Automated Assistant Proactive Content Incorporation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated assistants are typically reactive and do not proactively provide unsolicited content of potential interest to users during human-to-computer dialog sessions, requiring users to initiate interactions for information.
Innovation Solution
Configuring automated assistants to incorporate unsolicited content based on user profiles, conversation data, and aggregate user behavior, using natural language processing to select and present relevant facts tangentially related to the conversation context.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If automated assistants remain reactive and only respond to user queries, then computing resources are conserved and system simplicity is maintained, but user engagement and information discovery are reduced
Solution Approach 1:
The system performs preliminary actions by proactively generating and presenting unsolicited content before users would naturally request it. The automated assistant identifies entities mentioned in conversation, retrieves related facts from databases, and presents them as supplemental information without waiting for explicit user queries, thereby enhancing engagement while managing resource usage through selective activation.
2Loss of information
If automated assistants proactively provide unsolicited content, then user engagement and information value are enhanced, but system complexity and processing requirements increase
Solution Approach 1:
The system segments the information provision process into distinct modules: entity identification from conversation, fact retrieval from specialized databases, relevance filtering based on user profile, and selective presentation. This modular segmentation manages system complexity by allowing each component to handle specific tasks independently while maintaining overall coordination.
Solution Approach 2:
The system introduces an intermediary filtering layer that sits between the fact database and the user presentation. This intermediary evaluates retrieved facts against user profiles, conversation context, and relevance criteria before presenting information, thereby managing complexity through controlled mediation rather than direct connection between all system components.
3Loss of information
If automated assistants present multiple facts about mentioned entities, then comprehensiveness of information is improved, but information overload and user attention loss may occur
Solution Approach 1:
The system applies local quality by tailoring the quantity and type of presented facts to each specific user and conversation context. Rather than presenting a uniform set of facts, the system selectively presents facts based on individual user profiles, previous interactions, and current conversation relevance, ensuring comprehensiveness where needed while avoiding overload in other cases.
Solution Approach 2:
The system employs partial action by presenting only a subset of available facts rather than all possible information. It selectively retrieves and presents the most relevant facts based on user profile matching and conversation context, providing sufficient information value without presenting excessive content that would cause user overload or disengagement.
Data Source
AI summary
Methods, apparatus, and computer readable media are described related to automated assistants that proactively incorporate, into human-to-computer dialog sessions, unsolicited content of potential interest to a user. In various implementations, based on content of an existing human-to-computer dialog session between a user and an automated assistant, an entity mentioned by the user or automated assistant may be identified. Fact(s)s related to the entity or to another entity that is related to the entity may be identified based on entity data contained in database(s). For each of the fact(s), a corresponding measure of potential interest to the user may be determined. Unsolicited natural language content may then be generated that includes one or more of the facts selected based on the corresponding measure(s) of potential interest. The automated assistant may then incorporate the unsolicited content into the existing human-to-computer dialog session or a subsequent human-to-computer dialog session.