Learning Projects

This study proposes a Bayesian framework to identify the key variables influencing ride quality of a 24-row Joh Deere Exact-EmergeTM row crop planter. A robust predictive model is developed that employes Bayesian inference to estimate ride quality with reasonable accuracy and provides highly interpretable results with probability distributions. 

This project involves analyzing a dataset containing metadata about movies. The main goal is to understand what factors affect the IMDb score of a movie. The project is divided into several parts, each focusing on different aspects of data analysis and machine learning.

This project focuses on analyzing data breach cases using various data analytics techniques. The project is structured into four main parts: data import and quality assessment, data cleaning and transformation, business intelligence through data visualization and exploratory data analysis, and storytelling based on the analysis.

This paper presents applied statistical methods for orthro rectification of imagery entirely eliminating the use of GCPs. We formulated multiple statistical models and analyzed them using maximum likelihood estimation, polynomial regression, regression with basis function and boosted gradient regression. The prediction accuracies evaluated for each model show that the proposed method can be used for accurate georeferencing of photogrammetric models.