Parallel Neural Network Embeddings for Personalized Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Configuring multi-application software platforms to provide personalized navigational targets for users is a time-consuming process, as existing methods lack efficient algorithms to accurately recommend relevant applications and services based on user attributes and navigational target characteristics.
Innovation Solution
A machine learning model utilizing parallel neural networks is trained on user and navigational target attributes, generating embeddings to identify relationships and recommend navigational targets by comparing similarity values with historical navigation data, thereby optimizing the display of user interface elements for frequently accessed functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual configuration of navigational targets is performed for each user, then personalized navigation is achieved, but time consumption and administrative effort increase significantly
Solution Approach 1:
The system automatically generates personalized navigational target recommendations for users without requiring manual configuration. The machine learning model processes user attributes and navigational target attributes independently, then combines them to produce ranked recommendations, enabling the system to serve itself rather than requiring administrator intervention for each user's navigation setup
Solution Approach 2:
The system transforms the configuration process from manual parameter setting to automated parameter generation. By changing user attributes and navigational target attributes into embedding vectors and processing them through neural networks, the system dynamically generates optimized navigational recommendations based on learned patterns from historical data, rather than requiring static manual configuration
2Measurement precision
If existing recommendation methods are used, then some navigational targets are suggested, but accuracy and relevance of recommendations are insufficient
Solution Approach 1:
The system segments the recommendation process into independent components: user attribute processing through one neural network, navigational target attribute processing through another neural network, and combination of results. This segmentation allows each component to be optimized independently, improving overall recommendation accuracy by capturing complex relationships in user and target attributes separately before integrating them
Solution Approach 2:
The system replaces traditional rule-based or simple algorithmic recommendation methods with machine learning-based neural networks. The neural networks learn complex patterns and relationships from historical navigation data, automatically generating more accurate and relevant recommendations without requiring explicit programming of recommendation logic
Data Source
AI summary
Techniques for generating navigational target recommendations for a user are disclosed. A system propagates sets of user attributes through one neural network and sets of navigational target attributes through another neural network. The neural networks are configured to generate, as outputs, vectors mapped to a same vector space. The system trains the neural networks to identify relationships between the sets of user attributes and the sets of navigational targets. Once the neural networks have been trained, the system generates an embedding for a user by propagating the user's attributes through the trained user attribute neural network. The system also generates embeddings for different navigational targets by propagating the attributes for the different navigational targets through the navigational target neural network. The system identifies relationships between the user and the navigational targets based on the embeddings. The system generates recommendations for navigational targets for the user based on the identified relationships.