17964117. RECOMMENDATION METHOD AND APPARATUS BASED ON AUTOMATIC FEATURE GROUPING simplified abstract (HUAWEI TECHNOLOGIES CO., LTD.)
RECOMMENDATION METHOD AND APPARATUS BASED ON AUTOMATIC FEATURE GROUPING
Organization Name
Inventor(s)
RECOMMENDATION METHOD AND APPARATUS BASED ON AUTOMATIC FEATURE GROUPING - A simplified explanation of the abstract
This abstract first appeared for US patent application 17964117 titled 'RECOMMENDATION METHOD AND APPARATUS BASED ON AUTOMATIC FEATURE GROUPING
Simplified Explanation
The patent application describes an AI-based recommendation method that automatically groups features to improve the accuracy of recommendations. Here are the key points:
- Obtaining candidate recommended objects and their association features.
- Performing multi-order automatic feature grouping on the association features of each candidate recommended object.
- Generating a multi-order feature interaction set for each candidate recommended object.
- Calculating the interaction feature contribution value for each candidate recommended object based on the multi-order feature interaction set.
- Calculating the prediction score for each candidate recommended object based on the interaction feature contribution value.
- Identifying one or more candidate recommended objects with high prediction scores as target recommended objects.
Potential Applications
This technology can have various applications, including:
- E-commerce platforms: Improving product recommendations to enhance customer satisfaction and increase sales.
- Content streaming services: Enhancing personalized content recommendations to improve user engagement.
- Social media platforms: Providing more relevant suggestions for connections, groups, or content to enhance user experience.
- Online advertising: Optimizing ad recommendations to increase click-through rates and conversions.
Problems Solved
The technology addresses the following problems:
- Accuracy of recommendations: By automatically grouping features and considering multi-order feature interactions, the method improves the accuracy of recommendations.
- Scalability: The method can handle a large number of candidate recommended objects and association features, making it suitable for real-world applications.
- Personalization: By considering individual feature interactions, the method provides personalized recommendations tailored to each user's preferences.
Benefits
The technology offers several benefits:
- Improved recommendation accuracy: By considering multi-order feature interactions, the method provides more accurate recommendations.
- Enhanced user experience: Personalized recommendations based on individual feature interactions lead to more relevant and engaging content.
- Increased efficiency: The automatic feature grouping reduces the computational complexity, making the recommendation process more efficient.
- Scalability: The method can handle a large number of candidate recommended objects, making it suitable for platforms with a vast user base.
Original Abstract Submitted
This application relates to the field of artificial intelligence. A recommendation method based on automatic feature grouping includes: obtaining a plurality of candidate recommended objects and a plurality of association features of each of the plurality of candidate recommended objects; performing multi-order automatic feature grouping on the plurality of association features of each candidate recommended object, to obtain a multi-order feature interaction set of each candidate recommended object; obtaining an interaction feature contribution value of each candidate recommended object through calculation based on the plurality of association features in the multi-order feature interaction set of each candidate recommended object; obtaining a prediction score of each candidate recommended object through calculation based on the interaction feature contribution value of each candidate recommended object; and determining one or more corresponding candidate recommended objects with a high prediction score as a target recommended object.
