Data Analysis System Feature Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The analysis of megadata in industries like cloud computing and e-commerce is inefficient, leading to a significant burden on systems due to the need to search and process large volumes of data, which complicates the identification of major factors affecting events.
Innovation Solution
A method and system that integrate multiple queries to obtain features, adjust data volume based on predetermined ranges by reducing or increasing data through feature elimination, narrowing conditions, or sampling, and utilize an analysis unit to analyze correlations between features and events using machine learning techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple queries are integrated to search various features and obtain large volume of data for analysis, then the completeness of data for identifying major factors affecting events is improved, but the system burden and processing time increase significantly
Solution Approach 1:
The patent segments the large-scale data analysis process into multiple independent feature analysis tasks. Each feature is analyzed separately with its own search conditions, allowing parallel processing and reducing the time required to analyze the complete dataset while maintaining data completeness for identifying major factors affecting events.
2Measurement precision
If the system searches various features and obtains large volume of data, then the accuracy of identifying major factors is improved, but the device complexity and computational burden increase
Solution Approach 1:
The system divides the complex analysis task into separate feature analysis modules, each handling a specific feature with defined search conditions. This segmentation reduces system complexity by creating manageable, independent analysis units while maintaining the ability to identify major factors accurately through comprehensive feature coverage.
Solution Approach 2:
The patent dynamically adjusts search conditions and data volume parameters based on the analysis requirements. By changing parameters such as search conditions for each feature and the volume of data retrieved, the system optimizes the balance between analysis accuracy and computational burden without requiring excessive system complexity.
3Measurement precision
If all searched data is processed for analysis, then the precision of data mining is improved, but the energy consumption and system burden increase
Solution Approach 1:
The patent extracts and analyzes only the relevant features and their corresponding data based on defined search conditions, rather than processing all available data. This extraction approach maintains data mining precision by focusing on meaningful features while reducing energy consumption by excluding unnecessary data processing.
Data Source
AI summary
A method and a device for analyzing data are provided. The method includes following steps. A plurality of queries for an event stored in a database are integrated to obtain a plurality of features. Each feature is limited at a searching condition. A plurality of items of searched data are obtained from the database according to respective searching condition of each feature. Whether a data volume of the searched data is higher or lower than a predetermined range is determined. If the data volume is higher than the predetermined range, the data volume of the searched data is reduced according to the features. If the data volume is lower than the predetermined range, the data volume of the searched data is increased according to the features. A correlation between the features and the event is analyzed according to the searched data.