Browse Prior Art Database

“FlagX”, Revised Keyword Classification Method of Commodity Categorization for Emptoris - Spend Enrichment Manager Tool

IP.com Disclosure Number: IPCOM000249200D
Publication Date: 2017-Feb-09
Document File: 5 page(s) / 326K

Publishing Venue

The IP.com Prior Art Database

Abstract

There is an existing method claimed for Auto classification for Spend Enrichment Manager. We have found a modification to this method as FlagX in which we have modified the sequence of keyword pattern and input data to get higher classification accuracy. The new method calculates word count and segregates input data into multiple buckets and then gets used with new keyword sequence for accurate classification.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

1

“FlagX”, Revised Keyword Classification Method of Commodity Categorization for Emptoris - Spend Enrichment Manager Tool

BACKGROUND:

Organizations worldwide use Spend analysis tools to leverage buying power, reduction in cost, sourcing,  supplier management and to develop an informed procurement strategy. For execution of Spend  analysis, the data needs to be extracted out of different ERP systems and analyzed to gain visibility. This  data comes from multiple source systems. Spend analysis includes identification, collection; cleansing,  grouping, categorization and analysis of all spend data for the goods and services purchased for the  organization.

Emptoris (SBU of IBM) has developed Spend Enrichment Manager (SEM) tool for the automated  classification of this procurement data. The most important and tedious task is to classify those data  accurately and assigning the right commodities.

Emptoris has developed keyword and data pattern based automation tool called ‘Spend Enrichment  Manager (SEM) tool’ . Whenever you run data across this tool, It auto classifies up to 40% of your data  based on input data quality. This tool assigns categories to the input data which then needs to be  manually verified for its accuracy.

Emptoris team had patented  keyword classification method for commodity categorization in Emptoris  Spend Enrichment manager. This method helped to categorize millions of data in automated way using  keyword patterns. This method had developed keyword patterns (KAN‐>KA‐>KN‐>K) for automation run  in sequential manner. It auto classifies the input data and assigns the commodity. But, this method does  not ensure the accuracy of assigned commodity.

Generally the client data contains millions of transactions which need to be classified. Manual review of  those assigned commodities for its accuracy is critical, time consuming and tedious task to complete.

Hence, we have revised the keyword classification method which auto classifies the input data with  100% accuracy.

Drawback of Existing Keyword Classification Method

The manual verification of assigned categories can be easily done for smaller set of data. But, In case of  larger datasets containing millions of records, The Spend Enrich...