DECISION MAKING IN DYNAMIC ENVIRONMENTS AN APPLICATION OF MACHINE LEARNING TO THE ANALYTICAL HIERARCHY PROCESS
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
The purpose of this work is to propose a method of algorithmic decision making that builds on the Analytical Hierarchy Process by applying reinforcement learning. Decision making in dynamic environments requires adaptability as new information becomes available. The Analytical Hierarchy Process (AHP) provides a method for comparative decision making but is insufficient to handle information that becomes available over time. Using the opinions of one or many subject matter experts and the AHP, the relative importance of evidence can be quantified. However, the ability to explicitly measure the interdependencies is more challenging. The interdependency between the different evidence can be exploited to improve the model accuracy, particularly when information is missing or uncertain. To establish this ability within a decision-making tool, the AHP method can be optimized through a stochastic gradient descent algorithm. To illustrate the effectiveness of the proposed method, an experiment was conducted on air target threat classification in time series developing scenarios.
How to Cite
Downloads
##plugins.themes.bootstrap3.article.details##
Threat assessments, AHP, Machine learning, Multi Criteria Decision Making
Liebhaber, M.J., Kobus, D.A., & Feher, B.A. (2002). Studies of U.S. navy cues, information order, and impact of conflicting data. Technical Report 1888. San Diego: SPAWAR Systems Center. Doi: https://doi.org/10.21236/ada406334
Menon, A., Mehrotra, K., Mohan, C.K., & Ranka, S. (1996). Characterization of a class of sigmoid functions with applications to neural networks. Neural Networks, 9(5), 819-835. Doi: https://doi.org/10.1016/0893-6080(95)00107-7
Mu, E. (2006). A unified framework for site selection and business forecasting using ANP. Journal of Systems Science and Systems Engineering, 15(2), 178–188. Doi: https://doi.org/10.1007/s11518-006-5006-6
Ruder, S. (2017). An overview of gradient descent optimization algorithms. Computer Research Repository (CoRR), 1609.04747.
Saaty, T.L. (1987). The Analytical Hierarchy Process – What it is and how it is used. Pittsburgh, PA: Pergamon Journals Ltd.
Copyright of all articles published in IJAHP is transferred to Creative Decisions Foundation (CDF). However, the author(s) reserve the following:
- All proprietary rights other than copyright, such as patent rights.
- The right to grant or refuse permission to third parties to republish all or part of the article or translations thereof. In case of whole articles, such third parties must obtain permission from CDF as well. However, CDF may grant rights with respect to journal issues as a whole.
- The right to use all or parts of this article in future works of their own, such as lectures, press releases, reviews, textbooks, or reprint books.
- The authors affirm that the article has been neither copyrighted nor published, that it is not being submitted for publication elsewhere, and that if the work is officially sponsored, it has been released for open publication.
The only exception to the statements in the paragraph above is the following: If an article published in IJAHP contains copyrighted material, such as a teaching case, as an appendix, then the copyright (and all commercial rights) of such material remains with the original copyright holder.
CDF will receive permission for publication of copyrighted material in IJAHP. This permission is not transferable to third parties. Permission to make electronic and paper copies of part or all of the articles, including all computer files that are linked to the articles, for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage.
This permission does not apply to previously copyrighted material, such as teaching cases. In paper copies of the article, the copyright notice and the title of the publication and its date should be visible. To copy otherwise is permitted provided that a per-copy fee is paid.
To republish, to post on servers, or redistribute to lists requires that you post a link to the IJAHP article, which is available in open access delivery mode. Do not upload the article itself.
Authors are permitted to present a talk, based on a paper submitted to or accepted by IJAHP, at a conference where the paper would not be published in a copyrighted publication either before or after the conference and where the author did not assign copyright to the conference or related publisher.