PROJECT TITLE :

An Enhanced Visualization Method to Aid Behavioral Trajectory Pattern Recognition Infrastructure for Big Longitudinal Data - 2018

ABSTRACT:

Big longitudinal knowledge offer a lot of reliable data for call creating and are common in all sorts of fields. Trajectory pattern recognition is in an urgent want to find vital structures for such data. Developing higher and additional computationallyefficient visualization tool is crucial to guide this technique. This Project proposes an enhanced projection pursuit (EPP) methodology to raised project and visualize the structures (e.g., clusters) of huge high-dimensional (HD) longitudinal knowledge on a lower-dimensional plane. Unlike classic PP strategies doubtless helpful for longitudinal data, EPP is constructed upon nonlinear mapping algorithms to compute its stress (error) operate by balancing the paired weights for between and within structure stress whereas preserving original structure membership within the high-dimensional space. Specifically, EPP solves an NP hard optimization drawback by integrating gradual optimization and non-linear mapping algorithms, and automates the searching of an optimal number of iterations to show a stable structure for varying sample sizes and dimensions. Using publicized UCI and real longitudinal clinical trial datasets also simulation, EPP demonstrates its better performance in visualizing big HD longitudinal information.