NORMA eResearch @NCI Library

Mimicking the Actions of a Support Employee Using Machine Learning: Data Science Report

O'Donovan, Jordan (2023) Mimicking the Actions of a Support Employee Using Machine Learning: Data Science Report. Undergraduate thesis, Dublin, National College of Ireland.

[thumbnail of Bachelor of Science]
PDF (Bachelor of Science)
Download (1MB) | Preview


The purpose of this report is to utilise supervised classification models to predict the next step needed to be taken for a telecommunications test which occurred at Spearline Labs. This report details each step taken from cleaning the data, analysing and pre-processing it, to model training, testing and hyperparameter tuning. The impact of each feature and hyperparameter on the models was detailed and visualised.

This report found that once feature selection and hyperparameter tuning had occurred, there was less overfitting during the training of the models, thus their performance when using the testing dataset increased. The metrics measured in this paper are accuracy, precision, recall and the f-score.

The conclusions from this report are that these supervised classification algorithms are very effective in creating accurate predictions, and that the feature selection and hyperparameter tuning techniques utilised not only drastically reduce the computational power required to create the models, but they also increase the performance of said models.

Item Type: Thesis (Undergraduate)
Clifford, William
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Telecommunications Industry
Divisions: School of Computing > Bachelor of Science (Honours) in Computing
Depositing User: Tamara Malone
Date Deposited: 16 Jan 2024 16:51
Last Modified: 16 Jan 2024 16:51

Actions (login required)

View Item View Item