How I Spent My Day

Hello everyone, it is another beautiful day of the week, and I'm delighted to share with you all how I spent my day. It was a Friday and the last working day of the week, and I'm delighted that I spent that day very well because I was able to do a whole lot for myself.

We were to have a meeting in the school that very day because of the collaboration we are having to work on 3 papers for the Centre of Intelligence of Things. I'm in one of the papers, and my target is to get more publications so that after my graduation, I can apply for global talent.

If you have been following my blog, you will notice that one of my papers, for which I was the lead author, was accepted a few days ago with major corrections, which I have been working on for the past 2 days now. I have successfully learnt and improved the credibility of the research with all the corrections made on the paper.

Yesterday, while working on the collaborative project, I was also working hand in hand with my paper. I was able to include a statistical model in my analysis to improve the result, and I also incorporated a hybrid model. I then tested to check between the standalone model and the hybrid model, which performs better.

I also learnt about cross-validation and paired t-tests, which help me to check for the statistical difference in the accuracy of the models, which were close to each other in terms of prediction. My ensemble models, which have to do with Random Forest and XGBoost Regressor, were very close, so I had to check if there was a statistical difference between their results.

The p-value was around 0.0382, which is less than 0.05 and implies that there was a statistical difference at a 95% confidence level between the results obtained from Random Forest and XGBoost. The finding concluded that Random Forest wasn't only statistically better in the prediction, but also it is perfect for the provided dataset.

I was also able to address some of the few minor observations raised by the reviewers, such as inconsistency in citation, error in figure description, and a few other things. Because of the additional work for the result analysis, I had to rewrite the conclusion, future work, and recommendations. So that was how I spent my day.