We have an interdisciplinary team of faculty members involved with our Data Mine program. A brief description of each of our capabilities and expertise is given below.
Munni Begum
Munni Begum is the Jack A. Rice Distinguished Professor of Mathematical Sciences at Ball State University, an honor she received in August 2025. She has served as Director of the Data Science and Data Analytics programs since January 2022. Munni Begum teaches graduate‑level courses in statistics, biostatistics, and data analytics and is actively engaged in interdisciplinary collaborative research in public health and biomedical sciences. Her research focuses on the development of statistical and data‑analytic methodology with applications to biomedical and health science problems. Her areas of expertise include Bayesian and hierarchical modeling, survival analysis, models for correlated responses, subgroup analysis, multivariate methods, statistical learning, predictive modeling, and big‑data analytics. Munni Begum has an extensive record of graduate mentoring and has supervised more than fifty master’s theses and research projects in statistics, biostatistics, and public health.
Drew Lazar
Drew Lazar’s research and applied expertise spans survival analysis, machine learning and time series forecasting. I have experience building predictive models and data pipelines in Python and R, including time series forecasting and anomaly detection methods, and custom loss function development for threshold-based prediction problems. Drew Lazar teaches computational statistics, statistical programming, and probability at both the graduate and undergraduate level, and contribute to Ball State's MS in Data Science program on Coursera. Drew Lazar has led applied data science projects with university and community partners through the BSU Data Mine program, supervising student teams, meeting regularly with partners, and incorporating their feedback into project direction.
Liping (Lucie) Lu
Liping (Lucie) Lu, PhD, MD, is an Assistant Professor in the Department of Nutrition and Health Science at Ball State University. Her research focuses on nutritional and environmental epidemiology, with particular interests in chronic disease and maternal and child health. She received advanced training in epidemiology and biostatistics at Columbia University Irving Medical Center, where she served as a postdoctoral research scientist from 2019 to 2023. She holds an MS in Biostatistics from Columbia University and a PhD in Pediatrics with a concentration in Nutrition, as well as an MD degree from Shanghai Jiao Tong University School of Medicine.
Faezeh Soleimani
Faezeh Soleimani has been an enthusiastic member of the Ball State University family as an Assistant Professor of Mathematical Sciences since September 2021. She teaches a broad range of courses in data science, data analytics, and mathematics. Faezeh Soleimani earned her Ph.D. in Mathematics/Data Science from The University of Texas at Arlington, where she also completed a second master’s degree in mathematics/data science in 2020. Her academic journey is enriched by a master's degree in operations research and a bachelor's degree in applied mathematics. Her research is driven by a passion for optimization, machine learning, data mining, numerical analysis, and mathematical modeling. Faezeh Soleimani is interested in exploring new ideas, solving challenging problems, and contributing to advancements in these exciting areas.
Tengfei Ma
Tengfei Ma is an Assistant Professor in the Department of Nutrition and Health Science at Ball State University, with expertise in epidemiology, biostatistics, and data-driven health research. His work focuses on applying advanced statistical and computational approaches to large and complex health datasets, including human microbiome, multi-omics, and population-based cohort data. Tengfei Ma is particularly interested in translating data analytics into meaningful insights for public health and biomedical research, and he brings experience in interdisciplinary collaboration across health science, biostatistics, and informatics.
Roza Aceska
Roza Aceska is an Associate Professor of Mathematical Sciences at Ball State University, where she teaches Linear Algebra, Calculus, Mathematical Modeling, and Quantitative Reasoning. She holds a Ph.D. in Mathematics from the University of Vienna, Austria, and has explored the recovery of signals in evolutionary systems, fusion frame structured signals, and the scalability of frames generated by dynamical operators. Her research is centered on the intersections of applied harmonic analysis and signal processing, and her research interests have recently extended into graph theory and applications.
