Contextual Help Chat in Multi-Tenant Database Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional online chat systems in cloud computing environments often fail to provide effective assistance due to users' inability to accurately describe their issues, lack of contextual information, and inadequate questioning by AI bots or customer support agents, leading to insufficient problem-solving.
Innovation Solution
A multi-tenant database system that determines a user's context based on their profile and tenant information, providing contextually relevant assistance through user-selectable options, recommendations for knowledge articles or product features, and customizable privacy settings to control data exposure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional online chat systems are used, then basic communication is maintained, but problem-solving effectiveness deteriorates due to lack of contextual information
Solution Approach 1:
The system performs preliminary action by automatically determining user context before the chat session begins. It retrieves user profile information, tenant data, and recent activity records in advance, so that when the user initiates a chat, the contextual information is already prepared and immediately available to the support agent or AI bot, eliminating the need for users to manually provide basic contextual details.
Solution Approach 2:
The system introduces an intermediary mechanism - a contextual information retrieval and processing layer - that sits between the user and the support system. This intermediary automatically gathers relevant information from multiple sources (user profiles, tenant data, activity logs) and presents it to the support agent or AI bot, serving as a bridge that supplies necessary context without requiring direct user input.
2Reliability
If AI bots ask questions to gather information, then problem-solving improves, but interaction time increases due to inadequate questioning
Solution Approach 1:
The system performs preliminary information gathering by automatically determining user context before the chat session starts. It retrieves user profiles, tenant information, and recent activity records in advance, so that when the AI bot begins questioning, it already has a foundation of relevant information, allowing it to ask more targeted and efficient questions rather than starting from scratch.
Solution Approach 2:
The system implements feedback by continuously monitoring the chat interaction and using the retrieved contextual information to dynamically adjust the AI bot's questioning strategy. The bot receives feedback about the user's responses and the pre-retrieved context, allowing it to refine its questions in real-time to gather additional necessary information more efficiently.
3Loss of information
If users manually describe their issues, then problem-specific information is obtained, but accuracy deteriorates due to users' inability to accurately describe issues
Solution Approach 1:
The system enables self-service by automatically determining and providing contextual information about the user without requiring manual input. The system retrieves user profile data, tenant information, and activity records autonomously, presenting this pre-gathered context to the user or support agent, thereby eliminating the burden on users to manually describe their own situation while maintaining high accuracy.
Solution Approach 2:
The system introduces an intermediary information retrieval mechanism that automatically gathers contextual data about the user from multiple sources. This intermediary layer supplies accurate problem-related information without requiring users to manually describe their issues, acting as a bridge that provides precise contextual data while maintaining ease of operation.
4Reliability
If more contextual information is gathered, then recommendations become more relevant, but system complexity increases
Solution Approach 1:
The system applies segmentation by dividing the context determination process into separate, manageable modules: user profile retrieval, tenant information gathering, and recent activity recording. Each module handles a specific aspect of context collection independently, making the overall complex task of gathering comprehensive user context more manageable and maintainable through modular architecture.
Solution Approach 2:
The system implements universality by creating a multi-functional context determination mechanism that serves multiple purposes: it gathers information for support chat, generates recommendations, and provides context for analytics. The same underlying infrastructure for retrieving user profiles, tenant data, and activity records serves all these functions, reducing overall system complexity through shared components.
Data Source
AI summary
Disclosed are some implementations of systems, apparatus, methods and computer program products for providing contextually relevant recommendations based on a context of the user. The context of the user may be determined according to a set of privacy settings of the user, where the set of privacy settings indicates contextual features for which values are permitted to be accessed by a recommendation system. The contextual features may include user-related features and/or tenant features pertaining to a tenant of a multi-tenant database.