DATA MINING - Federal Efforts Cover a Wide Range of Uses [Implementing Data Mining Systems]

8/9/2019 DATA MINING - Federal Efforts Cover a Wide Range of Uses [Implementing Data Mining Systems]

1/71

a

GAOUnited States General Accounting Office

Report to the Ranking Minority Member,Subcommittee on Financial Management,the Budget, and International Security,Committee on Governmental Affairs,U.S. Senate

May 2004 DATA MINING

Federal Efforts Covera Wide Range of Uses

GAO-04-548


2/71

Federal agencies are using data mining for a variety of purposes, rangingfrom improving service or performance to analyzing and detecting terrorist

patterns and activities. Our survey of 128 federal departments and agencieson their use of data mining shows that 52 agencies are using or are planningto use data mining. These departments and agencies reported 199 datamining efforts, of which 68 are planned and 131 are operational. The figurehere shows the most common uses of data mining efforts as described byagencies. Of these uses, the Department of Defense reported the largest

number of efforts aimed at improving service or performance, managinghuman resources, and analyzing intelligence and detecting terroristactivities. The Department of Education reported the largest number ofefforts aimed at detecting fraud, waste, and abuse. The National Aeronauticand Space Administration reported the largest number of efforts aimed atanalyzing scientific and research information. For detecting criminalactivities or patterns, however, efforts are spread relatively evenly amongthe agencies that reported having such efforts.

In addition, out of all 199 data mining efforts identified, 122 used personalinformation. For these efforts, the primary purposes were improving serviceor performance; detecting fraud, waste, and abuse; analyzing scientific andresearch information; managing human resources; detecting criminal

activities or patterns; and analyzing intelligence and detecting terroristactivities.

Agencies also identified efforts to mine data from the private sector and datafrom other federal agencies, both of which could include personalinformation. Of 54 efforts to mine data from the private sector (such ascredit reports or credit card transactions), 36 involve personal information.Of 77 efforts to mine data from other federal agencies, 46 involve personalinformation (including student loan application data, bank account numberscredit card information, and taxpayer identification numbers).

Top Six Purposes of Data Mining Efforts in Departments and Agencies

Both the government and theprivate sector are increasinglyusing data miningthat is, theapplication of database technologyand techniques (such as statisticalanalysis and modeling) to uncoverhidden patterns and subtlerelationships in data and to inferrules that allow for the prediction

of future results. As has beenwidely reported, many federal datamining efforts involve the use ofpersonal information that is minedfrom databases maintained bypublic as well as private sectororganizations.

GAO was asked to survey datamining systems and activities infederal agencies. Specifically, GAOwas asked to identify planned andoperational federal data miningefforts and describe their

characteristics.

www.gao.gov/cgi-bin/getrpt?GAO-04-548

To view the full product, including the scopeand methodology, click on the link above.For more information, contact Linda Koontz at(202) 512-6240 or [email protected].

Highlights of GAO-04-548, a report to theRanking Minority Member, Subcommitteeon Financial Management, the Budget,and International Security, Committee on

Governmental Affairs, U.S. Senate

May 2004

DATA MINING

Federal Efforts Cover a Wide Range ofUses
http://www.gao.gov/cgi-bin/getrpt?GAO-04-548http://www.gao.gov/cgi-bin/getrpt?GAO-04-548http://www.gao.gov/cgi-bin/getrpt?GAO-04-548http://www.gao.gov/cgi-bin/getrpt?GAO-04-548http://www.gao.gov/cgi-bin/getrpt?GAO-04-548http://www.gao.gov/cgi-bin/getrpt?GAO-04-548


3/71

Page i GAO-04-548 Data Minin

Contents

LetterResults in BriefBackgroundAgencies Identified Numerous Data Mining Efforts with Various

AimsSummary 1

AppendixesAppendix I: Objective, Scope, and Methodology 1

Appendix II: Surveyed Departments and Agencies 1

Appendix III: Departments and Agencies Reporting No Data MiningEfforts 2

Appendix IV: Inventories of Efforts 2

Tables Table 1: Top Six Purposes of Data Mining Efforts in Departmentsand Agencies and Number of Efforts Reported

Table 2: Department of Agricultures Inventory of Data MiningEfforts 2

Table 3: Department of Commerces Inventory of Data Mining

Efforts 2Table 4: Department of Defenses Inventory of Data Mining

Efforts 2Table 5: Department of Educations Inventory of Data Mining

Efforts 3Table 6: Department of Energys Inventory of Data Mining

Efforts 4Table 7: Department of Health and Human Services Inventory of

Data Mining Efforts 4Table 8: Department of Homeland Securitys Inventory of Data

Mining Efforts 4Table 9: Department of the Interiors Inventory of Data Mining

Efforts 4Table 10: Department of Justices Inventory of Data Mining

Efforts 4Table 11: Department of Labors Inventory of Data Mining Efforts 4Table 12: Department of States Inventory of Data Mining Efforts 5Table 13: Department of Transportations Inventory of Data Mining

Efforts 5


4/71

Contents

Page ii GAO-04-548 Data Minin

Table 14: Department of the Treasurys Inventory of Data MiningEfforts 5

Table 15: Department of Veterans Affairs Inventory of Data MiningEfforts 5

Table 16: Environmental Protection Agencys Inventory of DataMining Efforts 5

Table 17: Export-Import Bank of the United States Inventory of DataMining Efforts 5

Table 18: Federal Deposit Insurance Corporations Inventory of DataMining Efforts 5

Table 19: Federal Reserve Systems Inventory of Data Mining

Efforts 5Table 20: National Aeronautics and Space Administrations

Inventory of Data Mining Efforts 5Table 21: Nuclear Regulatory Commissions Inventory of Data

Mining Efforts 6Table 22: Office of Personnel Managements Inventory of Data

Mining Efforts 6Table 23: Pension Benefit Guaranty Corporations Inventory of Data

Mining Efforts 6Table 24: Railroad Retirement Boards Inventory of Data Mining

Efforts 6Table 25: Small Business Administrations Inventory of Data Mining

Efforts 6

Figures Figure 1: Top Six Purposes of Data Mining Efforts That InvolvePersonal Information 1

Figure 2: Top Six Purposes of Data Mining Efforts That InvolvePrivate Sector Data 1

Figure 3: Top Six Purposes of Data Mining Efforts That InvolveData from Other Federal Agencies 1


5/71

Contents

Page iii GAO-04-548 Data Minin

Abbreviations

CARDS Counterintelligence Analytical Research Data SystemCG Coast GuardCI-AIMS Counterintelligence Automated Investigative

Management SystemDHHS Department of Health and Human ServicesDOD Department of DefenseDOE Department of EnergyDOT Department of Transportation

EFTPS Electronic Federal Tax Payment SystemEOS Earth Observing SystemFARS Fatality Analysis Reporting SystemFDA Food and Drug AdministrationGENESIS Global Environmental and Earth Science Information

SystemGSFC Goddard Space Federal CenterHR Human ResourcesHRSA Health Resources and Services AdministrationMATRIX Multistate Anti-terrorism Information Exchange SystemNASA National Aeronautics and Space AdministrationNVO National Virtual Observatory

OIG Office of Inspector GeneralOLAP On-line Analytical ProcessingRSST Real Estate Stress TestSAA Spectral Analysis AutomationSAS Safety Automated SystemSMARTS Statistical Management Analysis and Reporting Tool

SystemSWC Space Warfare CenterTIMS Technical Information Management SystemTOP Treasury Offset Program VA Veterans AffairsVHA Veterans Health Administration

VISN Veterans Integrated Service Network

This is a work of the U.S. government and is not subject to copyright protection in theUnited States. It may be reproduced and distributed in its entirety without furtherpermission from GAO. However, because this work may contain copyrighted images orother material, permission from the copyright holder may be necessary if you wish toreproduce this material separately.


