Bullet-Screen Comment Recall Pool for High-Concurrency Personalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bullet-screen comment systems face challenges in providing personalized and high-quality recommendations in high-concurrency scenarios due to limited pool sizes, uneven distribution, and reliance on time-based elimination, leading to suboptimal user interaction experiences.
Innovation Solution
A new bullet-screen comment processing system utilizing a KV database for a recall pool, combined with a TiDB database, implements personalized recommendation by storing and managing bullet-screen comment materials and indexes separately, optimizing storage and calculation methods, and ensuring data consistency through Redis segment locks, allowing for dynamic segment sizing and model-based scoring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a traditional bullet-screen comment system is used, then the system structure is simple, but the recall pool capacity is limited and cannot handle high-concurrency scenarios
Solution Approach 1:
The patent divides the recall pool into two independent pools: a material pool for storing bullet-screen comment data and an index pool for storing model evaluation results. This segmentation allows each pool to be optimized independently, enabling the system to handle high-concurrency scenarios while maintaining manageable complexity through specialized storage structures
2Productivity
If model evaluation is performed for all bullet-screen comments, then comprehensive evaluation is achieved, but the elimination process becomes inefficient and redundant reasoning occurs
Solution Approach 1:
The patent implements pre-ranking evaluation where the model assesses bullet-screen comments before they are added to the recall pool. Evaluation results are stored in the index pool, allowing the system to perform efficient filtering and elimination operations without redundant reasoning during high-concurrency operations, thus maintaining both efficiency and evaluation completeness
3Reliability
If data consistency is ensured through traditional methods, then data stability is maintained, but the system cannot handle high-concurrency and hot scenarios effectively
Solution Approach 1:
The patent introduces Redis segment locks as an intermediary mechanism to ensure data consistency between the material pool and index pool during high-concurrency operations. The locks operate at the segment level rather than the entire pool level, allowing parallel processing of different segments while maintaining consistency within each segment, thus enabling high-concurrency handling without sacrificing data reliability
4Adaptability or versatility
If fixed segment sizes are used, then bandwidth usage is predictable, but the system cannot optimize for different application scenarios
Solution Approach 1:
The patent implements dynamic segment size adjustment where the segment size is not fixed but adapts based on different application scenarios and traffic patterns. This allows the system to optimize bandwidth usage by adjusting segment sizes dynamically - using smaller segments during high-traffic periods and larger segments during low-traffic periods, thereby achieving both adaptability and energy efficiency
Data Source
AI summary
A bullet-screen comment processing method is provided. The method includes: evaluating, by using a model, bullet-screen comment information obtained from a bullet-screen comment database, and storing the bullet-screen comment information in a bullet-screen comment recall pool; and obtaining corresponding bullet-screen comment information from the bullet-screen comment recall pool based on a video identifier of a video viewed by a user of a client and a time period in which the user views the video, performing screening based on a feature algorithm, and displaying bullet-screen comment information obtained through screening on the client.