NESC9797 PATTER.

1. NAME OR DESIGNATION OF PROGRAM:  PATTER.

3. DESCRIPTION OF PROGRAM OR FUNCTION

PATTER is an interactive program with extensive facilities for modeling analytical processes  and solving complex data analysis problems using statistical methods, spectral analysis, and pattern recognition techniques. PATTER addresses the type of problem generally stated as follows: given a set of objects and a list of measurements made on these objects, is it possible to find or predict a property of the objects which is not directly measurable but is known to define some unkown  relationship. When employed intelligently, PATTER will act upon a data set in such a way it becomes apparent if useful information, beyond that already discerned, is contained in the data.

4. METHOD OF SOLUTION

In order to solve the general problem, PATTER contains preprocessing techniques to produce new variables that are  related to the values of the measurements which may reduce the number of variables and/or reveal useful information about the "obscure" property; display techniques to represent the variable space in some way that can be easily projected onto a two- or three-dimensional plot for human observation to see if any significant clustering of points occurs; and learning techniques based on both unsupervised and supervised methods, to extract as much information from the data as possible so that the optimum solution can be found.

5. RESTRICTIONS ON THE COMPLEXITY OF THE PROBLEM:

6. TYPICAL RUNNING TIME:

NEA-DB could not produce an operational conversion of the source program from the Lawrence Livermore LRLTRAN dialect to standard FORTRAN in order to run the test case. The package is therefore distributed in an untested form.

7. UNUSUAL FEATURES OF THE PROGRAM

The use of PATTER's "process instructions" allows the user a great deal of flexibility in the data analysis procedure. The verbs of these instructions are grouped into the following functional areas: pattern analysis, data preprocessing, spectral analysis, display routines, supervised learning, unsupervised learning, and utility routines.

8. RELATED AND AUXILIARY PROGRAMS

PATTER is a smaller, faster, more efficient code than the earlier RECOG program with additional capabilities. RECOG-ORNL (NESC Summary 9967) is a modification of RECOG.

10. REFERENCES

- L.A. Cox, Jr., R.H. Pritchard, and C.F. Bender,
  RECOG: A Polyalgorithm for the Analysis of Generalized Data Sets,
  An Operator's Manual,
  UCID-16443, Rev. 1, April 9, 1975.

- L.A. Cox, Jr. and C.F. Bender:
  PATTER: A Polyalgorithm for the Analysis of Generalized Data Sets.
  Principles and Practice.
  UCID-16915  (October 1975)
- W.S. Derby, J.T. Engle and J.T. Martin:
  LRLTRAN Language Used with the CHAT and CIVIC Compilers.
  LCSD-302, Rev. 1  (June 1, 1982)

11. MACHINE REQUIREMENTS:

13. OPERATING SYSTEM UNDER WHICH PROGRAM IS EXECUTED:

14. OTHER PROGRAMMING OR OPERATING INFORMATION OR RESTRICTIONS:

15. NAME AND ESTABLISHMENT OF AUTHORS

         L.C. Cox, Jr. and C.F. Bender*
         Lawrence Livermore National Laboratory
         P.O. Box 808
         Livermore, California 94550
* Contact

File name	File description	Records
NESC9797_01.001	Information File	45
NESC9797_01.002	PATTER source (LRLTRAN language)	5530
NESC9797_01.003	PATTER source (translated into FORTRAN)	5816
NESC9797_01.004	PATTER sample case input data	104