Automated synthesis of data analysis programs: learning in logic

Research output: Chapter in Book/Report/Conference proceedingConference PaperOtherpeer-review

Abstract

Program synthesis is the systematic, usually automatic construction of correct and efficient executable code from declarative statements. Program synthesis is routinely used in industry to generate GUIs and for database support. I contend that program synthesis can be applied as a rapid prototyping method to the data mining phase of knowledge discovery. Rapid prototyping of statistical data analysis algorithms would allow experienced analysts to experiment with different statistical models before choosing one, but without requiring prohibitively expensive programming efforts. It would also smooth the steep learning curve often faced by novice users of data mining tools and libraries. Finally, it would accelerate dissemination of essential research results. For the synthesis task, development on such a system has used a specification language that generalizes Bayesian networks, a dependency model on variables. With decomposition methods and algorithm templates, the system transforms the network through several levels of representation into pseudo-code which can be translated into the implementation language of choice. The system applies computational logic to make learning work. In this talk, I will present the AutoBayes system developed through a long program of research and development primarily by Bernd Fischer, Johann Schumann and others [1,2] at NASA Ames Research Center, starting from a program of research by Wray Buntine [3] and Mike Lowry. I will explain the framework on a mixture of Gaussians model used in some commercial clustering tools, and present some more realistic examples.

Original languageEnglish
Title of host publication14th International Conference ILP 2004: Inductive Logic Programming;
Subtitle of host publicationPorto; Portugal; 6 September 2004 through 8 September 2004
PublisherSpringer
Number of pages1
Publication statusPublished - 2004
Externally publishedYes
Event14th International Conference ILP 2004: Inductive Logic Programming - Porto, Portugal
Duration: 6 Sep 20048 Sep 2004

Publication series

NameLecture Notes in Computer Science
PublisherSpringer-Verlag London Ltd.
Volume3194
ISSN (Print)0302-9743

Conference

Conference14th International Conference ILP 2004: Inductive Logic Programming
CountryPortugal
CityPorto
Period6/09/048/09/04

Cite this

Buntine, W. (2004). Automated synthesis of data analysis programs: learning in logic. In 14th International Conference ILP 2004: Inductive Logic Programming; : Porto; Portugal; 6 September 2004 through 8 September 2004 (Lecture Notes in Computer Science; Vol. 3194). Springer.
Buntine, Wray. / Automated synthesis of data analysis programs : learning in logic. 14th International Conference ILP 2004: Inductive Logic Programming; : Porto; Portugal; 6 September 2004 through 8 September 2004. Springer, 2004. (Lecture Notes in Computer Science).
@inproceedings{1396395307714596a11079f32d16805d,
title = "Automated synthesis of data analysis programs: learning in logic",
abstract = "Program synthesis is the systematic, usually automatic construction of correct and efficient executable code from declarative statements. Program synthesis is routinely used in industry to generate GUIs and for database support. I contend that program synthesis can be applied as a rapid prototyping method to the data mining phase of knowledge discovery. Rapid prototyping of statistical data analysis algorithms would allow experienced analysts to experiment with different statistical models before choosing one, but without requiring prohibitively expensive programming efforts. It would also smooth the steep learning curve often faced by novice users of data mining tools and libraries. Finally, it would accelerate dissemination of essential research results. For the synthesis task, development on such a system has used a specification language that generalizes Bayesian networks, a dependency model on variables. With decomposition methods and algorithm templates, the system transforms the network through several levels of representation into pseudo-code which can be translated into the implementation language of choice. The system applies computational logic to make learning work. In this talk, I will present the AutoBayes system developed through a long program of research and development primarily by Bernd Fischer, Johann Schumann and others [1,2] at NASA Ames Research Center, starting from a program of research by Wray Buntine [3] and Mike Lowry. I will explain the framework on a mixture of Gaussians model used in some commercial clustering tools, and present some more realistic examples.",
author = "Wray Buntine",
year = "2004",
language = "English",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
booktitle = "14th International Conference ILP 2004: Inductive Logic Programming;",

}

Buntine, W 2004, Automated synthesis of data analysis programs: learning in logic. in 14th International Conference ILP 2004: Inductive Logic Programming; : Porto; Portugal; 6 September 2004 through 8 September 2004. Lecture Notes in Computer Science, vol. 3194, Springer, 14th International Conference ILP 2004: Inductive Logic Programming, Porto, Portugal, 6/09/04.

