May 15, 1991 - This report describes the development of a user interface to the Construction Appropriations. Control and Execution System (CAPCES) using.
US Army Corps of Engineers
USACERL Technical Report P-91/15 May 1991 Analysis Cycle Army Expert Systems for Miitary Construction,
Construction Engineering Research Laboratory
043 AD-A237 11111 11111ltil 11111111111 1111111 ll~l~ 111 A Knowledge-Based Natural Language Database Interface by Sandra F. Kappes Arthur B. Baskin Robert E. Reinke Larry B. Holder This report describes the development of a user interface to the Construction Appropriations Control and Execution System (CAPCES) using artificial intelligence techniques. CAPCES is a database used by the U.S. Army Corps of Engineers to store data generated during the Military Construction, Army (MCA) program development process. The goal of this research was to produce an interface to simplify access to CAPCES data, increasing the data's usefulness throughout the construction programming process. This research developed Expert-MCA, a proto-
type system that allows users to query CAPCES
ELECT JUN
l911U
using ordinary English. Expert-MCA combines knowledge of the MCA Cycle and the CAPCES database to translate English queries into FOCUS database commands that are executed in the CAPCES database.
Approved for public release; distribution is unlimited.
91-02448 I
I i
'The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names (loes not constitute anl official indlorsement or approval of tile use of such commercial prolUictS. Thle findings of this report are not to be construed as an official Departmini of the Army' position, unless so designated by other authorized documents.
DESTROY THIS REPORT WHEN IT IS NO LONGER NEEDED DO NOT RETURN IT TO THlE ORIGINATOR
REPORT DOCUMENTATION PAGE
I
A
F
OUB No. 0704-0188
Pubic reporting burden for this o Ketion of infornalion is eaii d to average 1 hour per reeponse, Inoluding the time for reviwng instructions. searching existing data sources, gathering and rmintaining the data needed, and corroeting and revieving the oollection of information. Send cormlwnts regarding this burden estimate or any other aspect of this collection of information, including auggestioi for reduing this burden, to Washington Headquarters Services. Directorate for informtion Operations and Reports. 1215 Jefferson Davis Highway. Suite 1204, Alington. VA 202-4302. and to the Office of Management and Budget. PaperwoAk Reduction Project (0704-0188). Washington, DC 20603. 1. AGENCY USE ONLY (Leave Blank)
2. REPORT DATE I
May 1991
3. REPORT TYPE AND DATES COVERED
Final
4. TITLE AND SUBTITLE
5. FUNDING NUMBERS
A Knowledge-Based Natural Language Database Interface
PE
4A62734
PR AT41
6. AUTHOR(S)
Sandra F. Kappes, Arthur B. Baskin, Robert E. Reinke, and Larry B. Holder 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
WU SA-AB9 S. PERFORMING ORGANIZATION REPORT NUMBER
U.S. Army Construction Engineering Research Laboratory (USACERL) 2902 Newmark Drive, PO Box 9005 Champaign, IL 61826-9005 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)
10. SPONSORING/MONITORING AGENCY REPORT NUMBER
HQUSACE ATrN: CEMP-P Washington, DC 20314-1000 11. SUPPLEMENTARY NOTES
Copies are available from the National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161 12a. DISTRIBUTIONAVAILABILITY STATEMENT
12b. DISTRIBUTION CODE
Approved for public release; distribution is unlimited.
13. ABSTRACT (Maximum 200 words)
This report describes the development of a user interface to the Construction Appropriations Control and Execution System (CAPCES) using artificial intelligence techniques. CAPCES is a database used by the U.S. Army Corps of Engineers to store data generated during the Military Construction, Army (MCA) program development process. The goal of this research was to produce an interface to simplify access to CAPCES data, increasing the data's usefulness throughout the construction programming process. This research developed Expert-MCA, a prototype system that allows users to query CAPCES using ordinary English. Expert-MCA combines knowledge of the MCA Cycle and the CAPCES database to translate English queries into FOCUS database commands that are executed in the CAPCES database.
