Evaluation of a semi-autonomous assistant for exploratory data analysis

Robert St. Amant, Paul R Cohen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

AIDE is a knowledge-based planning assistant for intelligent data exploration that draws on research in mixed-initiative planning and collaborative systems. AIDE incrementally explores a dataset, guided by user directives and its own evaluation of the data. The system is mixed-initiative: it semi-autonomously pursues high- and low-level goals but allows the user to review and potentially override its decisions. This paper briefly describes the exploratory task, AIDE's architecture, and how the system interacts with the user. The bulk of the paper is devoted to an experiment in which we compared the performance of human subjects analyzing data with and without AIDE. Although subjects each worked with AIDE for only a few hours, the system clearly influenced the efficiency and coherence of their exploration. We surmise that AIDE facilitates data analysis primarily by helping analysts navigate through the large space of decisions involved in exploring a dataset.

Original languageEnglish (US)
Title of host publicationProceedings of the International Conference on Autonomous Agents
EditorsW.L. Johnson
Pages355-362
Number of pages8
StatePublished - 1997
Externally publishedYes
EventProceedings of the 1997 1st International Conference on Autonomous Agents - Marina del Rey, CA, USA
Duration: Feb 5 1997Feb 8 1997

Other

OtherProceedings of the 1997 1st International Conference on Autonomous Agents
CityMarina del Rey, CA, USA
Period2/5/972/8/97

Fingerprint

Planning
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

St. Amant, R., & Cohen, P. R. (1997). Evaluation of a semi-autonomous assistant for exploratory data analysis. In W. L. Johnson (Ed.), Proceedings of the International Conference on Autonomous Agents (pp. 355-362)

Evaluation of a semi-autonomous assistant for exploratory data analysis. / St. Amant, Robert; Cohen, Paul R.

Proceedings of the International Conference on Autonomous Agents. ed. / W.L. Johnson. 1997. p. 355-362.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

St. Amant, R & Cohen, PR 1997, Evaluation of a semi-autonomous assistant for exploratory data analysis. in WL Johnson (ed.), Proceedings of the International Conference on Autonomous Agents. pp. 355-362, Proceedings of the 1997 1st International Conference on Autonomous Agents, Marina del Rey, CA, USA, 2/5/97.
St. Amant R, Cohen PR. Evaluation of a semi-autonomous assistant for exploratory data analysis. In Johnson WL, editor, Proceedings of the International Conference on Autonomous Agents. 1997. p. 355-362
St. Amant, Robert ; Cohen, Paul R. / Evaluation of a semi-autonomous assistant for exploratory data analysis. Proceedings of the International Conference on Autonomous Agents. editor / W.L. Johnson. 1997. pp. 355-362
@inproceedings{11fd9bfc5c0441689d5e050167dd715b,
title = "Evaluation of a semi-autonomous assistant for exploratory data analysis",
abstract = "AIDE is a knowledge-based planning assistant for intelligent data exploration that draws on research in mixed-initiative planning and collaborative systems. AIDE incrementally explores a dataset, guided by user directives and its own evaluation of the data. The system is mixed-initiative: it semi-autonomously pursues high- and low-level goals but allows the user to review and potentially override its decisions. This paper briefly describes the exploratory task, AIDE's architecture, and how the system interacts with the user. The bulk of the paper is devoted to an experiment in which we compared the performance of human subjects analyzing data with and without AIDE. Although subjects each worked with AIDE for only a few hours, the system clearly influenced the efficiency and coherence of their exploration. We surmise that AIDE facilitates data analysis primarily by helping analysts navigate through the large space of decisions involved in exploring a dataset.",
author = "{St. Amant}, Robert and Cohen, {Paul R}",
year = "1997",
language = "English (US)",
pages = "355--362",
editor = "W.L. Johnson",
booktitle = "Proceedings of the International Conference on Autonomous Agents",

}

TY - GEN

T1 - Evaluation of a semi-autonomous assistant for exploratory data analysis

AU - St. Amant, Robert

AU - Cohen, Paul R

PY - 1997

Y1 - 1997

N2 - AIDE is a knowledge-based planning assistant for intelligent data exploration that draws on research in mixed-initiative planning and collaborative systems. AIDE incrementally explores a dataset, guided by user directives and its own evaluation of the data. The system is mixed-initiative: it semi-autonomously pursues high- and low-level goals but allows the user to review and potentially override its decisions. This paper briefly describes the exploratory task, AIDE's architecture, and how the system interacts with the user. The bulk of the paper is devoted to an experiment in which we compared the performance of human subjects analyzing data with and without AIDE. Although subjects each worked with AIDE for only a few hours, the system clearly influenced the efficiency and coherence of their exploration. We surmise that AIDE facilitates data analysis primarily by helping analysts navigate through the large space of decisions involved in exploring a dataset.

AB - AIDE is a knowledge-based planning assistant for intelligent data exploration that draws on research in mixed-initiative planning and collaborative systems. AIDE incrementally explores a dataset, guided by user directives and its own evaluation of the data. The system is mixed-initiative: it semi-autonomously pursues high- and low-level goals but allows the user to review and potentially override its decisions. This paper briefly describes the exploratory task, AIDE's architecture, and how the system interacts with the user. The bulk of the paper is devoted to an experiment in which we compared the performance of human subjects analyzing data with and without AIDE. Although subjects each worked with AIDE for only a few hours, the system clearly influenced the efficiency and coherence of their exploration. We surmise that AIDE facilitates data analysis primarily by helping analysts navigate through the large space of decisions involved in exploring a dataset.

UR - http://www.scopus.com/inward/record.url?scp=0030710869&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030710869&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0030710869

SP - 355

EP - 362

BT - Proceedings of the International Conference on Autonomous Agents

A2 - Johnson, W.L.

ER -