6/71


7/71

Page 2 GAO-04-548 Data Minin

To address our objective to identify and describe operational and planneddata mining systems and activities in federal agencies, we surveyed chiefinformation officers or comparable officials at 128 federal departments andagencies to determine whether the agencies had operational and planneddata mining systems or activities.2 We then conducted telephone interviewswith the reported system managers to obtain information on thecharacteristics of the identified data mining efforts. To verify theinformation we received, we sent follow-up letters to agencies thatresponded as well as to those that did not respond, we asked responsibleofficials to verify the information, and we performed random assessmentsof the means that these officials used to verify the information.

In addition, we conducted a search of technical literature and periodicalsto develop a comprehensive list of federal government data mining effortsand then compared these efforts with data mining efforts reported byfederal agencies. If the data mining efforts on our lists were not reported onthe survey, we contacted the appropriate chief information officers and,with their concurrence, added the efforts.

We performed our work from May 2003 to April 2004 in accordance withgenerally accepted government auditing standards. Additional details onour scope and methodology are provided in appendix I.

Results in Brief Federal agencies are using data mining for a variety of purposes, rangingfrom improving service or performance to analyzing and detecting terroristpatterns and activities. Our survey of 128 federal departments and agencieson their use of data mining shows that 52 agencies are using or are planningto use data mining. These departments and agencies reported 199 datamining efforts, of which 68 were planned and 131 were operational. Themost common uses of data mining efforts were described by agencies as

improving service or performance;

detecting fraud, waste, and abuse;

analyzing scientific and research information;

2That is, we asked about both systems explicitly dedicated to data mining and activitiesusing automated tools to mine databases that are part of other systems. In this report, weuse the word efforts to refer to both systems and activities, unless otherwise specified.


8/71


managing human resources;

detecting criminal activities or patterns; and

analyzing intelligence and detecting terrorist activities.

The Department of Defense reported having the largest number of datamining efforts aimed at improving service or performance and at managinghuman resources. Defense was also the most frequent user of efforts aimedat analyzing intelligence and detecting terrorist activities, followed by theDepartments of Homeland Security, Justice, and Education.

The Department of Education reported the largest number of efforts aimedat detecting fraud, waste, and abuse, while the National Aeronautics andSpace Administration targets most of their data mining efforts (21 out of23) toward analyzing scientific and research information. Data miningefforts for detecting criminal activities or patterns, however, were spreadrelatively evenly among the reporting agencies.

In addition, out of all 199 data mining efforts identified, 122 used personalinformation. For these efforts, the primary purposes were detecting fraud,waste, and abuse; detecting criminal activities or patterns; analyzingintelligence and detecting terrorist activities; and increasing taxcompliance.

Agencies also identified efforts to mine data from the private sector anddata from other federal agencies, both of which could include personalinformation. Of 54 efforts to mine data from the private sector (such ascredit reports or credit card transactions), 36 involve personal informationOf 77 efforts to mine data from other federal agencies, 46 involve personalinformation (including student loan application data, bank accountnumbers, credit card information, and taxpayer identification numbers).

Background Data mining enables corporations and government agencies to analyzemassive volumes of data quickly and relatively inexpensively. The use ofthis type of information retrieval has been driven by the exponentialgrowth in the volumes and availability of information collected by thepublic and private sectors, as well as by advances in computing and datastorage capabilities. In response to these trends, generic data mining toolsare increasingly available foror built intomajor commercial databaseapplications. Today, mining can be performed on many types of data,


9/71


including those in structured, textual, spatial, Web, or multimedia forms.Data mining is becoming a big business; Forrester Research has estimatedthat the data mining market is passing the billion dollar mark.

Although the use and sophistication of data mining have increased in boththe government and the private sector, data mining remains an ambiguousterm. According to some experts, data mining overlaps a wide range ofanalytical activities, including data profiling, data warehousing, onlineanalytical processing, and enterprise analytical applications.3 Some of theterms used to describe data mining or similar analytical activities includefactual data analysis and predictive analytics. We surveyed technical

literature and developed a definition of data mining based on the mostcommonly used terms found in this literature. Based on this search, wedefine data mining as the application of database technology andtechniquessuch as statistical analysis and modelingto uncover hiddenpatterns and subtle relationships in data and to infer rules that allow for theprediction of future results. We used this definition in our initial survey ofchief information officers; these officials found the definition sufficient toidentify agency data mining efforts.

Data mining has been used successfully for a number of years in the privateand public sectors in a broad range of applications. In the private sector,these applications include customer relationship management, marketresearch, retail and supply chain analysis, medical analysis and diagnosticsfinancial analysis, and fraud detection. In the government, data mining wasinitially used to detect financial fraud and abuse. For example, data mininghas been an integral part of GAO audits and investigations of federalgovernment purchase and credit card programs.4 Data mining and relatedtechnologies are also emerging as key tools in Department of HomelandSecurity initiatives.

3Lou Agosta, Data Mining Is DeadLong Live Predictive Analytics! (Forrester Research,Oct. 30, 2003), http://www.forrester.com/Research/LegacyIT/0,7208,33030,00.html(downloaded Jan. 26, 2004).

4For more information on the uses of data mining in GAO audits, see U.S. GeneralAccounting Office,Data Mining: Results and Challenges for Government Programs,Audits, and Investigations, GAO-03-591T (Washington, D.C: Mar. 25, 2003).
http://www.forrester.com/Research/LegacyIT/0http://www.forrester.com/Research/LegacyIT/0,7208,33030,00.htmlhttp://www.gao.gov/cgi-bin/getrpt?GAO-03-591Thttp://www.gao.gov/cgi-bin/getrpt?GAO-03-591Thttp://www.gao.gov/cgi-bin/getrpt?GAO-03-591Thttp://www.forrester.com/Research/LegacyIT/0http://www.forrester.com/Research/LegacyIT/0,7208,33030,00.html


10/71


Data Mining Poses PrivacyChallenge

Since the terrorist attacks of September 11, 2001, data mining has beenseen increasingly as a useful tool to help detect terrorist threats byimproving the collection and analysis of public and private sector data. In arecent report on information sharing and analysis to address the challengesof homeland security, it was noted that agencies at all levels of governmentare now interested in collecting and mining large amounts of data fromcommercial sources.5 The report noted that agencies may use such data noonly for investigations of known terrorists, but also to perform large-scaledata analysis and pattern discovery in order to discern potential terroristactivity by unknown individuals. Such use of data mining by federal

agencies has raised public and congressional concerns regarding privacy.

One example of a large-scale development effort launched in the wake ofthe September 11 attacks is the Multistate Anti-terrorism InformationExchange System, known as MATRIX. MATRIX, currently used in fivestates,6 provides the capability to store, analyze, and exchange sensitiveterrorism-related and other criminal intelligence data among agencieswithin a state, among states, and between state and federal agencies.Information in MATRIX databases includes criminal history records,drivers license data, vehicle registration records, incarceration records,and digitized photographs. Public awareness of MATRIX and of similarlarge-scale data mining or data mining-like projects has led to concerns

about the governments use of data mining to conduct a massdataveillance7a surveillance of large groups of peopleto sift throughvast amounts of personally identifying data to find individuals who might fita terrorist profile.