Automated synthesis of data analysis programs : learning in logic. / Buntine, Wray.

14th International Conference ILP 2004: Inductive Logic Programming; : Porto; Portugal; 6 September 2004 through 8 September 2004. Springer, 2004. (Lecture Notes in Computer Science; Vol. 3194).

Research output: Chapter in Book/Report/Conference proceedingConference PaperOtherpeer-review

TY - GEN

T1 - Automated synthesis of data analysis programs

T2 - learning in logic

AU - Buntine, Wray

PY - 2004

Y1 - 2004

N2 - Program synthesis is the systematic, usually automatic construction of correct and efficient executable code from declarative statements. Program synthesis is routinely used in industry to generate GUIs and for database support. I contend that program synthesis can be applied as a rapid prototyping method to the data mining phase of knowledge discovery. Rapid prototyping of statistical data analysis algorithms would allow experienced analysts to experiment with different statistical models before choosing one, but without requiring prohibitively expensive programming efforts. It would also smooth the steep learning curve often faced by novice users of data mining tools and libraries. Finally, it would accelerate dissemination of essential research results. For the synthesis task, development on such a system has used a specification language that generalizes Bayesian networks, a dependency model on variables. With decomposition methods and algorithm templates, the system transforms the network through several levels of representation into pseudo-code which can be translated into the implementation language of choice. The system applies computational logic to make learning work. In this talk, I will present the AutoBayes system developed through a long program of research and development primarily by Bernd Fischer, Johann Schumann and others [1,2] at NASA Ames Research Center, starting from a program of research by Wray Buntine [3] and Mike Lowry. I will explain the framework on a mixture of Gaussians model used in some commercial clustering tools, and present some more realistic examples.

AB - Program synthesis is the systematic, usually automatic construction of correct and efficient executable code from declarative statements. Program synthesis is routinely used in industry to generate GUIs and for database support. I contend that program synthesis can be applied as a rapid prototyping method to the data mining phase of knowledge discovery. Rapid prototyping of statistical data analysis algorithms would allow experienced analysts to experiment with different statistical models before choosing one, but without requiring prohibitively expensive programming efforts. It would also smooth the steep learning curve often faced by novice users of data mining tools and libraries. Finally, it would accelerate dissemination of essential research results. For the synthesis task, development on such a system has used a specification language that generalizes Bayesian networks, a dependency model on variables. With decomposition methods and algorithm templates, the system transforms the network through several levels of representation into pseudo-code which can be translated into the implementation language of choice. The system applies computational logic to make learning work. In this talk, I will present the AutoBayes system developed through a long program of research and development primarily by Bernd Fischer, Johann Schumann and others [1,2] at NASA Ames Research Center, starting from a program of research by Wray Buntine [3] and Mike Lowry. I will explain the framework on a mixture of Gaussians model used in some commercial clustering tools, and present some more realistic examples.

UR - http://www.scopus.com/inward/record.url?scp=22944471424&partnerID=8YFLogxK

M3 - Conference Paper

AN - SCOPUS:22944471424

T3 - Lecture Notes in Computer Science

BT - 14th International Conference ILP 2004: Inductive Logic Programming;

PB - Springer

ER -

Buntine W. Automated synthesis of data analysis programs: learning in logic. In 14th International Conference ILP 2004: Inductive Logic Programming; : Porto; Portugal; 6 September 2004 through 8 September 2004. Springer. 2004. (Lecture Notes in Computer Science).