How I Spent My Day | 06-05-2025

It was a Tuesday morning, so I rushed to the school after my normal morning routine, which has to do with my cleaning of the room, taking my bath, and also writing my article for the day. Recall that my system hard disc had an issue, and for that reason I lost some of my documents because I didn't save them on my OneDrive.

predicted_vs_actual_all_models.png

My aim of going to the school that early was to check if I could recover some of the documents, but all efforts to see them proved abortive. I started afresh from the beginning using the school system. I was able to write the introductory part of the research because I already had hard materials stored in my RefWorks, so sourcing materials wasn't much of an issue.

While on it, I felt the need to get a new system, and then I went to the shop close to the school here in Bolton, and I was able to get a very good system, which I believe will save me in all that I will be doing for my master's and PhD research. I bought the system at the rate of £500, which is 1M+ in Nigerian currency.

Immediately after the purchase I hurried home to set up the system because I needed to work on my assessment because my deadline for submission was fast approaching. I have two assessments to submit on the 9th of this month, and from today I only have two days to make them ready and available for submission.

I worked throughout the night, and I was able to write the literature review, and currently I'm working on the methodology. I have initially carried out the analysis because in my project I tried to check which machine learning algorithm can do well in the prediction of uterine cancer. In my research I compared 4 different machine learning models, namely, Decision Tree, Random Forest, Support Vector Machine (SVM) and the XGBoost Regressor.

From my analysis I checked for metrics performance of MAE, RMSE and R² score, and from the analysis random forest regressor performed better with an R² score of 0.655, which means that when compared to the others, it can be adopted for clinical prediction. So that is what my analysis was all about. I'm still working on the entire paper, and I also intend to have it published thereafter.

Sort:  

Upvoted! Thank you for supporting witness @jswit.