Intelligent Assistant for Expert Matching via Intent Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users often struggle to identify when they need help with tasks and finding suitable experts due to unawareness of available expertise and hesitation in approaching strangers, compounded by difficulty in articulating their needs and tracking the effectiveness of expert assistance.
Innovation Solution
An intelligent system that integrates with email and digital assistants to detect user intent, assemble a database of expert profiles, match user needs with relevant experts, and track the outcome of expert interactions using natural language processing and machine learning, providing user interfaces to facilitate connections and feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If users manually search for experts, then they can find suitable help, but it requires significant time and effort to identify needs and locate experts
Solution Approach 1:
The system performs preliminary actions by continuously monitoring user interactions, detecting potential help needs before users actively search, and pre-assembling expert profiles. This proactive approach eliminates the need for users to manually search and reduces time to connection.
Solution Approach 2:
The system serves itself by automatically detecting user needs through interaction analysis, retrieving relevant expert information, and presenting matched experts without requiring user initiation. This automation transforms manual expert search into an automated service.
2Loss of information
If users actively search for experts, then they can find help, but they often lack awareness of available expertise and how to approach experts
Solution Approach 1:
The system acts as an intermediary between users and experts, translating user interactions into detected needs, matching them with relevant expert profiles, and presenting them through user interfaces. This mediator function bridges the information gap and simplifies the interaction process.
Solution Approach 2:
The system provides feedback by monitoring user interactions, detecting help needs, and presenting matched experts. This continuous feedback loop keeps users informed of available expertise and guides them through the expert connection process without requiring manual search.
3Ease of operation
If users approach experts directly, then they can seek help, but hesitation and awkwardness reduce effectiveness
Solution Approach 1:
The system serves as an intermediary that eliminates the awkwardness of direct approaches by automatically matching users with experts based on detected needs. This removes the social friction and hesitation associated with cold calls, making expert engagement more natural and effective.
4Measurement precision
If users manually track expert assistance outcomes, then they can evaluate effectiveness, but it requires ongoing manual monitoring and feedback
Solution Approach 1:
The system performs self-service by automatically monitoring expert interactions, detecting task completion, and evaluating outcomes without requiring manual tracking. This automation provides precise measurement of expert assistance effectiveness while eliminating the time investment needed for manual monitoring.
Data Source
AI summary
Representative embodiments disclose intelligent help systems that monitor user interactions through email, digital assistants, and other applications and recognize when a user can utilize the help of an expert with a task. The system detects user intent and a category of problem from the interactions (i.e., email communications, etc.) and searches a database of user profiles to find experts with the proper expertise to help the user with the category of problem. User intent can be detected by parsing communications, extracting features from the communications, and using the extracted features to identify intent, such as through matching or machine learning. A social score and an expertise score are calculated for expert profiles from the database. The social score is based on a degree of separation and expert and the expertise score is based on a level of expertise. Experts and areas of commonality are presented to the user.