ML Session Slot Adaptation for Recommendation Homogeneity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing recommendation models struggle to create homogeneous sessions due to varying user interaction patterns, leading to sessions with mixed items that may not accurately reflect individual user preferences.
Innovation Solution
An electronic device employs a machine learning model that uses reinforcement learning to determine customizable session slots based on individual user interaction patterns, and applies a sequential model to recommend items that align with these patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If fixed time duration sessions are used for all users, then the session structure is simple and consistent, but the session homogeneity deteriorates due to varying user interaction patterns
Solution Approach 1:
The patent applies dynamics by making session slot durations adaptable rather than fixed. The system dynamically adjusts the number and duration of session slots based on user interaction patterns, allowing the session structure to evolve from static to dynamic. This resolves the contradiction by enabling the session structure to maintain simplicity through automation while achieving homogeneity through pattern-based adaptation.
Solution Approach 2:
The patent changes the parameter of session duration from a fixed value to a variable determined by user behavior patterns. By using machine learning models to analyze interaction patterns and adjust session slot durations accordingly, the system transforms the rigid fixed-time structure into a flexible parameter-based structure that maintains both simplicity and homogeneity.
2Ease of manufacture
If fixed time duration sessions are used, then the implementation is simple, but the recommendation accuracy deteriorates due to mixed items in sessions
Solution Approach 1:
The patent applies segmentation by dividing fixed time sessions into multiple session slots, each potentially containing items of a single dominating category. This segmentation allows the system to maintain the simple fixed-time structure while improving recommendation accuracy by ensuring each slot contains homogeneous items, thus separating the simplicity of implementation from the precision of recommendations.
Solution Approach 2:
The patent introduces machine learning models as intermediaries between the fixed-time session structure and the recommendation process. These models analyze user interaction patterns and determine session slot boundaries, acting as a mediator that enables accurate recommendations without requiring complex custom session structures for each user.
3Measurement precision
If customized session slots based on user patterns are used, then the recommendation accuracy is improved, but the system complexity increases
Solution Approach 1:
The patent applies self-service by enabling the system to automatically analyze user interaction patterns and generate customized session slots without manual intervention. The machine learning models autonomously process user behavior data and determine optimal session structures, reducing the perceived complexity for users while maintaining high recommendation accuracy through automated pattern recognition.
Solution Approach 2:
The patent implements feedback mechanisms where user interaction patterns continuously inform and refine session slot configurations. The system monitors user behavior, adjusts session structures based on observed patterns, and uses this feedback to improve recommendation accuracy over time, managing complexity through iterative learning rather than static complex rules.
4Productivity
If fixed time sessions are used, then the processing is efficient, but the user preference representation deteriorates due to varying interaction speeds
Solution Approach 1:
The patent applies dynamics by making session slot durations adaptive to user interaction speeds. Instead of forcing all users into fixed-time sessions, the system dynamically adjusts slot durations based on observed interaction patterns, preserving user preference information while maintaining processing efficiency through automated pattern-based segmentation.
Solution Approach 2:
The patent changes the time parameter from fixed to variable based on user behavior. By analyzing interaction speeds and adjusting session slot durations accordingly, the system preserves user preference information that would be lost in fixed-time sessions, while maintaining processing efficiency through systematic parameter adjustment rather than individual customization.
Data Source
AI summary
An electronic device and a method for creation of homogeneous sessions using machine learning (ML) model is disclosed. The electronic device receives interaction information of a set of users for a set of items. The electronic device receives a textual description of each item. The electronic device determines an embedding vector associated with each item. The electronic device clusters the determined embedding vector. The electronic device applies a reinforcement learning model on the set of labels. The electronic device determines a set of session slots associated with the received interaction information of the set of users. The electronic device applies a sequential model on the set of session slots associated with the received interaction information of the set of users. The electronic device determines a set of recommended items for a user of the set of users. The electronic device renders the determined set of recommended items.