Honors College Senior Thesis Presentations

Comparing Data and Algorithmic Modeling Approaches for Count Data

Presenter Information

Andraya HackFollow

Academic Level at Time of Presentation

Senior

Major

Mathematics

Minor

Biology

List all Project Mentors & Advisor(s)

Dr. Christopher Mecklin; Dr. Gopal Nath

Presentation Format

Oral Presentation

Abstract/Description

Throughout the world there is numerous reasons a person may want to predict an outcome. A type of data that can be predicted is count data. As the name suggests count data involves the number of an occurrence or object that is in a countable form. An example of count data is the number of expected insurance claims to be received during a certain period of time. Another example is the anticipated number or articles a professor might publish. There are several approaches that can be taken in order to predict count data. There is the traditional statistical model's approach. These models are the Poisson distribution and the negative binomial distribution. A different procedure is to use computer algorithms also known as machine learning for predictions. Two specific algorithms are artificial neural networks (ANN) and k-nearest neighbors (KNN). To choose which method to use, the accuracy must be accessed for each approach.

Spring Scholars Week 2022 Event

Honors College Senior Thesis Presentations

This document is currently not available here.

Share

COinS
 

Comparing Data and Algorithmic Modeling Approaches for Count Data

Throughout the world there is numerous reasons a person may want to predict an outcome. A type of data that can be predicted is count data. As the name suggests count data involves the number of an occurrence or object that is in a countable form. An example of count data is the number of expected insurance claims to be received during a certain period of time. Another example is the anticipated number or articles a professor might publish. There are several approaches that can be taken in order to predict count data. There is the traditional statistical model's approach. These models are the Poisson distribution and the negative binomial distribution. A different procedure is to use computer algorithms also known as machine learning for predictions. Two specific algorithms are artificial neural networks (ANN) and k-nearest neighbors (KNN). To choose which method to use, the accuracy must be accessed for each approach.