Date of Award

12-2020

Degree Name

Doctor of Philosophy

Department

Computer Science

First Advisor

Dr. Li Yang

Second Advisor

Dr. Ala Al-Fuqaha

Third Advisor

Dr. Alvis Fong

Fourth Advisor

Dr. Jun-Seok Oh

Keywords

Visualization, multi resolution analysis, support vector machine, data mining, machine learning, entropy discretization

Abstract

Big data analysis is essential for many smart applications in areas such as connected healthcare, intelligent transportation, human activity recognition, environment, and climate change monitoring. Traditional data mining algorithms do not scale well to big data due to the enormous number of data points and the velocity of their generation. Mining and learning from big data need time and memory efficiency techniques, albeit the cost of possible loss in accuracy. This research focuses on the mining of big data using aggregated data as input. We developed a data structure that is to be used to aggregate data at multiple resolutions. The structure is built once, updated incrementally, and serves as data input for data mining algorithms instead of accessing the raw data. We adapted several existing data mining algorithms and developed new algorithms to accept multiresolution aggregated data instead of raw data as input. Example data mining algorithms include Bayes classifiers, entropy discretization, decision trees, and support vector machines. We studied visual analytics of high dimensional big data by visually exploring the multiresolution data aggregation structure. We have developed techniques for pruning and visualizing data cube lattices. Learning from multiple levels of data aggregation enables users to compromise between time, memory, and accuracy, depending on available resources and application requirements. The multiresolution aggregation gives opportunities to develop new data mining methods and algorithms.

Access Setting

Dissertation-Open Access

Download

Share

COinS