Neural Net Recommendation Engine for E-Commerce
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing e-commerce recommendation systems often fail to accurately identify a user's current desires due to complexity and excessive inputs, masking the user's actual needs and becoming unwieldy, while also not effectively incorporating current behavior alongside past behavior.
Innovation Solution
A system and method utilizing a neural net-based recommendation engine that applies multiple models, such as CNN and transformer-based models, for analyzing both current and past user behavior to provide personalized recommendations, featuring user behavioral logging and concatenating outputs to determine user needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional recommendation systems include actions of other users to surface newly desired goods, then the system can help users discover new items, but the actual current desire of the user may be masked
Solution Approach 1:
The patent segments user behavior analysis into two distinct models: one for current online behavior and another for past online behavior. This segmentation allows the system to analyze current desires separately from historical patterns, preventing the masking effect while still incorporating both types of data for comprehensive recommendations.
Solution Approach 2:
The patent introduces an intermediary mechanism that processes current behavior data separately before combining it with past behavior analysis. This intermediary step ensures that current user desires are not overshadowed by historical data, maintaining measurement precision while still allowing collaborative filtering to surface new items.
2Productivity
If recommendation systems use excessively complicated models and too many inputs, then they can handle vast volumes of online purchases, but the systems become unwieldy
Solution Approach 1:
The patent divides the recommendation system into multiple specialized models (current behavior model and past behavior model) that can be processed independently and in parallel. This segmentation maintains productivity by distributing computational load while reducing the complexity of individual model components.
Solution Approach 2:
The patent applies partial action by selecting only the most relevant behavioral factors for analysis rather than processing all possible inputs. This approach maintains system productivity while avoiding the unwieldy complexity that would result from incorporating excessive inputs and overly complicated models.
Data Source
AI summary
A system and method for recommendations for online e-commerce to a user, which provide a number of different methods for surfacing a current desire or need of the user in terms of a purchase, through a recommendation engine that applies one or more neural net models. One non-limiting example of such a method is to apply a first model for analyzing the current online behavior of the user, and a second model for analyzing past online behavior of the user, then concatenating the outputs of both models to a single output, to determine a recommendation for the user. The two models may be of the same or different type. For example and without limitation, the first model may comprise a CNN, while the second model may comprise a CNN and/or a transformer-based model. The transformer-based model may comprise an encoder alone, rather than the art-known combination of an encoder and a decoder. If the model comprises a CNN, preferably the CNN features a plurality of overlapping filters of different shapes. Optionally the CNN comprises a single convolutional layer. Optionally each such model may comprise a plurality of different models. Optionally the recommendation engine may also comprise another type of AI or machine learning algorithm, in addition to one or more neural net models.