Automated Assistant Source Rules for Lower-Cost Information Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
User interactions with automated assistants often result in prolonged dialogues and wastage of computational resources due to the automated assistant providing results from non-preferred sources, necessitating reformulation of inputs and additional searches to meet user preferences.
Innovation Solution
Implementing source parameter rules that prioritize preferred sources for user requests, allowing selective searching and rendering of results, thereby reducing unnecessary computations and interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the automated assistant searches multiple sources to provide comprehensive results, then the quality and completeness of information is improved, but the computational resources and time consumed increase
Solution Approach 1:
The system performs preliminary actions by pre-establishing user source preferences and organizing them into structured rules before actual information retrieval occurs. When a user interacts with the assistant, the pre-configured source preference rules are directly applied to filter and select sources, eliminating the need to search all available sources and thereby reducing computational resources while maintaining information quality.
Solution Approach 2:
The system applies different source selection strategies to different users based on their individual preferences. Each user has customized source preference rules that locally optimize the information retrieval process for their specific needs, rather than using a uniform approach for all users. This allows the system to concentrate computational resources on retrieving from preferred sources only.
2Loss of information
If the automated assistant provides results from all available sources, then the completeness of information is improved, but the user interaction time increases due to needing to reformulate inputs
Solution Approach 1:
The system incorporates user feedback about source preferences into its operation. By monitoring which sources users prefer and adjusting source selection accordingly, the system learns from user interactions and progressively improves its ability to provide complete information in fewer turns, reducing interaction time while maintaining information completeness.
Solution Approach 2:
User source preferences are captured and configured in advance through preliminary setup processes. This preliminary action ensures that when information retrieval occurs, the system already knows which sources to prioritize for each user, eliminating the need for users to reformulate their inputs to specify source preferences and thereby reducing interaction time.
3Loss of information
If the automated assistant renders all search results, then the completeness of information delivery is improved, but the device resources and user interface complexity increase
Solution Approach 1:
The system applies different rendering strategies based on user preferences and device capabilities. By using source preference rules to determine which sources to prioritize, the system can selectively render results from preferred sources with higher prominence while potentially summarizing or omitting results from non-preferred sources, thereby delivering complete information without overwhelming the user interface.
Solution Approach 2:
The system segments search results based on source preferences, organizing and presenting them in a hierarchical or categorized manner. Results from preferred sources are presented first or with greater emphasis, while results from other sources are presented subsequently or in a condensed format. This segmentation allows complete information delivery while maintaining manageable interface complexity.
Data Source
AI summary
Implementations relate to interactions between a user and an automated assistant during a dialog between the user and the automated assistant. Some implementations relate to processing received user request input to determine that it is of a particular type that is associated with a source parameter rule and, in response, causing one or more sources indicated as preferred by the source parameter rule and one or more additional sources not indicated by the source parameter rule to be searched based on the user request input. Further, those implementations relate to identifying search results of the search(es), and generating, in dependence on the search results, a response to the user request that includes content from search result(s) of the preferred source(s) and/or content from search result(s) of the additional source(s). Generating the response further includes including, in the response, some indication that indicates whether the source parameter rule was followed or violated in generating the response.