Applied Data Science & AI

From Data to Informed Decisions

My approach to applied data science emphasizes understanding problems, evaluating evidence, and using analytical methods to support informed decision-making.

As an Information Systems scholar and educator, I view data science and artificial intelligence not simply as collections of computational techniques, but as approaches to extracting meaningful insights from information while recognizing the assumptions, uncertainties, and limitations inherent in analytical processes.

My interests encompass predictive and exploratory analytics, computational analysis of textual information, and the evaluation of AI-assisted outputs. Across these areas, I emphasize methodological rigor, information quality, interpretability, and the responsible use of analytical results.

Areas of Applied Expertise

Data Quality and Analytical Problem-Solving

Meaningful analysis begins with a well-defined problem and an understanding of the underlying data. My work emphasizes data preparation, exploratory analysis, identifying relevant patterns, and evaluating whether available information is appropriate for the questions being investigated.

I approach analytical work as a process of connecting business or research questions with suitable evidence and methods, rather than applying techniques without considering their purpose.

Predictive and Exploratory Analytics

Predictive and exploratory methods provide different ways to understand patterns, relationships, and potential outcomes. My approach to machine learning is guided by the CRISP-DM framework, emphasizing the full analytical process—from understanding business problems and preparing data to developing, evaluating, and interpreting models.

I emphasize the distinction between a model that performs well according to a statistical metric and one that produces meaningful, useful, and defensible insights.

Text Analytics and Information Representation

Textual information presents analytical challenges that differ from those encountered with structured data. My work explores how textual information can be represented, compared, and analyzed for tasks such as classification and semantic similarity.

I am particularly interested in how different representation approaches capture meaning, where they fall short, and how these limitations influence analytical conclusions.

AI Evaluation and Human Judgment

AI systems can assist with information processing, analysis, and decision-making, but their outputs require critical evaluation. My interests include the reliability, accuracy, completeness, and relevance of AI-generated information, as well as the role of human judgment in interpreting and verifying AI-assisted outputs.

This area connects my applied analytical interests with my broader research on information quality, trust, and human–AI interaction.

Analytical Work and Applications

Evaluating Alternative Approaches to Text Representation

An exploratory analytical project comparing how different approaches represent and measure similarities between textual statements. The analysis examines paraphrases, negation, and semantic relationships to illustrate how representation choices can affect analytical results. It also demonstrates why numerical similarity should not automatically be interpreted as agreement in meaning.

Machine Learning for Business Problems

Teaching-oriented analytical projects involving predictive modeling, exploratory analysis, and model evaluation. These activities emphasize the full analytical process: defining a question, preparing information, selecting appropriate methods, evaluating results, and communicating findings in a business context.

Connecting Research, Analytics, and Education

My work in applied analytics complements my research and teaching in Information Systems. In research, computational approaches can help examine complex patterns in digital information and technology-mediated interactions. In teaching, I emphasize helping students understand not only how analytical methods work, but also when to use them, how to evaluate their limitations, and how to interpret their results responsibly.

Across these activities, my focus remains on the quality of analytical reasoning and the defensibility of conclusions—not the tools used to produce them.