Random Number Sampling for Distributed Document Instance Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large organizations, determining the number of copies of documents across various computer systems is time-consuming, requires significant processing power and memory, and is prone to duplication sensitivity, especially in distributed systems without a central server.
Innovation Solution
The method employs order statistics by pairing random numbers with document instances on each computer system, allowing them to communicate and eliminate lower values, ultimately determining the approximate number of instances using a graph of highest values after several passes, thereby avoiding the need for full counting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional document copying methods are used, then document instances can be counted, but it takes a significant amount of time and requires substantial processing power and memory
Solution Approach 1:
The patent applies partial action by using sampling techniques to count only a representative portion of document instances rather than enumerating all copies. The system selects random samples of documents and estimates the total copy count based on the sample statistics, significantly reducing the time and computational resources required while maintaining acceptable measurement precision for large-scale document inventories
2Measurement precision
If conventional counting methods are used, then document copies can be enumerated, but it requires significant processing power and memory resources
Solution Approach 1:
The system uses partial action through statistical sampling to estimate document instance counts without processing every single document copy. By analyzing a random sample of documents and applying statistical formulas, the system achieves sufficient measurement precision while dramatically reducing processing power and memory requirements compared to complete enumeration methods
3Measurement precision
If conventional document counting methods are used, then copy numbers can be determined, but they are prone to duplication sensitivity where copies are counted more than once
Solution Approach 1:
The patent applies preliminary action by implementing deduplication mechanisms before counting occurs. The system uses document metadata, checksums, or unique identifiers to recognize and eliminate duplicate document copies from the sample set before statistical analysis is performed. This preliminary deduplication step ensures that each unique document instance is counted exactly once, eliminating duplication sensitivity while maintaining measurement precision
Data Source
AI summary
Embodiments of the present invention pertain to determining an approximate number of instances of an item for an organization. According to one embodiment, instances of items that reside on computer systems associated with the organization are determined. Instances of the same item can reside on different computers and an identification uniquely identifies an item. Random numbers are associated with identifications of the items. An approximate number of instances of the item is determined based on a highest random number associated with the item. The highest random number is the highest of the random numbers that were generated for the instances of the item.