Data Grouping Method for Unbounded Datasets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Client devices with limited storage resources face challenges when accessing large data sets, either by sacrificing interactivity or consuming excessive storage and network resources, especially when performing binning processes.
Innovation Solution
A computer system and method that groups data elements into bins without accessing the data set, allowing each data element to be associated with a group based on value ranges, with the option to merge groups and represent data as floating-point integers, minimizing storage and network usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the large data set is stored by server storage resources, then the client device can access the data set, but the client device sacrifices interactivity with the large data set for obtaining binning results
Solution Approach 1:
The patent segments the data set into multiple data groups based on value ranges, allowing the client device to work with smaller, manageable groups rather than the entire large data set. This enables interactivity while reducing the quantity of data that needs to be transmitted and stored on the client device.
Solution Approach 2:
The server performs preliminary grouping of the data set into data groups with defined minimum and maximum values before transmission to the client device. This preliminary action allows the client device to receive and process only the necessary grouped data, maintaining interactivity without handling the complete large data set.
2Ease of operation
If the server transmits the large data set to the client device, then the client device can store and process the data set, but the transmission consumes a large amount of network resources and the client device consumes a large amount of storage resources
Solution Approach 1:
The patent extracts only the essential grouping information (minimum and maximum values defining data groups) from the large data set and transmits this extracted information to the client device. This extraction approach enables the client device to perform binning operations without receiving or storing the complete large data set, significantly reducing network and storage resource consumption.
Solution Approach 2:
Instead of transmitting the actual large data set, the server transmits a simplified representation or copy consisting of data group definitions (minimum and maximum values). This copy contains sufficient information for the client device to perform binning operations without requiring the original large data set, reducing resource consumption while maintaining processing capability.
3Adaptability or versatility
If the data set is grouped into multiple data groups with unknown bounds, then the system can handle unbounded data sets, but the grouping process becomes more complex
Solution Approach 1:
The patent implements a dynamic grouping process where data groups are created and merged based on incoming data values. The system can dynamically adjust the number of data groups, their minimum and maximum values, and merge groups when appropriate, allowing handling of unbounded data sets without requiring predetermined group structures or bounds.
Solution Approach 2:
The system changes parameters such as the number of data groups, their minimum and maximum values, and merging thresholds based on the characteristics of the incoming data. This parameter adaptation allows the grouping process to handle unbounded data sets effectively, managing complexity through dynamic parameter adjustment rather than fixed structures.
Data Source
AI summary
A computer system to group a data set is provided, in which the computer system includes a memory configured to store instructions and a processor disposed in communication with the memory. The processor upon execution of the instructions is configured to access a data set including a plurality of data elements and associate the plurality of data elements with a plurality of data groups. The data set is assignable to a plurality of bins that are grouped differently than the plurality of data groups, wherein assignment of the data set to the plurality of bins is performed by accessing the plurality of data groups without accessing the data set or the plurality of data elements.