated into subsequent interactions. The system learns from user feedback to improve question generation and content delivery, creating a closed-loop system that continuously adapts to user needs and preferences.

Inventive Principle:
Principle #23Feedback

2Productivity

If passive information reception is required, then information delivery is straightforward, but user engagement and interpretation efficiency decrease

Engineering Contradiction:
Improveuser interpretation efficiencyVSAvoiduser interaction mode
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The patent applies preliminary action by pre-generating relevant questions based on content analysis before users need to interpret the information. The system proactively structures information through question-generation, guiding users through key points and facilitating more efficient comprehension without requiring users to actively structure the information themselves.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If automatic conversation generation is implemented, then user interaction becomes more intuitive and efficient, but system complexity increases

Engineering Contradiction:
Improveuser interaction efficiencyVSAvoidconversation generation system
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary layer (the automatic question generation system) that mediates between the content and the user. This intermediary automatically processes content, generates appropriate questions, and manages the interaction flow, shielding users from system complexity while enabling intuitive and efficient interaction.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS10891322B2Automatic conversation creator for news
Publication Date: 2021.01.12 MICROSOFT TECHNOLOGY LICENSING LLC
  • US10891322B2 patent drawing
  • US10891322B2 patent drawing
  • US10891322B2 patent drawing

AI summary

An example environment can automatically generate content for a conversation in response to a presented content item. An example computing device receives information associated with a content item produced by a source system, the content item being accessible to other the computing devices via a network. In response to receiving the information, the computing device identifies a start comment. In response to identifying the start comment, the computing device identifies a reply to the start comment using at least a ranking model.