14. SUBJECT TERMS
15. NUMBER OF PAGES
Construction Appropriations Control and Execution System (CAPCES)
Expert-MCA
natural language
56 16. PRICE CODE
knowledge based systems 17. SECURITY CLASSIFICATION OF REPORT
Unclassified NSN 7540-01-280-5500
18. SECURITY CLASSIFICATION OF THIS PAGE
Unclassified
19. SECURITY CLASSIFICATION OF ABSTRACT
Unclassified
20. LIMITATION OF ABSTRACT
SAR SMWxd Form 298 (R. Prescribed byANSI S 298-102
2-8)
2309-18
FOREWORD This work was conducted for the Directorate of Military Programs, Programming and Execution Support Office, Headquarters, U.S. Army Corps of Engineers (HQUSACE), under Project 4A 162734AT4 1, "Military Facilities Engineering Technology"; Work Unit SA-AB9, "Expert System for MCA Cycle Analysis." The HQUSACE technical monitor was Mr. John Shechey, CEMP-P. The research was performed by the Facility Systems Division (FS) of the U.S. Army Construction Engineering Research Laboratory (USACERL). Dr. Michael J. O'Connor is Chief, USACERL-FS. The USACERL principal investigator was Ms. Sandra Kappes. Dr. Arthur Baskin is an associate professor and Mr. Robert Reinke and Mr. Larry Holder are research assistants in the Department of Computer Science at the University of Illinois, Urbana, IL. The USACERL technical editor was Mr. William J. Wolfe, Information Management Office. COL Everett R. Thomas is Commander and Director of USACERL, and Dr. L.R. Shaffer is Technical Director.
2
CONTENTS
Pg
SF298 FOREWORD LIST OF FIGURES
1 2 4
1 INTRODUCTION ..................................................... Background Objective Approach Mode of Technology Transfer 2
3
5
ANALYSIS OF THE CONSTRUCTION APPROPRIATIONS CONTROL AND EXECUTION SYSTEM (CAPCES)........................................ Overview Database Structure Database Content EXPERT-MCA....................................................... Overview User Interface and Functional Capabilities Knowledge-Base Structure Virtual-Memory Frame System The Query Parser
4 CONCLUSION ....................................................... APPENDIX APPENDIX APPENDIX APPENDIX
A: B: C: D:
14
33
Project Monitoring Master File (PMMFILE) Map of CAPCES Data Elements by Functional Categories Map of CAPCES, Data Elements by Information Type MCA Events and CAPCES Data Relationships
DISTRIBUTION
acoeslon yor gRA&r DTIC TAB TIS N Un~announced -
Justiloactt Di St rIbutlIoq/ Availability Codes
3
10
34 40 44 47
FIGURES Page
Number
6
1
MCA program development flow chart
2
Sample CADETS data field entry
10
3
Map of functional category "Planning"
11
4
Map of information-type categories
12
5
DD Form 1391 submission entry
13
6
The Expert-MCA interface
16
7
The organization of Expert-MCA knowledge
22
8
Graphical structure of "Approval Date"
25
9
Graphical structure of "Advertisement"
26
10
Dictionary meanings for sample query
28
11
Sample query parse
30
12
Complete parse for sample query
32
13
Generated FOCUS code for sample query
32
4
A KNOWLEDGE-BASED NATURAL LANGUAGE DATABASE INTERFACE
1
INTRODUCTION
Background Military Construction, Army (MCA) projects must follow steps defined by Army regulations, policies, and procedures before being accepted into the MCA Program. This program includes projects that meet the requirements of 10 major defense programs: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Strategic Forces General Purpose Forces Intelligence and Communications Airlift and Sealift Guard and Reserve Forces Research and Development Central Supply and Maintenance Training, Medical, and Other General Personnel Activities Administration and Associated Activities Support of Other Nations.
Before construction can begin, each project must meet the requirements defined by the Military Construction, Army (MCA) Program Development Process (Figure 1). As a project steps through the MCA development cycle, it must be justified, reviewed, revised, programmed, and budgeted. Before construction can begin, it must be approved at the Congressional level. This process ensures that every project follow the guidelines established by the Department of Defense (DOD) to meet combat capability through the balanced allocation of resources. The program development process generates a considerable amount of data that is stored in the Construction Appropriations, Control and Execution System (CAPCES). At any given time, the CAPCES database contains information on more than 15,000 active projects with approximately 500 data elements per project, tracking each project from inception to disposal. To access the data in CAPCES, a user must have considerable knowledge of the structure and content of the database, the MCA cycle, and the FOCUS database command language. This complex retrieval process can inhibit the efficient use of CAPCES. To improve access to CAPCES data, USACERL developed a prototype knowledge-based natural language interface (Expert-MCA) to the CAPCES database. Expert-MCA enables users to pose questions to CAPCES in ordinary English and provides assistance in formulating and evaluating queries. This prototype demonstrates how the addition of a knowledge-based interface to a database can allow even casual users to quickly and easily generate accurate and meaningful reports.
5
90
E
zn
D L
0
U
0:
E
0)0
661
0)
toI
00
>
00
0 CD
O
0.
r-
0
cio
-0
CA4 Cfl
4
z
0
a.I-
0LL
o 07
0
0 0
o U 1
0
WU
CCL
4
c
nm
41M1 C)C.7
-
I
z.9
(
8
E~
'
3 t
ao 400'-
0
