mmanunpat1005 Goto Github PK

air_pollution_global_health_problems

bike_sharing_forecasting

Predicting the number of available bikes at the bike-sharing stations in Morges for every 10 minutes on a predefined date using the data from a large Swiss bike-sharing company with a broad range of models and techniques (e.g. STL decomposition, ETS, ARIMA, Fourier transform, cross-validation techniques) using R language.

consumer_loan_default_prediction

Applying various machine learning classification models to predict the credit quality of new loan applicants (KNN, Logistic regression, Naïve Bayes, Decision Trees, Random Forest, Neural Networks, SVM, Linear, and Quadratic Discriminant Analysis)

customized_optimizing_mcdonald-smeal_application-

A nutrition optimization application for McDonald’s meal plans using Python, user-specific data (age, weight, activity), and key libraries/tools such as pandas, NumPy, Google OR-Tools, and Tkinter, enhancing meal preferences based on personalized health metrics.

instacart_ml

master_thesis

By employing Extreme Value Theory, this research utilizes data from EM-DAT, World Bank, and IPCC databases to uncover insights into the impact of carbon emissions on exacerbating extreme flood events and the relationship between GDP per capita and flood severity.

mmanunpat1005

oil_price_prediction

tm_ted

Utilizing machine learning and text mining, this study conducts sentiment and topic analysis on TED talk transcripts, demonstrating a high accuracy in predicting video topics and showcasing a positive sentiment trend, aligning clusters closely with established TED categories while offering insights for future research.

unilsports

This project focuses on developing an interactive dashboard, UnilSports, utilizing R Shiny and various statistical methodologies to suggest personalized sports class timetables at UNIL/EPFL Sports Center, incorporating user preferences and calorie estimations for an enhanced class-search experience.