Inspiration
Our inspiration came from different machine learning tutorials, thus each of us started with different approaches and methods, but later combined and tested our algorithms together to find out the best one with the highest accuracy
What it does
We finally agreed that the Random forest classification algorithm has the best performance of all algorithms. With the highest learning accuracy and test accuracy, RFC proved its potential in learning and analyzing data with the correct data clustering, thus we focused on the RFC algorithm and made it our final submission
How we built it
We built the RFC model with the following steps
- split the dataset and pre-processing data
- build the model
- train the model
- evaluate the model with metrics
- tune the hyperparameters
Challenges we ran into
- choose the best algorithm base on performance
- analyze data and select the most important columns to feed the algorithm
- improve the performance of the RFC model
Accomplishments that we're proud of
- we tried a variety of algorithms at the beginning stage
- the RFC model is promising and has the highest accuracy
- our way of analyzing input data is efficient
What we learned
- how to split and cluster original data
- more ML algorithms and their coding technique
- how to test and improve ML algorithms
What's next for the AI dataset
there should be a better way to clean the data, so if we have more time, we can try to make the input data cleaner to achieve higher accuracy
Built With
- matplotlib
- pandas
- python
- scikit-learn
Log in or sign up for Devpost to join the conversation.