5Creating a Trusted Information Network for Homeland Security (New York City: TheMarkle Foundation, December 2003),

http://www.markletaskforce.org/Report2_Full_Report.pdf(downloaded Mar. 8, 2004).6Five states are currently participating in the MATRIX pilot project: Connecticut, Florida,Michigan, Ohio, and Pennsylvania.

7Roger Clarke, Information Technology and Dataveillance, Communications of the ACMvol. 31, issue 5 (New York City: ACM Press, May 1988),http://www.anu.edu.au/people/Roger.Clarke/DV/CACM88.html (downloaded Mar. 5, 2004).Clarke defines mass dataveillance as the systematic use of personal data systems in theinvestigation or monitoring of the actions or communications of groups of people.
http://www.markletaskforce.org/Report2_Full_Report.pdfhttp://www.anu.edu.au/people/Roger.Clarke/DV/CACM88.htmlhttp://www.anu.edu.au/people/Roger.Clarke/DV/CACM88.htmlhttp://www.markletaskforce.org/Report2_Full_Report.pdfhttp://www.anu.edu.au/people/Roger.Clarke/DV/CACM88.htmlhttp://www.markletaskforce.org/Report2_Full_Report.pdf


11/71


12/71


Agencies IdentifiedNumerous Data MiningEfforts with VariousAims

Of 128 federal departments and agencies surveyed for information on theirplanned and operational data mining efforts (listed in app. II), 52 agenciesreported 199 data mining efforts, and 69 agencies reported that they werenot engaged in data mining and were not planning such efforts (listed inapp. III). Of the 199 data mining efforts, 68 were planned and 131 wereoperational. Seven agencies did not respond to our survey.10 Appendix IVlists the 199 data mining efforts reported, along with key characteristics.

Agencies described the most common purposes of data mining efforts as

improving service or performance;

detecting fraud, waste, and abuse;

analyzing scientific and research information;

managing human resources;

detecting criminal activities or patterns; and

analyzing intelligence and detecting terrorist activities.

As shown in table 1, the Department of Defense reported the largestnumber of efforts aimed at improving service or performance (with 19 outof 65 reported efforts) and at managing human resources (with 14 out of 17efforts). Defense was also the most frequent user of efforts aimed atanalyzing intelligence and detecting terrorist activities, with 5 of 14 effortsfollowed by the Departments of Homeland Security and Justice, with 4 and3 efforts, respectively. The Department of Education has the largestnumber of efforts aimed at detecting fraud, waste, and abuse (9 out of 24efforts reported). The National Aeronautics and Space Administrationaccounts for 21 of the 23 identified efforts for analyzing scientific andresearch information. Efforts are spread relatively evenly among theagencies that reported using data mining efforts for detecting criminal

10Agencies that did not respond to our survey are (1) the Central Intelligence Agency; (2) theCorporation for National and Community Services; (3) the Department of Army, Departmenof Defense; (4) the Equal Employment Opportunity Commission; (5) the National ParkService, Department of the Interior; (6) the National Security Agency, Department ofDefense; and (7) the Rural Utilities Service, Department of Agriculture.


13/71


activities or patterns. Table 1 summarizes the top six uses of data miningefforts among the responding agencies.

Table 1: Top Six Purposes of Data Mining Efforts in Departments and Agencies and Number of Efforts Reported

Source: GAO analysis of agency-provided data.

Department or agency

Improvingservice or

performance

Detectingfraud, waste,

and abuse

Analyzingscientific and

researchinformation

Managinghuman

resources

Detectingcriminal

activities orpatterns

Analyzinintelligenc

and detectinterroris

activitie

Department of Agriculture 8 1

Department of Commerce

Department of Defense 19 1 1 14 1

Department of Education 6 9 3

Department of Energy 3

Department of Health and HumanServices 4 1

Department of Homeland Security 5 2 2

Department of the Interior 1

Department of Justice 1 1 3

Department of Labor 3 1

Department of State 2

Department of Transportation 1

Department of the Treasury 4 1 2

Department of Veterans Affairs 5 5 1

Environmental Protection Agency 1

Export-Import Bank of the UnitedStates 1

Federal Deposit Insurance Corporation 1

Federal Reserve System 1

National Aeronautics and SpaceAdministration 1 1 21

Nuclear Regulatory Commission 1Office of Personnel Management 1

Pension Benefit Guaranty Corporation 2

Railroad Retirement Board 1

Small Business Administration 1

Total 65 24 23 17 15 1


14/71


Some data mining purposes focus on human activities and therefore areinherently likely to involve personal information; examples of thesepurposes are detecting fraud, waste, and abuse; detecting criminalactivities or patterns; managing human resources; and analyzingintelligence. The following are examples of data mining efforts for each ofthese purposes:

Detecting fraud, waste, and abuse. The Veterans BenefitsAdministrations C & P Payment Data Analysis effort mines veteranscompensation and pension data for evidence of fraud.

Detecting criminal activities or patterns. The Department ofEducations Title IV Identity Theft Initiative effort focuses on identitytheft cases involving education loans.

Managing human resources. The U.S. Air Forces Oracle HR (HumanResources) uses data mining to provide information on promotions, paygrades, clearances, and other information relevant to human resourcesplanning.

Analyzing intelligence and detecting terrorist activities. The DefenseIntelligence Agencys Verity K2 Enterprise mines data from theintelligence community and Internet sources to identify foreignterrorists or U.S. citizens connected to foreign terrorism activities.

On the other hand, other categories of efforts do not necessarily focus onhuman activities or involve personal information, such as many of theefforts aimed at analyzing scientific and research information. The NationaAeronautics and Space Administration, for example, mines large, complexearth science data sets to find patterns and relationships to detect hiddenevents (the system is called Machine Learning and Data Mining forImproved Data Understanding of High Dimensional Earth Sensed Data).

Similarly, many efforts aimed at improving service or performance (the

most frequently cited purpose of data mining efforts) do not involvepersonal information. For example, the Department of the Navys SupplyManagement System Multidimensional Cubes system includes a datawarehouse containing data on every ship part that has been ordered sincethe 1980s, with multidimensional information on each part. The Navy usesdata mining to calculate failure rates and identify needed improvements;according to the Navy, this system reduces downtime on ships byimproving parts replacement.


15/71


However, some efforts aimed at improving service or performance doinvolve personal information. For example, the Veterans AdministrationsVISN (Veterans Integrated Service Network) 16 Data Warehouse is minedfor a variety of information, including patient visits, laboratory tests, andpharmacy records, to provide management with health care systemperformance information.

Overall, 122 of the 199 data mining efforts involve personal information.Figure 1 shows the top six purposes of these efforts, as well as theirdistribution.

Figure 1: Top Six Purposes of Data Mining Efforts That Involve Personal Informatio

Of the 199 data mining efforts, 54 use or plan to use data from the privatesector. Of these, 36 involve personal information. The personal informationfrom the private sector included credit reports and credit card transactionrecords. Figure 2 shows the distribution of the top six purposes of the 54

efforts involving data from the private sector.

0 10 20 30 40

Purposes

Source: GAO analysis of agency data.

Number of data mining efforts

Managing human resources

Analyzing intelligence and detectingterrorist activities

Increasing tax compliance

Detecting criminal activities orpatterns

Improving service or performance

Detecting fraud, waste, and abuse

7

10

15

15

24

33


16/71


Figure 2: Top Six Purposes of Data Mining Efforts That Involve Private Sector Data

Of the 199 data mining efforts, 77 efforts use or plan to use data from otherfederal agencies. Of the 77 efforts, 46 involve personal information. Thepersonal information from other federal agencies included student loanapplication data, bank account numbers, credit card information, andtaxpayer identification numbers. Figure 3 shows the top six uses for the 77efforts involving data from other federal agencies and their distribution.

