Collecting customer feedback is a great way for businesses to understand the strengths and weaknesses of their products and services; At the same time, quickly grasp the mind and needs of customers to bring them the most perfect products and services. Machine Learning has revolutionized the way people solve these problems by bringing tools that allowed companies to analyze people's reviews sentiments with highly reliable results. We use Data analytic tools and Data visualizations techniques have tried to provide eventful insights into data set. And general findings from data sets have been used by machine learning process to generate a prediction on the rating on Foody based on people's criticism.
By analysing the rating on Foody based on people's reviews, the target of this project is building an optimal model that can classify each review into 2 classes of Rating: 1(Positive) and 0(Negative).
# initialize develop environment
conda create env -n sentiment python=3.7
# install dependencies
conda install --file requirements.txt
or
pip install -r requirements.txtMake sure download and unarchived the image data set in this link before running the notebook
After modify the right path to Image dataset, you can work around with image_training.ipynb
First run dataset_preprocessing.ipynb to get the preprocessing data set
Then work around with the model_training.ipynb to see what models do
- Vu Duc Cuong - 20020282 (@jvs47)
- Vu Minh Vuong - 20020314 (@UnicornSaga)