ML_azure_notebooks
This repository contains projects that focus on building Machine Learning models with Python. The projects were made based on Miscrosoft Learn in the Azure Notebooks environment.
-
Prediction of flight delay; Data set: flightdata.csv - information about flights for a major U.S. airline Data was loaded, cleaned and prepared with Pandas Then, a machine-learning model was built with scikit-learn. Using Matplotlib, the output from the model was visualized.
-
Sentiment analysis - customer reviews; Data set: https://keras.io/api/datasets/imdb/ The main goal was to perform sentiment analysis. Neural network has been built and trained with Keras. Neural network has been used to analyze and score text. Finally model can predict sentiment in given text. An input such as "Fantastic service, I can fully recommend" will score close to 1.0 - it means that the sentiment is positive. An input such as "The worst food I have ever had" will score closer to 0.0.