0 10 20 30 40

Purposes



Improving safety

Detecting criminal activities or

patterns

Analyzing intelligence and

detecting terrorist activities

Analyzing scientific and research

information



4

4

5

8

9

14


17/71


Figure 3: Top Six Purposes of Data Mining Efforts That Involve Data from Other

Federal Agencies

Summary Driven by advances in computing and data storage capabilities and bygrowth in the volumes and availability of information collected by thepublic and private sectors, data mining enables government agencies toanalyze massive volumes of data. Our survey shows that data mining isincreasingly being used by government for a variety of purposes, rangingfrom improving service or performance to analyzing and detecting terroristpatterns and activities.

Although this survey provides a broad overview of the emerging uses ofdata mining in the federal government, more work is needed to shed lighton the privacy implications of these efforts. In future work, we plan toexamine selected federal data mining efforts and their implications.

As agreed with your office, unless you publicly announce the contents ofthe report earlier, we plan no further distribution until 30 days from thereport date. At that time, we will send copies of this report to the Chairmenand Ranking Minority Members of the House Committee on GovernmentReform; Subcommittee on Civil Service and Agency Organization, HouseCommittee on Government Reform; Select Committee on HomelandSecurity, House of Representatives; Senate Committee on Governmental

0 10 20 30 40

Purposes




Analyzing scientific and researchinformation

Analyzing intelligence and detecting

terrorist activities

Detecting criminal activities or patterns

Managing human resources


5

6

7

12

13

20


18/71


Affairs; and the Subcommittee on Oversight of Government Management,the Federal Workforce and the District of Columbia, Senate Committee onGovernmental Affairs. We will also make copies available to others onrequest. In addition, this report will be available at no charge on the GAOWeb site at http://www.gao.gov.

If you have any questions concerning this report, please call me at (202)512-6240 or Mirko J. Dolak, Assistant Director, at (202) 512-6362. We canalso be reached by e-mail at [email protected] and [email protected],respectively. Key contributors to this report were Camille M. Chaires,Barbara S. Collier, Orlando O. Copeland, Nancy E. Glover, Stuart M.

Kaufman, Lori D. Martinez, Morgan F. Walts, and Marcia C. Washington.

Sincerely yours,

Linda D. KoontzDirector, Information Management Issues
http://www.gao.gov./mailto:[email protected]:[email protected]:[email protected]:[email protected]://www.gao.gov./mailto:[email protected]:[email protected]:[email protected]:[email protected]


19/71


Appendix I

Objective, Scope, and Methodology

Our objective was to identify and describe planned and operational federadata mining efforts. As a first step in addressing this objective, wedeveloped a definition of data mining. Because this expression has arange of meanings, we surveyed the technical literature to develop adefinition based on the most commonly used terms found in this literatureWe defined data mining as the application of database technology andtechniquessuch as statistical analysis and modelingto uncover hiddenpatterns and subtle relationships in data and to infer rules that allow for theprediction of future results. In our initial survey of chief informationofficers, these officials found the definition sufficient to identify agencydata mining efforts.

We then surveyed chief information officers or comparable officials at 128federal departments and agencies (see app. II) and asked them to identifywhether their agency had operational and planned data mining efforts. Weachieved a 95 percent response rate. Of the 121 agencies that responded, 69reported that they did not have any data mining efforts (see app. III). Wefollowed up with these 69 agencies and gave them another opportunity toreport data mining efforts.

To obtain information on the characteristics of the identified operational orplanned data mining efforts, we conducted structured telephoneinterviews1 with the identified system owners or activity managers. Theinterviews were designed to obtain detailed information about each datamining system, including the purpose and size, the use of personalinformation, and the use of data from the private sector or other federalorganizations. We pretested the structured interview to ensure relevanceand clarity.

We aggregated these data by agency and sent them back to the chiefinformation officer, comparable official, or their designee and asked thatthey review the characteristics for completeness and accuracy. One of the52 departments and agencies that reported data mining systemstheDepartment of Homeland Securityhas not responded to our request to

review the reported data for completeness and accuracy.

1In a structured interview, the interviewer asks the same questions of numerous individualsor individuals representing numerous organizations in a precise manner, offering eachinterviewee the same set of possible responses.


20/71

Appendix IObjective, Scope, and Methodology


We performed random assessments of the means that these officials usedto verify the information. Based on these assessments, we concluded thatthe agencies verification methods were reasonable and that as a result, wecould rely on the accuracy of the reported data. We also conducted asearch of technical literature and periodicals to develop a list of federalgovernment data mining efforts and then compared the efforts on this listwith the data mining efforts reported by federal agencies. If the data miningefforts on our list were not reported on the survey, we contacted the chiefinformation officer or comparable official to determine whether that datamining effort should be included in our survey.

Because this was not a sample survey, there are no sampling errors.However, the practical difficulties of conducting any survey may introduceerrors, commonly referred to as nonsampling errors. For example,difficulties in how a particular question is interpreted, in the sources ofinformation that are available to respondents, or in how the data areentered into a database or were analyzed can introduce unwantedvariability into the survey results. We took steps in the development of thestructured interview, the data collection, and the data analysis to minimizethese nonsampling errors. Among these steps, we pretested the structuredinterview instrument, contacted nonresponding agencies as well asagencies not identifying data mining efforts, and sent the aggregated datato the agency chief information officer for review.

We conducted our work from May 2003 to April 2004 in accordance withgenerally accepted government auditing standards.


