Applying Data Clustering and Data Reduction Methods in High-Dimensional Survival Data Analysis

• Activity Number: 566
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Risk Analysis
• Author(s): Keivan Sadeghzadeh*+ and Nasser Fard
• Companies: Northeastern University and Northeastern University
• Keywords: big data ; data clustering ; data reduction ; high-dimensional data ; survival data

Abstract

Advancement in information technology has led to accessibility of massive and complex data in many fields. Proper management and utilization of valuable data could significantly increase knowledge and reduce cost by preventive actions, whereas erroneous and misinterpreted data could lead to poor inference and decision making, sometimes irreversible and catastrophic events, even fatalities, specifically in the field of health science. In this field, survival data analysis is a kernel of prognostication of disease, treatment and medication efficiency. Therefore, optimal methods for analysis of high-dimensional and complex large-scale survival data, which is measured and generated rapidly today, are necessary.

This paper presents proper recommendations to apply data clustering and data reduction methods and techniques as practical solutions to reduce volume of high-dimensional survival data appropriately in order to avoid aforementioned difficulties and facilitate survival data analysis.