Data cleaning using machine learning

Ahmed Hassan

Bachelor's thesis

Abstract:

This thesis provides a general outlook on Data Analysis, Data Cleaning, Artificial Intelligence & Machine Learning. The main part focuses on building up a machine learning model that is used to find errors in any data set, and that will be considered with an application of machine learning known as predictive analytics. After the model cleans the errors, we will be able to evaluate the accuracy of the model practically. Firstly, Data are going to be collected from an online resource e.g., Kaggle.com, These data will then be analyzed to figure-out the type of errors needed to be considered e.g., Deleting duplicate data, Dropping/replacing missing data, Typos fixing, etc. Then we will apply the data preprocessing techniques e.g., Data cleaning, Sampling data, Imbalanced data. Then we will develop and apply the machine learning model to clean the data we pre-processed. Lastly, we will go with prediction and evaluate the model’s efficiency for appraising its usability.

Keywords

Data cleaning, Predictive models, Python, Machine Learning, Pre-processing, Data analysis

Language used: English

Date on which the thesis was submitted / produced: 18. 3. 2023

Thesis defence

• Date of defence: 16. 5. 2023
• Supervisor: Ing. Jan Pavlík, Ph.D.
• Reader: John Phillip Sabou, Ph.D.

Czech University of Life Sciences Prague

Faculty of Economics and Management

Bachelor programme:
Informatics