Collaborative Filtering with Exponential Smoothing for Time-Aware Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional collaborative filtering methods fail to accurately predict user preference degrees over time, as they do not consider the factor of time in user behavior analysis, leading to inaccurate predictions when user behavior changes over time.
Innovation Solution
The method involves establishing an exponential smoothing model to fuse user preference degrees over multiple time cycles, generating a sparse matrix, and using a collaborative filtering model to calculate predictive values, effectively incorporating the time factor into user preference degree predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional collaborative filtering methods are used to predict user preference degrees, then the prediction process is simple, but the prediction accuracy deteriorates over time as user behavior changes
Solution Approach 1:
The patent applies dynamics by transforming the static collaborative filtering model into a dynamic one that adapts to changing user preferences over time. The exponential smoothing model continuously updates user preference degrees based on temporal patterns, allowing the system to adapt to evolving user behavior while maintaining prediction accuracy
Solution Approach 2:
The patent changes the parameter representation by introducing time-dependent parameters into the collaborative filtering model. User preference degrees are no longer static values but dynamic parameters that evolve over time according to the exponential smoothing model, enabling accurate predictions despite temporal changes in user behavior
2Ease of manufacture
If user preference degrees from past time periods are used directly for prediction, then the calculation is straightforward, but the prediction result cannot reflect current user behavior changes
Solution Approach 1:
The patent applies preliminary action by pre-processing user behavior data through exponential smoothing calculations before feeding it into the collaborative filtering model. This preliminary transformation of raw preference data into smoothed temporal sequences enables the model to account for time factors while maintaining computational efficiency
Solution Approach 2:
The exponential smoothing model serves as an intermediary between raw user behavior data and the collaborative filtering prediction model. This intermediate processing layer transforms historical preference data into time-aware representations, bridging the gap between simple historical data and reliable predictions without requiring complex direct modeling
Data Source
AI summary
A method of collaborative filtering in combination with time factor includes: establishing an exponential smoothing model; acquiring a time period proposed for the exponential smoothing model, the time period includes a plurality of time cycles; acquiring a plurality of user identifiers and user preference degree values of the user identifiers over a specified product during the plurality of time cycles; performing iterative calculations of the user preference degree values utilizing the exponential smoothing model, and obtaining smoothing results corresponding to the time cycles; generating a sparse matrix utilizing the user identifiers and the smoothing result corresponding to the time cycles, the sparse matrix includes a plurality of user preference degrees to be predicted; acquiring a collaborative filtering model and inputting the smoothing results corresponding to the time cycles into the collaborative filtering model; and training through the collaborative filtering model, calculating and obtaining predictive values of the plurality of user preference degrees to be predicted in the sparse matrix.