Date of Graduation

Summer 2020

Degree

Master of Science in Computer Science

Department

Computer Science

Committee Chair

Ajay Katangur

Keywords

cloud, resource prediction, machine learning, neural networks, explainable, cooperative agents, regression

Subject Categories

Computer Sciences

Abstract

This work proposes a system for predicting cloud resource utilization by using runtime assembled cooperative artificial neural networks (RACANN). RACANN breaks up the problem into smaller contexts, each represented by a small-scale artificial neural network (ANN). The relevant ANNs are joined together at runtime when the context is present in the data for training and predictions. By analyzing the structure of a complete ANN, the influence of inputs is calculated and used to create linguistic descriptions (LD) of model behavior, so RACANN becomes explainable (eRACANN). The predictive results of eRACANN are compared against its prototype and a single deep ANN (DNN). The DNN is shown to outperform eRACANN in terms of accuracy, though eRACANN shows specialized ANN topologies facilitate more specific LDs than singular DNNs.

Copyright

© Nathan R. Nelson

Open Access

Share

COinS