21/71


Appendix II

Surveyed Departments and Agencies

Department of Agriculture

Agricultural Marketing Service

Agricultural Research Service

Animal and Plant Health Inspection Service

Cooperative State Research, Education, and Extension Service

Farm Service Agency

Food and Nutrition Service

Food Safety and Inspection Service

Foreign Agricultural Service

Forest Service

National Agricultural Statistics Service

Natural Resources Conservation Service

Risk Management Agency

Rural Utilities Service


Bureau of the Census

Economic Development Administration

International Trade Administration

National Oceanic and Atmospheric Administration

U.S. Patent and Trademark Office


22/71

Appendix IISurveyed Departments and Agencies


Department of Defense

Missile Defense Agency

Defense Advanced Research Projects Agency

Defense Commissary Agency

Defense Contract Audit Agency

Defense Contract Management Agency

Defense Information Systems Agency

Defense Intelligence Agency

Defense Legal Services Agency

Defense Logistics Agency

Defense Security Cooperation Agency

Defense Security Service

Defense Threat Reduction Agency

Department of the Air Force

Department of the Army

Department of the Navy

National Geospatial-Intelligence Agency

National Security Agency

U.S. Marine Corps

Department of Education


23/71



Department of Energy

Bonneville Power Administration

Southeastern Power Administration

Southwestern Power Administration

Western Area Power Administration

Department of Health and Human Services

Administration for Children and Families

Agency for Healthcare Research and Quality

Centers for Disease Control and Prevention

Centers for Medicare and Medicaid Services

Food and Drug Administration

Health Resources and Services Administration

Indian Health Service

National Institutes of Health

Program Support Center

Department of Homeland Security

Border and Transportation Security Directorate

Bureau of Citizenship and Immigration Services

Emergency Preparedness and Response Directorate

Information Analysis and Infrastructure Protection Directorate

Management Directorate


24/71



Science and Technology Directorate

U.S. Coast Guard

U.S. Secret Service

Department of Housing and Urban Development

Department of the Interior

Bureau of Indian Affairs

Bureau of Land Management

Bureau of Reclamation

Minerals Management Service

National Park Service

Office of Surface Mining Reclamation and Enforcement

U.S. Fish and Wildlife Service

U.S. Geological Survey

Department of Justice

Bureau of Alcohol, Tobacco, Firearms, and Explosives

Drug Enforcement Administration

Federal Bureau of Investigation

Federal Bureau of Prisons

U.S. Marshals Service

Department of Labor

Department of State


25/71



Department of Transportation

Federal Aviation Administration

Federal Highway Administration

Federal Motor Carrier Safety Administration

Federal Railroad Administration

Federal Transit Administration

National Highway Traffic Safety Administration

Department of the Treasury

Bureau of Engraving and Printing

Bureau of the Public Debt

Financial Management Service

Internal Revenue Service

Office of the Comptroller of the Currency

Office of Thrift Supervision

U.S. Mint

Department of Veterans Affairs

Veterans Benefits Administration

Veterans Health Administration

Agency for International Development

Central Intelligence Agency

Corporation for National and Community Service


26/71



Environmental Protection Agency

Equal Employment Opportunity Commission

Executive Office of the President

Export-Import Bank of the United States

Federal Deposit Insurance Corporation

Federal Energy Regulatory Commission

Federal Reserve System

Federal Retirement Thrift Investment Board

General Services Administration

Legal Services Corporation

National Aeronautics and Space Administration

National Credit Union Administration

National Labor Relations Board

National Science Foundation

Nuclear Regulatory Commission

Office of Management and Budget

Office of Personnel Management

Peace Corps

Pension Benefit Guaranty Corporation

Railroad Retirement Board

Securities and Exchange Commission


27/71



Small Business Administration

Smithsonian Institution

Social Security Administration

U.S. Postal Service


28/71


Appendix III

Departments and Agencies Reporting No DataMining Efforts

The following 69 departments and agencies reported that they have nooperational or planned data mining efforts:

Department of Agriculture

Agricultural Marketing Service

Agricultural Research Service

Animal and Plant Health Inspection Service

Cooperative State Research, Education, and Extension Service

Farm Service Agency

Foreign Agricultural Service

Forest Service

National Agricultural Statistics Service

Food Safety and Inspection Service


Economic Development Administration

Bureau of the Census

International Trade Administration

Department of Commerce Headquarters

National Oceanic and Atmospheric Administration

Department of Defense

Defense Contract Audit Agency

Missile Defense Agency

Defense Legal Services Agency


29/71

Appendix IIIDepartments and Agencies Reporting NoData Mining Efforts


Defense Security Service

Defense Threat Reduction Agency

Defense Logistics Agency

Defense Advanced Research Projects Agency

Defense Contract Management Agency

Defense Security Cooperation Agency

Department of Energy

Bonneville Power Administration

Southeastern Power Administration

Southwestern Power Administration

Western Area Power Administration

Department of Health and Human Services

Centers for Medicare and Medicaid Services

Administration for Children and Families

National Institutes of Health

Indian Health Service

Department of Homeland Security

Science and Technology Directorate

Management Directorate

Bureau of Citizenship and Immigration Services

Department of Homeland Security Headquarters


30/71



Department of Housing and Urban Development

Department of the Interior

Bureau of Reclamation

Bureau of Land Management

U.S. Geological Survey

Fish and Wildlife Service

Office of Surface Mining Reclamation and Enforcement

Bureau of Indian Affairs

Department of the Interior Headquarters

Department of Justice

Bureau of Alcohol, Tobacco, Firearms, and Explosives

Department of Transportation

Federal Aviation Administration

Federal Transit Administration

Federal Railroad Administration

Federal Motor Carrier Safety Administration

Federal Highway Administration

Department of the Treasury

Comptroller of the Currency

Bureau of the Public Debt

Office of Thrift Supervision


31/71



Department of the Treasury Headquarters

Bureau of Engraving and Printing

Agency for International Development

Executive Office of the President

Federal Energy Regulatory Commission

Federal Retirement Thrift Investment Board

General Services Administration

Legal Services Corporation

National Credit Union Administration

National Labor Relations Board

National Science Foundation

Office of Management and Budget

Peace Corps

Security and Exchange Commission

Smithsonian Institution

Social Security Administration

U.S. Postal service


32/71


Appendix IV

Inventories of Efforts

The following tables present selected information from our survey of 128major federal departments and agencies on their use of data mining. Thetables list the purpose of each data mining effort, whether the system isplanned or operational, and whether the system uses personal informationdata from the private sector, or data from other federal agencies. Thesurvey shows that 52 departments and agencies are using or are planning touse data mining. These departments and agencies reported 199 data miningefforts, of which 68 were planned and 131 were operational.

Table 2: Department of Agricultures Inventory of Data Mining Efforts

Features

Organization/system name Description Purpose Status

Personalinformation

Privatesector data

Otheragencydata

Department of Agriculture Headquarters

Travel Data Mart Will consolidate employee travelinformation from financial andtravel systems. Will allow for agovernmentwide e-travel systemand provide the department withinformation on the financial

ramifications of its travel.

Improvingservice orperformance

Planned Yes No No

Financial StatementsData Warehouse

Is used in the production ofconsolidated financial statements.Provides information for productsthat are used to satisfy externalreporting requirements, such asOffice of Management and Budgetand Department of the Treasuryrequirements.

Financialmanagement

Operational No No No

Financial DataWarehouse

Is the departments internalfinancial management reportingsystem. Data mining is done for adhoc and on-demand reports.

Financialmanagement

Operational Yes No No

Food and Nutrition Service

Grantee MonitoringActivitiesSoutheastRegional Office

Assists in monitoring the financialstatus of grant holders. Granteesare required to provideexpenditure reports, and analysisis performed quarterly thatmatches stated draws to theactual draws from the U.S.Treasury.




33/71

Appendix IVInventories of Efforts


Source: Department of Agriculture.

Grantee MonitoringActivitiesMountainPlains RegionalOffice

Assists in monitoring themanagement and distribution ofIndian funds for major food benefitprograms, such as food stamps, in10 grantee states.



Grantee MonitoringActivities

Southwest RegionalOffice

Maximizes on-site monitoringefforts by confirming the accuracy

of grantee accounting. Reduceson-site time, maximizes time tocomplete reviews, and hasachieved a 50 percent travelsavings.

Improvingservice or

performance


Grantee MonitoringActivitiesMidwestRegional Office

Will be a reporting system toprovide reports and automate theaudit process. Plans are toacquire data mining tools to reviewand compare budgets, reports,and plans.


Planned No No Yes

Grantee MonitoringActivitiesNortheastRegional Office

Supports on-site reviews ofanalyses to confirm financialreport information.


Operational Yes Yes No

Integrated ProgramAccounting SystemData Integrity

Will create ad-hoc reportingcenters to validate accountinginformation.


Planned No No No

Natural Resources Conservation Service

National ResourceInventory Used forStatistical Analysis ofPast Soil SurveyDatabases.

Is a trending database that tracksmore than 200 resource issuessuch as monitoring erosion. Alsoprocesses statistical technology.



Risk Management Agency

CAE Is part of a congressionallymandated project to assist theRisk Management Agency incontrolling fraud, waste, and

abuse in the Federal CropInsurance Corporation program.

Detectingfraud, waste,and abuse

Operational Yes Yes Yes

(Continued From Previous Page)

Features


Personalinformation

Privatesector data

Otheragencydata


34/71



Table 3: Department of Commerces Inventory of Data Mining Efforts

Source: Department of Commerce.

