Multivariate Analysis Commons^™

Predicting Remaining Useful Life Using Multivariate Time-Series Data, Anayah Smith, Victoria Gaibor

Discovery Day - Daytona Beach

Accurate prediction of Remaining Useful Life (RUL) is critical for enabling predictive maintenance, improving system reliability, and reducing operational costs in degrading systems. This project addresses the problem of modeling and predicting RUL using multivariate time-series sensor data from the NASA CMAPSS turbofan engine dataset, with a focus on understanding how predictive performance changes across datasets of varying complexity. The objective is to develop a reproducible machine learning pipeline that captures degradation patterns and produces reliable time-to-failure predictions. The approach includes data preprocessing, exploratory data analysis, feature engineering, dimensionality reduction, and model evaluation. RUL values are computed and capped to …

A Machine-Learning Tool-Supported Methodology For Nonprofit Donor Analysis, Corbin Weiss

Campus Research Month

We developed a machine-learning tool-supported methodology for modeling the nonprofit donor relationship. This approach was demonstrated in the case of a US-based nonprofit. Conclusions were drawn from this example and tool-support provided for use by other nonprofits.

Kroger Post-Pandemic Customer Segmentation, Mario Mata, Joey Truitt, Renn Spigelmyer, Dhanuja Kasturiratna, Lisa Holden, Nitish Baidya, Hanna Tafari

Posters-at-the-Capitol

The grocery retail industry landscape has changed greatly in the wake of the pandemic. Specifically, delivery and pickup services have become more popular and customer buying habits have evolved. At the same time, improvements in data collection and analysis have allowed grocery marketing strategies to become highly individualized.

We worked with 84.51, an analytics firm, to identify customer segments for the Kroger Company based on data from 2023. Using clustering techniques, we organized customers into groups, or segments, based on similar characteristics. We identified and profiled four distinct groups of customers. Three segments were characterized by high frequency and spending …

If You Can’T Beat Them Join Them: Empirical Assessment Into How Integrating Conventional Taxis On The Uber App Impacts Conventional Taxi Ridership, Shahmeer Mohsin

CBER Conference

Since the emergence of ride-hailing platforms like Uber, conventional taxi ridership has taken a severe hit. Taxi-hailing apps like Curb and Arro have allowed conventional taxis to jump on the platform economy bandwagon and offer a similar service to ride-hailing platforms. Despite the emergence of these taxi-hailing apps, strong lock-in effects and high switching costs of popular ride-hailing platforms (Uber, Lyft, etc.) restrict the ridership volumes of conventional taxis. Recently, the ride-hailing platform, Uber has started to add conventional taxis on its app under increasing pressure from Cities and conventional taxi associations. Such integrations have the potential of increasing conventional …

Principal Component Analysis With Application To Credit Card Data, Eleanor Cain, Semhar Michael, Gary Hatfield

SDSU Data Science Symposium

Principal Component Analysis (PCA) is a type of dimension reduction technique used in data analysis to process the data before making a model. In general, dimension reduction allows analysts to make conclusions about large data sets by reducing the number of variables while retaining as much information as possible. Using the numerical variables from a data set, PCA aims to compute a smaller set of uncorrelated variables, called principal components, that account for a majority of the variability from the data. The purpose of this poster is to understand PCA as well as perform PCA on a large sample credit …