Automated Item Description Standardization for Procurement Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large-scale procurement systems, determining consistent descriptions of items across different transactions is challenging due to varying descriptions used by buyers and suppliers, making it difficult to derive meaningful price trends and comparisons automatically.
Innovation Solution
A computer-implemented method using machine learning models to standardize item information by extracting and comparing attributes across transactions, enabling statistical analysis and generating recommendations for contract negotiation and system configuration changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If different buyers and suppliers use different descriptions for the same commodity in different transactions, then the system can accommodate diverse user preferences and terminology, but deriving meaning from the mass of community data for transactions becomes extremely difficult to execute on an automated basis
Solution Approach 1:
The system transforms the parameter of item description from free-text varying formats to standardized structured attributes. Machine learning models analyze transaction data to automatically extract and normalize item attributes (such as commodity type, specifications, quantity) from diverse descriptions, converting unstructured text into consistent parameterized data that enables automated comparison and trend analysis while preserving the original diverse descriptions for reference
Solution Approach 2:
The patent introduces an intermediary layer of machine learning-based attribute extraction and normalization between the diverse transaction descriptions and the analytical processing system. This intermediary automatically standardizes item information by extracting key attributes and mapping them to a common schema, enabling automated derivation of meaning from community data without requiring manual intervention or losing the flexibility to accommodate diverse user terminology
2Measurement precision
If manual methods are used to determine consistent item descriptions and perform statistical comparisons, then accuracy can be maintained, but the process cannot be executed automatically and scales poorly with large volumes of transaction data
Solution Approach 1:
The system implements self-service automation where machine learning models automatically perform attribute extraction, item matching, and statistical analysis without manual intervention. The models learn from transaction data to autonomously identify consistent item descriptions, perform price comparisons, and generate insights, enabling the system to handle large volumes of community data automatically while maintaining measurement precision through learned patterns and validation mechanisms
Solution Approach 2:
The patent replaces manual mechanical processes of data analysis with automated machine learning systems. Instead of human analysts manually reviewing transactions to derive price trends, the system uses trained models to automatically extract attributes, match items across transactions, compute statistical comparisons, and generate insights, achieving both automation and precision through computational intelligence
3Loss of information
If all transaction data is processed and stored for analysis, then comprehensive insights can be derived, but data privacy and security concerns arise from exposing sensitive transaction information
Solution Approach 1:
The system extracts only the essential analytical attributes from transaction data (such as item type, price, quantity, time) while leaving out sensitive identifying information (such as buyer/supplier names, specific contract terms). This extraction approach enables comprehensive price trend analysis and statistical comparisons using the extracted attributes while maintaining data privacy by not storing or exposing the sensitive portions of the transaction data
Data Source
AI summary
A computer-implemented method of executing a programmed spend management computer system. The computer system comprises a data pre-processor that is communicatively coupled to a plurality of the application instances and accesses historic transaction data from any of the instances and thereby has access to a large community of data across all tenants. The data pre-processor is programmed to normalize transaction descriptions and determine line spend values, unit price values, quantity values, and buyer country data for a plurality of commodities, and to store the data in item sets in digital storage. A statistical processor is coupled to the digital storage to access the item sets and executes statistical calculation on the item sets to generate pricing insight data. Pricing insights and/or prescriptions are generated automatically under stored program control and provided to a presentation processor for output to and/or rendering to an end-user device.