Table 4: Department of Defenses Inventory of Data Mining Efforts

Features


Personalinformation

Privatesector data

Otheragencydata

U.S. Patent and Trademark Office

CompensationProjection Model inthe Enterprise DataWarehouse

Generates and makes availablecompensation projection data,both salary and benefits, oncurrent employees and onplanned hires. It also accounts forplanned attritions.

Managinghumanresources

Operational Yes No Yes

Features


Personalinformation

Privatesectordata

Otheragency data

Defense Commissary Agency

DeCA ElectronicRecordsManagement andArchive System

Will be a corporate informationsystem for managing unstructureddata. It will allow for electronicrecord keeping, documentmanagement, and automatedreceipt processes.


Planned Yes Yes Yes

Corporate DecisionSupport System/CommissaryOperationsManagement System

Mines data to produce analyticaldata on commissary operations.Provides information such as whatitems stores are selling and helpsdetermine whether cashiers arebeing honest.



Defense Information Systems Agency

Enterprise BusinessIntelligence System

Will replace the currentmanagement informationenvironment, which includesoperations, reporting, billing,statistics, and other managementinformation activities.


Planned No No No


35/71



Defense Intelligence Agency

Insight SmartDiscovery

Will be a data mining knowledgediscovery tool to work againstunstructured text. Will categorizenouns (names, locations, events)and present information in images.

Analyzingintelligenceand detectingterroristactivities

Planned Yes No Yes

Verity K2 Enterprise Mines data from the intelligence

community and Internet searchesto identify foreign terrorists or U.S.citizens connected to foreignterrorism activities.

Analyzing

intelligenceand detectingterroristactivities


PATHFINDER Is a data mining tool developed foranalysts that provides the ability toanalyze government and privatesector databases rapidly. It cancompare and search multiplelarge databases quickly.



Autonomy Is a large search engine tool thatis used to search hundreds ofthousands of word documents. Isused for the organization and

knowledge discovery ofintelligence.

Analyzingintelligenceand detectingterrorist

activities

Operational No No Yes

Department of the Air Force

ANG DataWarehouseGuardian

Will be used to measure militaryreadiness. It incorporatesinformation on all disciplines toprovide management informationneeded to assess militaryreadiness.

Measuringmilitaryreadiness

Planned Yes No No

Integrated SpaceWarfare Center(SWC) InformationSystem

Will be an internal databasecontaining information on alldevelopment/execution activitieswithin the SWC. Will be used by allmanagement and analystpersonnel to track and align thecenters activities to warfighterneeds, report on execution status,financial status, schedule status,and performance measurements.


Planned Yes No No

Safety AutomatedSystem (SAS)

Will query databases to findautomation mishaps. Governed byDirective 920124 and will allow forthe investigation and reporting ofidentified automation mishaps.

Improvingsafety

Planned Yes No No


Features


Personalinformation

Privatesectordata

Otheragency data


36/71



Enterprise BusinessSystem

Will support strategic planning,assist in building scientific andtechnical budgets for the AirForce, and serve as a launch pointfor all new programs. Researchand development case files will bemaintained for 75 years; theactivity indexes, catalogs, and

tracks these files.


Planned No No Yes

Genomic andProteomic ResultsAnalysis

Analyzes National Institutes ofHealths genetic data.

Analyzingscientific andresearchinformation


IG CorporateInformation System

Enhances combat readiness andmission capabilities for Air CombatCommand units and commanders.It assists in preparing for andconducting inspections.



Computer NetworkDefense System

Evaluates network activities tocreate rules for intrusion detectionsystem signature sets.

Improvinginformationsecurity


FAME Will serve as a central repositoryfor Air Force manpowerinformation. Will track manpowerand unit authorization funding.


Planned No No Yes

Resource Wizard Serves as a manpower trackingsystem. Tracks positions andcaptures data for specific fundingpurposes.



GovernmentPurchase Card

Is used in overseeing purchasesmade by Air Force personnel withgovernment-provided credit cards.



Ambulatory DataSystem Queries

Tracks the initial diagnosis ofpatients with the results of furthertesting and diagnosis. Allows forearly notification of diseases andinjuries.

Monitoringpublic health


Modus OperandiDatabase

Is an investigative tool used toidentify and track trends incriminal behavior. It linkscharacteristics of crimes andprovides details on crime scenesand other crime factors.

Detectingcriminalactivities orpatterns



Features


Personalinformation

Privatesectordata

Otheragency data


37/71



Executive DecisionSupport System

Takes data from all functionalmetric balances. Processes chartsand graphs to identify trends andto make sure goals areaccomplished.



Inspire Is a tool that assists in providing anarrative description of all

research and development that isbeing conducted within the AirForce. Provides cost andmilestone information on researchand development projects.

Performingstrategic

planning


Discoverer Is used to manage personnelrecords, including individualaliases and histories.



Requirements andConcepts System

Will serve as a repository for newsystem projects and systemrequirements. It will be availablefor consultation for information onall project requests and identifiedrequirements.


Planned No No No

Business Objects Is a commercial off-the-shelf toolthat is used to analyze and reporton human resources activities.



THRMIS Uses commercial off-the-shelfsoftware to maintain a datawarehouse of integrated inventoryand manpower data for the TotalForce: active duty (officer andenlisted), Air Force Reserve, AirNational Guard, and civilians. Isused to assess and analyze thehealth of the Air Force.



SAS Is a Web-enabled personnel datasystem that gives authorizedusers worldwide the ability to

tabulate demographic data onrecruitment, promotion, andretention.



Oracle HR Is a personnel managementsystem that manages informationfor promotions, pay grades,clearances, and other informationrelevant to human resources.




Features


Personalinformation

Privatesectordata

Otheragency data


38/71



Health Modeling andInformatics DivisionData Mart

Provides information and decisionsupport to the Air Forceheadquarters surgeon general fordecision making, policydevelopment, and resourceallocation. It also providesperformance information andanalysis to medical field units in

support of performancemeasurement objectives.



FIRST EDV (BRIO) Will deal with Air Force budgetsand other components of itsfinancial environment. Historicalanalyses and trend analyses willbe performed on the budgetprocess.


Planned No Yes No

IG World Is used to store and track data andrequirements, such as lodging andaugmentee requirements, for thePAC inspector general.



Department of Defense Headquarters

AutomatedContinuingEvaluation System

Will be used to improve personnelsecurity continuing evaluationefforts within Department ofDefense (DOD) by identifyingissues of security concernbetween the normalreinvestigation cycle for those whohold DOD security clearances andhave signed a consent form that isstill in effect.



Department of the Navy

Human ResourceTrend Analysis

Is used to improve Navyreadiness. Data on personnelmanning levels are mined toensure that each Navy unit has

the correct number of trainingpersonnel aboard.




Features


Personalinformation

Privatesectordata

Otheragency data


39/71



U.S. Naval Academy Allows for the assessment ofacademic performance ofmidshipmen. It includesdemographic information,information on grades,participation in sports, leadershippositions, etc. It is an extension ofthe registrars system and is

mined for comparisons andtrends.



Navy Training MasterPlanning System

Provides overall Navy traininginformation to assist in deliveringNavy training in the most efficientmanner. Pertinent data frommultiple databases areconsolidated into a singledatabase that is mined.



DHAMSMultidimensionalCubes

Is a database that containsinformation on the time andattendance of 3,000 marinersacross 120 ships. Allowsmanagers to look at what people

were doing at a particular time andto look across the fleet as a wholeand compare ship activities.



