Personalized Recommendation Ranking via User-Specific Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online listing platforms face challenges in providing relevant item recommendations to users, as current methods often fail to accurately reflect user interests based on prior interactions and user profiles, leading to a suboptimal user experience and reduced average purchase size.
Innovation Solution
A computer-implemented method and system that generates item recommendations by analyzing mutual popularity scores of items co-purchased by prior users and ranking them based on user-specific parameters derived from user data, including browsing history and sociodemographic characteristics, using machine-learning algorithms to determine the likelihood of user interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If item recommendations are provided based on similarity to the given item, then the quantity of recommended items is increased, but the relevance of recommendations to the user deteriorates
Solution Approach 1:
The patent applies local quality by transitioning from uniform item similarity-based recommendations to user-specific personalized recommendations. The system analyzes user data (browsing history, demographics, purchase behavior) to determine individual preferences and tailors recommendations accordingly. This allows different quality levels of relevance for different users while maintaining a comprehensive set of recommended items.
Solution Approach 2:
The patent changes the parameter for recommendation generation from item-centric similarity metrics to user-centric preference models. By incorporating multiple user-specific parameters (browsing behavior, purchase history, demographic characteristics), the system transforms the recommendation approach to better match individual user needs while still providing a substantial quantity of recommendations.
2Reliability
If recommendations are based on prior user interactions and user profiles, then the relevance of recommendations is improved, but the complexity of the recommendation system deteriorates
Solution Approach 1:
The patent segments the recommendation system into distinct functional modules: user data collection module, user profile analysis module, item ranking module, and recommendation generation module. This segmentation allows each component to handle specific tasks independently, making the overall complex system more manageable and maintainable while preserving the ability to provide highly relevant personalized recommendations.
Solution Approach 2:
The patent introduces user profiles as an intermediary structure that mediates between raw user data and final recommendations. User profiles aggregate and process user interaction data into meaningful representations, simplifying the connection between complex user behavior patterns and the recommendation generation process. This intermediary layer reduces the direct complexity of processing all user data for each recommendation.
3Ease of operation
If user-specific ranking parameters are applied to rank recommendable items, then the user experience is improved, but the processing time deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing user profiles and item features before the actual recommendation request. User profiles are updated periodically in advance, and item metadata is pre-processed and indexed. This allows the system to quickly generate recommendations by retrieving pre-computed information rather than processing all user data in real-time, thus improving user experience while reducing processing time.
Solution Approach 2:
The patent uses partial action by selecting and processing only the most relevant user parameters and item features for each recommendation request, rather than processing all available data. The system dynamically determines which user profile attributes and item characteristics are most important for the current recommendation context, reducing processing time while maintaining high recommendation quality through targeted analysis.
Data Source
AI summary
A method and a system for recommending items in an online listing platform are provided. The method comprises: receiving, from an electronic device, a request for item recommendations triggered by the user having interacted with a given item of the online listing platform; identifying, in the plurality of items, based on the given item, a set of recommendable items; obtaining user data of the user; determining, based on the user data, for a given recommendable item, a respective value of a user-specific ranking parameter, which is indicative of a likelihood value of the user perceiving the given recommendable item as being of interest to the user; ranking the set of recommendable items according to respective values of the user-specific ranking parameter, thereby generating a ranked set of recommendable items; and selecting, from the ranked set of recommendable items, at least one recommendable for presentation thereof to the user.