Date of Award

3-2020

Document Type

Thesis

Degree Name

Master of Science in Operations Research

Department

Department of Operational Sciences

First Advisor

Jefferey D. Weir, PhD

Abstract

A holistic approach to the algorithm selection problem is presented. The “algorithm selection framework" uses a combination of user input and meta-data to streamline the algorithm selection for any data analysis task. The framework removes the conjecture of the common trial and error strategy and generates a preference ranked list of recommended analysis techniques. The framework is performed on nine analysis problems. Each of the recommended analysis techniques are implemented on the corresponding data sets. Algorithm performance is assessed using the primary metric of recall and the secondary metric of run time. In six of the problems, the recall of the top ranked recommendation is considered excellent with at least 95 percent of the best observed recall; the average of this metric is 79 percent due to two poorly performing recommendations. The top recommendation is Pareto efficient for three of the problems. The framework measures well against an a-priori set of criteria. The framework provides value by filtering the candidate of analytic techniques and, often, selecting a high performing technique as the top ranked recommendation. The user input and meta-data used by the framework contain information with high potential for effective algorithm selection. Future work should optimize the recommendation logic and expand the scope of techniques for other types of analysis problems. Further, the results of this proposed study should be leveraged in order to better understand the behavior of meta-learning models.

AFIT Designator

AFIT-ENS-MS-20-M-137

DTIC Accession Number

AD1192878

Share

COinS