National CargoTracking Plan CargoTracking Division

Is used to conduct predictiveanalysis for counterterrorism,small weapons of massdestruction proliferation, narcotics,alien smuggling, and other high-interest activities involvingcontainer shipping activity.


Operational No Yes No

Supply ManagementSystemMultidimensionalCubes

Reduces downtime on ships byallowing for the analysis of shipparts information. The datawarehouse contains data on everypart that has been ordered since

the 1980s, and hasmultidimensional information oneach part. Failure rates can becalculated and improvements canbe identified.




Features


Personalinformation

Privatesectordata

Otheragency data


40/71



Type CommandersReadinessManagement System

Is designed to provide a fullyintegrated environment for onlineanalytical processing of readinessindicators. Examples of readinessindicators include status ofsupplies available, equipment inoperation, health status, andcapabilities of the crew.

Measuringmilitaryreadiness


FATHOM (APMCHuman Resources)

Will be an internal program andproject tool used to improvestaffing, recruiting, and managingday-to-day operations.


Planned Yes No No

Navy Training QuotaManagement System

Is used for planning andforecasting training needs basedon skill requirements.



National Geospatial-Intelligence Agency

OLAP (On-LineAnalyticalProcessing)

Will provide aggregations ofimagery system performance datafor management officers andsenior source decision makers tocharacterize system performance

and contribution to intelligenceissues of national priority.


Planned No No No

CITO Data Mining Will evaluate and identify imagerysystem performance trends foroptimization, monitoring, orreengineering.


Planned No No No

InformationRelevance Prototype

Will establish an informationrelevancy prototype to serve as aframework for communityevaluation of commercialinformation relevanceapproaches, methods, andtechnology. The term informationrelevance refers to the ability ofusers to receive or extract, then

display and describe, informationwith measurable satisfactionaccording to their need.


Planned No No No

U.S. Marine Corps

Operational DataStore Enterprise

Is used for workforce planning. Managinghumanresources



Features


Personalinformation

Privatesectordata

Otheragency data


41/71



Source: Department of Defense.

Global CombatSupport SystemsMarine Corps

Will be a physical implementationof the IT enterprise architecturedesigned to support bothimproved and enhanced marineair/ground task force combatservice support functions andcommander and combatantcommander joint task force

combatant support informationrequirements. Data mining willallow for interoperability withlegacy Marine Corps systems andallow for a shared dataenvironment.


Planned No Yes No

Total Force DataWarehouse

Is a system whose primarypurpose is workforce planning andworkforce policy decision making.It contains current (after 30 days)and historical workforce data.



Marine CorpsRecruitingInformation Support

System

Is a Web-based informationsystem used for managing assetsand tracking enlisted and officer

accessions into the Marine Corps.




Features


Personalinformation

Privatesectordata

Otheragency data


42/71



Table 5: Department of Educations Inventory of Data Mining Efforts

Features


Personalinformation

Privatesectordata

Otheragency data

Citizenship of PLUSLoan BorrowersNational StudentLoan Data Systems

Looks for issues regardingcitizenship among its PLUS loanborrowers. Flags records basedon selected criteria and requestsadditional information from

schools.



Foreign SchoolsInitiatives NationalStudent Loan DataSystem/CentralProcessing

Is a proactive investigation effortthat looks at whether financial aidwas granted individuals attendingforeign institutions during periodsof nonenrollment.



ProfessionalJudgment Practices:Title IV Pell Grants,National StudentLoan Data

Used to determine whenprofessional judgment has beenexercised for special situationswhere families cannot affordcollege expenses.



Title IV ApplicantDeath DatabaseMatch

Compares Department ofEducation data with the SocialSecurity Administrations death

database to detect fraud orcriminal activity.



Title IV Loans withNo Applications

Will compare information from theFree Application for FederalStudent Aid Program with theFederal Family Education LoanProgram to identify fraud.


Planned Yes No No

OIGProjectStrikeback

Compares Department ofEducation and Federal Bureau ofInvestigation data for anomalies.Also verifies personal identifiers.



Accuracy of U.S.Department of

Education PersonalData

Audits and verifies personalinformation that is contained in the

Department of Educationspersonal data system.


and abuse


Impact of CohortDefault RateRedefinitionNational StudentLoan Data System

Audits data to determine theimpact of legislation that extendedthe college loan repayment defaultperiod from 180 to 270 days.

Legislativeimpact



43/71



CheckFreeSoftware/PurchaseCard Program

Takes monthly billing informationfrom the Bank of America tocreate reports on purchases,purchase quantity, and frequencyof purchases. Data are mined forinstances of fraud or abuse.



Improper Pell Grant

Payment Activity

Will compare Pell Grants issued

with the amounts received andlook at the eligibility of grantrecipients.

Detecting

fraud, waste,and abuse

Planned Yes No No

Title IV Identity TheftInitiative

Helps identify patterns and trendsin identity theft cases involvingloans for education. Provides aninvestigative resource for victimsof identity theft.



Title IV ApplicantUse of MultipleAddresses/CentralProcessing System

Reviews addresses listed on TitleIV applications to see if they arevalid. For example, jails oremployment addresses are notconsidered valid addresses.



Lapsed

Funds/ImproperDraw of FederalGrant Proceeds

Identifies funds that remain in the

grants and payment processingsystem beyond the time period forallocating the funds.

Improving

service orperformance


Decision SupportSystem with OnlineAnalytical ProcessingQuery

Will support the departmentsperformance-based initiative. Willallow custom queries of schoolsfrom state and local databases fordemographics and test scores.


Planned No No No

Grant Administrationand Payment System

Assists in managing grantactivities and aids in detectinginstances of fraud or abuse ingrant activities.



Budget ExecutionSupport

Uses information in the NationalStudent Loan Data System and asample drawn from it to estimatecohort distributions for financialactivities related to the FederalFamily Education Loan Programpursuant to the Credit Reform Act.

Financialmanagement


Pell Grant ModelAssumptions

Provides estimates on the totalcost of the Pell Grant program. Ituses data from previous years andmakes assumptions for futureyears.

Financialmanagement



Features


Personalinformation

Privatesectordata

Otheragency data


44/71



Source: Department of Education.

National StudentLoan Data System

Compiles student loan informationfrom the guaranteeing agencies.Is used for eligibility tracking andto calculate default rates.



Loan ModelAssumptions

Estimates the cost of loanprograms. Also analyzes loandefault behavior.

Financialmanagement


Office of theInspector General(OIG) Projects:Tumbleweed/Snowball

Is part of an OIG investigation todetermine potential fraud offinancial aid grants primarily inNew Hampshire.



Central ProcessingSystem

Processes applications for studentaid. Contains data on more than13 million applications. Data aremined for demographic trends.



Direct Loan ServicesSystem

Is used to track the life of studentdirect loans and to monitor loanrepayments.



CheckFreeSoftware/Travel Card

Program

Uses monthly billing informationfrom Bank of America to create

reports on travel expenditures tolook for improper use of travelcards.


and abuse



Features


Personalinformation

Privatesectordata

Otheragency data


45/71



Table 6: Department of Energys Inventory of Data Mining Efforts

Source: Department of Energy.

Features


Personalinformation

Privatesectordata

Otheragency data

CounterintelligenceAutomatedInvestigativeManagement System(CI-AIMS)

Is an investigative managementsystem used by Department ofEnergy (DOE) field sites to trackinvestigative cases on individualsor countries that threaten DOE

assets. Information stored in thisdatabase is also used to supportfederal and state law enforcementagencies in support of nationalsecurity.



