Optimal Path Prediction Model for User Intent Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current personalized recommendation services rely on analyzing user activity patterns to suggest content, but they often fail to provide an optimal path that aligns with user intent and may not leverage expert knowledge effectively, leading to inefficient user navigation through various services.
Innovation Solution
A method and apparatus that utilize an optimal path prediction model trained on user action trajectories to identify user intent and guide users through expert knowledge-based paths, incorporating AI models that learn from expert knowledge and user experiences to recommend personalized and efficient navigation paths across multiple services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If current personalized recommendation services analyze user activity patterns to suggest content, then content recommendations can be provided based on user behavior, but the services fail to provide an optimal path that aligns with user intent and do not leverage expert knowledge effectively
Solution Approach 1:
The patent introduces an optimal path prediction model as an intermediary between user activity analysis and content recommendation. This model incorporates expert knowledge graphs and multiple AI models (sequence model, classification model, generation model) to act as a mediator that translates user intent into optimized navigation paths, thereby improving ease of operation while maintaining personalization capability
Solution Approach 2:
The recommendation system is segmented into multiple specialized AI models: a sequence model for predicting user actions, a classification model for identifying user intent, and a generation model for creating optimal paths. Each model handles a specific aspect of the recommendation process, allowing the system to simultaneously provide personalized recommendations and optimized navigation paths without compromising either function
2Loss of information
If traditional recommendation systems suggest content based on similar users' activity patterns, then content discovery can be achieved, but users must navigate through unnecessary steps and cannot reach their destination efficiently
Solution Approach 1:
The system performs preliminary action by predicting user actions and identifying user intent before the user actually performs those actions. The optimal path prediction model anticipates user needs and prepares optimized navigation paths in advance, allowing users to skip unnecessary steps and reach their destination faster without losing understanding of their intent
Solution Approach 2:
The patent replaces the mechanical navigation process with an intelligent system that uses AI models to automatically generate optimal paths. Instead of users manually navigating through content based on recommendations, the system substitutes this mechanical process with automated path optimization that considers user intent and expert knowledge, significantly reducing navigation time while preserving intent understanding
3Measurement precision
If expert knowledge is integrated into the recommendation system, then more accurate and personalized paths can be provided, but the system complexity increases
Solution Approach 1:
The complex system is segmented into distinct functional modules: an expert knowledge graph that stores domain knowledge, a sequence model for action prediction, a classification model for intent identification, and a generation model for path creation. Each module has a specific function and can be independently trained and optimized, managing system complexity while maintaining high prediction accuracy through specialized processing in each segment
Solution Approach 2:
The patent introduces an expert knowledge graph as an intermediary layer between raw user data and the AI models. This knowledge graph serves as a mediator that structures and organizes expert knowledge in a standardized format, making it easier for the AI models to process and apply this knowledge without directly increasing the complexity of the core prediction algorithms
Data Source
AI summary
A method for providing an optimal path through a computer network includes predicting a subsequent action of a target user through an optimal path prediction model that is trained in a form of a graph representing a user action trajectory of a session unit and recommending a path of the predicted action as an optimal path. The recommendation includes identifying a user intent based on a previous action trajectory of the target user in a current session, and guiding a path corresponding to the user intent as one of the optimal paths.