Fermentation database mining by pattern recognition

Gregory Stephanopoulos, Georg Locher, Michael J. Duff, Roy Kamimura, George Stephanopoulos

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

A large volume of data is routinely collected during the course of typical fermentation and other processes. Such data provide the required basis for process documentation and occasionally are also used for process analysis and improvement. The information density of these data is often low, and automatic condensing, analysis, and interpretation ('database mining') are highly desirable. In this article we present a methodology whereby process variables are processed to create a database of derivative process quantities representative of the global patterns, intermediate trends, and local characteristics of the process. A powerful search algorithm subsequently attempts to extract the specific process variables and their particular attributes that uniquely characterize a class of process outcomes such as high- or low-yield fermentations. The basic components of our pattern recognition methodology are described along with applications to the analysis of two sets of data from industrial fermentations. Results indicate that truly discriminating variables do exist in typical fermentation data and they can be useful in identifying the causes or symptoms of different process outcomes. The methodology has been implemented in a user-friendly software, named dbminer, which facilitates the application of the methodology for efficient and speedy analysis of fermentation process data.

Original languageEnglish (US)
Pages (from-to)443-452
Number of pages10
JournalBiotechnology and Bioengineering
Volume53
Issue number5
DOIs
StatePublished - Mar 5 1997
Externally publishedYes

Fingerprint

Fermentation
Pattern recognition
Databases
Documentation
Software
Derivatives

Keywords

  • database mining
  • dbminer®
  • decision trees
  • fermentation
  • pattern recognition
  • wavelets

ASJC Scopus subject areas

  • Biotechnology
  • Microbiology

Cite this

Fermentation database mining by pattern recognition. / Stephanopoulos, Gregory; Locher, Georg; Duff, Michael J.; Kamimura, Roy; Stephanopoulos, George.

In: Biotechnology and Bioengineering, Vol. 53, No. 5, 05.03.1997, p. 443-452.

Research output: Contribution to journalArticle

Stephanopoulos, Gregory ; Locher, Georg ; Duff, Michael J. ; Kamimura, Roy ; Stephanopoulos, George. / Fermentation database mining by pattern recognition. In: Biotechnology and Bioengineering. 1997 ; Vol. 53, No. 5. pp. 443-452.
@article{216d2fe8c6b94bdfaadd40fa8f3cc8ac,
title = "Fermentation database mining by pattern recognition",
abstract = "A large volume of data is routinely collected during the course of typical fermentation and other processes. Such data provide the required basis for process documentation and occasionally are also used for process analysis and improvement. The information density of these data is often low, and automatic condensing, analysis, and interpretation ('database mining') are highly desirable. In this article we present a methodology whereby process variables are processed to create a database of derivative process quantities representative of the global patterns, intermediate trends, and local characteristics of the process. A powerful search algorithm subsequently attempts to extract the specific process variables and their particular attributes that uniquely characterize a class of process outcomes such as high- or low-yield fermentations. The basic components of our pattern recognition methodology are described along with applications to the analysis of two sets of data from industrial fermentations. Results indicate that truly discriminating variables do exist in typical fermentation data and they can be useful in identifying the causes or symptoms of different process outcomes. The methodology has been implemented in a user-friendly software, named dbminer, which facilitates the application of the methodology for efficient and speedy analysis of fermentation process data.",
keywords = "database mining, dbminer{\circledR}, decision trees, fermentation, pattern recognition, wavelets",
author = "Gregory Stephanopoulos and Georg Locher and Duff, {Michael J.} and Roy Kamimura and George Stephanopoulos",
year = "1997",
month = "3",
day = "5",
doi = "10.1002/(SICI)1097-0290(19970305)53:5<443::AID-BIT1>3.0.CO;2-H",
language = "English (US)",
volume = "53",
pages = "443--452",
journal = "Biotechnology and Bioengineering",
issn = "0006-3592",
publisher = "Wiley-VCH Verlag",
number = "5",

}

TY - JOUR

T1 - Fermentation database mining by pattern recognition

AU - Stephanopoulos, Gregory

AU - Locher, Georg

AU - Duff, Michael J.

AU - Kamimura, Roy

AU - Stephanopoulos, George

PY - 1997/3/5

Y1 - 1997/3/5

N2 - A large volume of data is routinely collected during the course of typical fermentation and other processes. Such data provide the required basis for process documentation and occasionally are also used for process analysis and improvement. The information density of these data is often low, and automatic condensing, analysis, and interpretation ('database mining') are highly desirable. In this article we present a methodology whereby process variables are processed to create a database of derivative process quantities representative of the global patterns, intermediate trends, and local characteristics of the process. A powerful search algorithm subsequently attempts to extract the specific process variables and their particular attributes that uniquely characterize a class of process outcomes such as high- or low-yield fermentations. The basic components of our pattern recognition methodology are described along with applications to the analysis of two sets of data from industrial fermentations. Results indicate that truly discriminating variables do exist in typical fermentation data and they can be useful in identifying the causes or symptoms of different process outcomes. The methodology has been implemented in a user-friendly software, named dbminer, which facilitates the application of the methodology for efficient and speedy analysis of fermentation process data.

AB - A large volume of data is routinely collected during the course of typical fermentation and other processes. Such data provide the required basis for process documentation and occasionally are also used for process analysis and improvement. The information density of these data is often low, and automatic condensing, analysis, and interpretation ('database mining') are highly desirable. In this article we present a methodology whereby process variables are processed to create a database of derivative process quantities representative of the global patterns, intermediate trends, and local characteristics of the process. A powerful search algorithm subsequently attempts to extract the specific process variables and their particular attributes that uniquely characterize a class of process outcomes such as high- or low-yield fermentations. The basic components of our pattern recognition methodology are described along with applications to the analysis of two sets of data from industrial fermentations. Results indicate that truly discriminating variables do exist in typical fermentation data and they can be useful in identifying the causes or symptoms of different process outcomes. The methodology has been implemented in a user-friendly software, named dbminer, which facilitates the application of the methodology for efficient and speedy analysis of fermentation process data.

KW - database mining

KW - dbminer®

KW - decision trees

KW - fermentation

KW - pattern recognition

KW - wavelets

UR - http://www.scopus.com/inward/record.url?scp=0031553815&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031553815&partnerID=8YFLogxK

U2 - 10.1002/(SICI)1097-0290(19970305)53:5<443::AID-BIT1>3.0.CO;2-H

DO - 10.1002/(SICI)1097-0290(19970305)53:5<443::AID-BIT1>3.0.CO;2-H

M3 - Article

C2 - 18634039

AN - SCOPUS:0031553815

VL - 53

SP - 443

EP - 452

JO - Biotechnology and Bioengineering

JF - Biotechnology and Bioengineering

SN - 0006-3592

IS - 5

ER -