Autonomy Will be used to mine a myriadintelligence-related databaseswithin the intelligence communityto uncover criminal or terroristactivities relating to DOE assets.


Planned Yes No No

CounterintelligenceAnalytical ResearchData System(CARDS)

Is used to log briefings anddebriefings given to DOEemployees who travel to foreigncountries or interact with foreignvisitors to DOE facilities. Data aremined to identify potential threatsto DOE assets.




46/71



Table 7: Department of Health and Human Services Inventory of Data Mining Efforts

Features


Personalinformation

Privatesector data

Otheragency data

Agency for Healthcare Research and Quality

National PatientSafety Network

Will contain reports on adversemedical events that are filed byhospitals. The planned networkspurpose is to take out patientpersonal identifiers and otheritems that may violate certainrules and create a warehouse thatcan be used by registered andunregistered users to evaluate andimplement patient safety andquality measures. The network willbe used to create tools thathospitals can use for makingquality improvements.


Planned No No No

Centers for Disease Control and Prevention

BioSense Enhances the nations capability torapidly detect bioterrorism events.

Analyzingintelligenceand detectingterrorist

activities

Operational No Yes Yes

Department of Health and Human Services Headquarters

DHHS BloodMonitoring Program

Monitors the countrys bloodsupply by keeping an inventory onred blood cells and platelets andmonitors blood supply shortages,the nature of the shortage, andsize of the shortages.

Monitoringpublic health


Food and Drug Administration

MissionAccomplishment andRegulatoryCompliance ServicesSystem

Is a comprehensive redesign andreengineering of two core mission-critical legacy systems at Foodand Drug Administration (FDA)that support the regulatory

functions that primarily take placein FDA's field offices.

Monitoringfood or drugsafety



47/71



Turbo EstablishmentInspection Report

Provides a standardized databaseof citations of regulations andstatutes, and help investigators inpreparing reports. It will collectdata on specific observationsuncovered during inspections andprovide a more uniform formatnationwide that will allow forelectronic searches and statistical

analysis to be performed bycitation.

Improvingsafety


PhoneticOrthographicComputer Analysis

Is a search engine that providesresults indicating how similar twodrug names are on a phonetic andorthographic basis. Its purpose isto help in the safety evaluation ofproposed proprietary names toreduce drug name confusion afteran application is approved by theFDA.

Improvingsafety


MPRIS DataWarehouse

Will provide data to support enduser ad-hoc query analysis andstandard reporting needs. It will

provide the foundation for a centralreporting repository that can beused to populate business-specificdata marts.


Planned No No No

Development andDeployment ofAdvanced AnalyticalTools for Drug SafetyRisk Assessment

Will develop advanced softwaretools for quantitative analysis ofdrug safety data. Medical officersand safety evaluators will usethese advances in software tools.

Analyzingscientific andresearchinformation

Planned Yes Yes Yes

Add data miningcapability to CFSANAdverse EventReporting System

Is a comprehensive system fortracking, reviewing, and reportingadverse event incidences involvingfoods, cosmetics, and dietarysupplements. Integrating andcentralizing the system and

eliminating patchwork systemsmake information on theseadverse events available tofederal, state, and localgovernments as well as to industryand the public in a more timelyand efficient manner.

Monitoringfood or drugsafety

Planned Yes Yes Yes


Features


Personalinformation

Privatesector data

Otheragency data


48/71



Source: Department of Health and Human Services.

Table 8: Department of Homeland Securitys Inventory of Data Mining Efforts

Health Resources and Services Administration

HRSA GeospatialData Warehouse

Data warehouse that primarilycollects programmatic,demographic, and statistical data.



Program Support Center

Employee AssistanceProgram Analysis

Uses information from a databaseof employee assistance programcase information that does notcontain client personal identifiers.Data are mined for qualityassurance and programmanagement information that isused to enhance the quality andcost effectiveness of services.



Features


Personalinformation

Privatesector data

Otheragency data

Border and Transportation Security Directorate

Workforce ProfileData Mart

Contains payroll and personneldata and is mined for workforcetrends.



Customs IntegratedPersonnel PayrollSystem Data Mart

Is a Customs data mart containedwithin Department of HomelandSecuritys workforce profile datamart. Personnel and payroll dataare mined for workforce trends.



Internal AffairsTreasuryEnforcementCommunicationsSystem Audit DataMart

Assists the Internal Affairs group bymining criminal activity data toascertain how Customs employeesare using the Treasury EnforcementSystem.



OperationsManagementReports Data Mart

Assists in managing the operationof all ports of entry for incomingcarriers, people, and cargo. Helpsin making resource (people andequipment) allocation andoperational improvement decisions.




Features


Personalinformation

Privatesector data

Otheragency data


49/71



Automated ExportSystem Data Mart

Mines data on export trade in theU.S. and produces reports onhistorical shipping and receivingtrends.



Seized Property/Forfeitures,Penalties, and FinesCase Management

Data Mart

Mines data to ensure data qualityand review work assignments.System has two components: onethat processes legal cases like a

law firm, and a second that servesas property and inventory control bytracking property seized.



Incident Data Mart Will look through incident logs forpatterns of events. An incident is anevent involving a law enforcementor government agency for which alog was created (e.g., traffic ticket,drug arrest, or firearm possession).The system may look at crimes in aparticular geographic location,particular types of arrests, or anytype of unusual activity.


Planned Yes Yes Yes

Case ManagementData Mart

Assists in managing lawenforcement cases, including

Customs cases. Reviews caseloads, status, and relationshipsamong cases.

Analyzingintelligence

and detectingterroristactivities


Emergency Preparedness and Response Directorate

Enterprise DataWarehouse

Will take data from multiple,disparate systems and integrate thedata into one reportingenvironment. The objective of theeffort is to allow for the reduction ofdata within the agency and toprovide an enterprise view ofinformation necessary to drivecritical business processes anddecisions. Data on internal human

resources, all aspects of disastermanagement, infrastructure,equipment location, etc., will beused.

Disasterresponse andrecovery

Planned Yes Yes Yes

Information Analysis and Infrastructure Protection Directorate

Analyst Notebook I2 Correlates events and people tospecific information




Features


Personalinformation

Privatesector data

Otheragency data


50/71



Source: Department of Homeland Security.

Automatic MessageHandling System(Verity)

Automatically takes messages fromexternal agencies and routes themto appropriate recipients


Planned No No Yes

U.S. Coast Guard

ReadinessManagement

System

Assists in ensuring readiness for allCoast Guard missions.

Improvingservice or

performance


CG Info Provides one-stop shopping forCoast Guard information. It is thecentral location and commoninterface for the entire Coast Guardto gain near real-time access todata from multiple, disparate CoastGuard information systems. Itprovides a single interface for usersto view mission-critical supportdata.



U.S. Secret Service

CriminalInvestigation

Division Data Mining

Mines data in suspicious activityreports received from banks to find

commonalities in data to assist instrategically allocating resources.

Detectingcriminal

activities orpatterns



Features


Personalinformation

Privatesector data

Otheragency data


51/71



Table 9: Department of the Interiors Inventory of Data Mining Efforts

Source: Department of the Interior.

Features


Personalinformation

Privatesectordata

Otheragency data

Minerals Management Service

Data Mining of theTechnical InformationManagement System(TIMS) Database

Is a corporate database for oil andgas leases. The database ismined in support of policydevelopment. One area of datamining is identification of leasesthat will be abandoned in the nearfuture. Data mini

Documents

DATA MINING - Federal Efforts Cover a Wide Range of Uses [Implementing Data Mining Systems]