Cloud Classification Microservice with Modular Rule Engine
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud-based classification systems lack efficiency in processing and integrating multiple classification algorithms and data sources, leading to suboptimal performance in classifying diverse textual data across different domains and taxonomies.
Innovation Solution
A cloud system with a classification microservice that employs a modular architecture, allowing multiple classifiers and rule engines to be plugged in, enabling real-time and batch classification of text terms through a REST API, and facilitating the selection of appropriate classifiers and execution orders based on specific requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a cloud-based classification system processes multiple classification algorithms and data sources, then classification accuracy improves, but system complexity increases
Solution Approach 1:
The patent divides the classification system into separate, independent microservices - each classification algorithm and data source is implemented as an independent microservice that can be developed, deployed, and scaled separately. This segmentation reduces system complexity by isolating components while maintaining the ability to process multiple algorithms and data sources simultaneously for improved classification accuracy.
Solution Approach 2:
The patent creates a universal classification framework that can accommodate multiple different classification algorithms and data sources through a common interface and standardized data structures. This multi-functional system allows various classifiers to work together seamlessly, improving overall classification accuracy without proportionally increasing system complexity through reuse of common infrastructure.
2Adaptability or versatility
If multiple classifiers are integrated into a single system, then classification versatility improves, but processing efficiency deteriorates
Solution Approach 1:
The patent implements a dynamic classifier selection mechanism that chooses which classifiers to apply based on the specific characteristics of the input data and the classification task at hand. Rather than always running all classifiers, the system dynamically selects the most appropriate subset, maintaining classification versatility while improving processing efficiency by avoiding unnecessary computations.
Solution Approach 2:
The patent introduces an intermediary layer - a classification orchestrator microservice - that manages the coordination between multiple classifiers. This intermediary handles the complexity of integrating multiple classifiers, manages their execution order, and aggregates results, thereby preserving classification versatility while improving processing efficiency through optimized coordination and result aggregation.
3Speed
If real-time classification is implemented, then response speed improves, but system resource consumption increases
Solution Approach 1:
The patent applies partial action by implementing real-time classification only for those data points that require immediate processing, while allowing batch processing for others. The system identifies which classifications need real-time response and applies classifiers selectively to those cases, improving response speed where needed while reducing overall system resource consumption by avoiding unnecessary real-time processing for all data.
Data Source
AI summary
Example embodiments for classification are described. In an example embodiment, a request including a text term to be classified is received from a source system via a communication network at a computer system. A rule associated with the text term is accessed, in which the rule indicates at least one classifier of a plurality of classifiers installed at the computer system. Classification information for the text term is generated at the computer system using the at least one classifier indicated by the rule. The generated classification information includes a classification selected from a taxonomy by the at least one classifier. The generated classification information is transmitted via the communication network to the source system.