Nimish Valvi
Nimish Valvi is an Assistant Professor in the Department of Nutrition and Health Science at Ball State University, with expertise in epidemiology. His work focuses on leveraging electronic health records for population health surveillance and research in the areas of chronic and rare diseases. Most of his research has been focused on hypertension, diabetes, cancer outcomes and is currently a co-PI on a CDC funded grant focused on the surveillance and outcomes among children, adolescents, and adults with congenital heart defects in the state of Indiana.
BSU Energy Modeling Project:
1) Original Business Problem
BSU Facilities Management collects hourly energy meter data from buildings across campus but lacked tools to analyze and act on it. The project originated through Dr. Jen Coy (Computer Science), who was developing an Open Energy Dashboard and connected us with Margaret Lo, BSU's Chief Sustainability Officer. We expanded the scope beyond visualization to include modeling, forecasting, and exploratory analysis, with a focus on predicting when campus energy usage would exceed a threshold that triggers a substantial overage penalty.
2) How It Was Scoped
The project runs as a Data Mine immersive learning engagement. Facilities Management staff attend classes, meet regularly with the student team, and provide feedback through two-week Agile sprints. Work is organized into data cleaning, exploratory analysis, time series forecasting, threshold detection, and web interface development.
3) What Was Delivered
The project is ongoing. The team has delivered a data pipeline integrating meter readings, weather, and academic calendar covariates; outlier detection tools that allow the partner to identify anomalous consumption; a forecasting framework comparing SARIMAX, Prophet, and NeuralProphet; and exploratory tools revealing relationships between weather, academic schedules, and energy usage. A reliable overage prediction system and consolidated web interface are still in active development, with plans to integrate with the Open Energy Dashboard for real-time campus data.
BSU Women’s Soccer Data Analytics
1) Original Business Problem
BSU Women’s Soccer coaching staff collects data via PlayerData devices, which use a GPS system to track total distance, speed, accelerations, decelerations, and more. The coaching staff, led by head coach Andy Stoots, wanted the datamine group to focus on how a variety of load (physical exertion) metrics impacted player and team performance. The goal was to use analytics to find valuable insights into team and player performance to improve match outcomes, i.e., winning.
2) How It Was Scoped
The project also runs as a Data Mine immersive learning engagement. The student team meets the faculty mentor regularly and the coaching staff as needed. Work is organized into data cleaning, exploratory analysis and visualization, and predictive modeling.
3) What Was Delivered
Students collected data on 22 players; 8 forwards, 6 midfielders, and 8 defenders. No data was collected for goalies. Based on the literature on soccer data analytics, they calculated Team Average Match Intensity using a novel algebraic formula. The team has delivered insights for the coaching staff based on extensive exploratory data analysis.
IU Health Ball Memorial Pharmacy Residency Consulting
1) Original business problem: In spring 2020, a PGY1 pharmacy resident at IU Health Ball Memorial Hospital needed to answer two operational questions about discharge management of Clostridium difficile patients: were patients discharged on oral vancomycin actually filling their outpatient prescriptions, and what was driving the choice between vancomycin capsules and the compounded oral solution at discharge? She was directed to us through our standing consulting relationship with the PGY1 Pharmacy Residency Program, under which the program connects residents needing statistical support to our statisticians in the Department of Mathematical Sciences.
2) How it was scoped: We worked directly with the resident to scope a retrospective chart review of 75 patients. She handled de-identified data collection and clinical interpretation; I along my students worked on the full analytic pipeline: cleaning, descriptive summaries, and association testing between compliance and clinical variables (severity, care management, SNF placement, days of treatment, formulation). The timeline and reporting standard were fixed by the Great Lakes Pharmacy Resident Conference, which set both the delivery date and the level of rigor the final analysis had to meet.
3) What was delivered: A written statistical report with the headline compliance rate, prescribing-pattern breakdown, and association results for each predictor. The resident used it as the basis of a successful May 2020 conference presentation. This is one of eight PGY1 resident projects statistics faculty at Ball State have led through this relationship since 2019.