Automated Assistant Multi-Corpus Search for Contextual Responses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automated assistants often fail to provide relevant information for follow-up user requests as they rely solely on general-purpose document corpuses, neglecting user-specific and context-specific documents that may be more relevant to the current conversation or user context.
Innovation Solution
Implementing automated assistants that expand or limit their search to user-specific and context-specific corpuses of active documents, which include currently open or recently accessed documents, and documents identified based on the user's context, allowing for more relevant information retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If automated assistants search only general-purpose document corpuses, then device complexity is reduced, but information relevance deteriorates
Solution Approach 1:
The patent segments the search system into multiple independent corpus components: general-purpose corpuses and user-specific corpuses (including active documents and context-specific documents). Each corpus type is searched separately and results are integrated, allowing the system to maintain simplicity in individual search modules while achieving high information relevance through multi-corpus integration.
2Loss of information
If automated assistants search multiple corpuses including user-specific documents, then information relevance is improved, but device complexity increases
Solution Approach 1:
The patent implements a universal search framework that handles multiple corpus types (general-purpose, user-specific, active documents, context-specific documents) through a single integrated architecture. The system uses unified search queries that can target multiple corpus types simultaneously, and employs a common result integration mechanism that works across all corpus sources, thereby managing complexity through multi-functionality rather than separate specialized systems.
3Ease of operation
If automated assistants use conventional search techniques, then ease of operation is maintained, but response accuracy deteriorates
Solution Approach 1:
The patent applies preliminary action by proactively identifying and indexing user-specific documents (active documents, context-specific documents) before they are needed for search. The system pre-processes these documents, extracts relevant features, and organizes them in searchable formats ahead of time. This preliminary preparation allows the system to quickly retrieve accurate information using simple search operations without requiring complex real-time processing.
4Ease of operation
If automated assistants reformulate follow-up requests based on conversational context, then ease of operation is improved, but information retrieval accuracy deteriorates
Solution Approach 1:
The patent adds another dimension to information retrieval by searching not only through general-purpose corpuses but also through user-specific corpuses that include active documents and context-specific documents. This dimensional expansion allows the system to maintain simple conversational reformulation while achieving high retrieval accuracy by accessing relevant information from multiple document dimensions (general knowledge, user's current context, user's browsing history) simultaneously.
Data Source
AI summary
Techniques are described herein for automated assistants that search various alternative corpora for information. In various implementations, a method may include receiving, by an automated assistant via an input component of a first client device, a free form input, wherein the free form input includes a request for specific information; searching a general purpose corpus of online documents to obtain a first set of candidate response(s) to the request for specific information; searching a user-specific corpus of active document(s) to obtain a second set of candidate response(s) to the request for specific information; comparing the first and second sets of candidate responses; based on the comparing, selecting a given response to the request for specific information from the first or second set; and providing, by the automated assistant, output indicative of the given response.