Maritime Mortality

Inspiration

I was inspired to build this because I was very interested in working with data, and as I learned about libraries used for data analysis, I found that I was not able to stop, so I kept adding more capabilities to my project until time ran out.

What it does

The project lets the user input year, age, and gender to find the probability of dying in a maritime disaster in the existing dataset or predict this probability with parameters that are not necessarily in the data set. One section has a dropdown menu, which is used to create 2 types of graphs for probability vs any one of the parameters mentioned above.

How we built it

I used the Python libraries numpy, pandas, matplotlib, and scikit-learn for creating the back end to find, graph, and predict using the dataset. I used streamlit to create the front end so the user would have an easier time using my program.

Challenges we ran into

I had to do a lot of research to figure out how to do what I wanted to do for each of the three functions, and sometimes I struggled with the logic in order to get them to work, but in the end, I fixed all of the problems and got a pretty decent final product.

Accomplishments that we're proud of

I am proud of making an app with multiple capabilities, but I am especially happy with the last part, which was the machine learning part. Originally, that was not even in my plan because I thought it would be too difficult for me to understand quickly. However, once I finished the other two functions, I thought that I at least had to try it, and luckily I was able to figure it out.

What we learned

I learned how to leverage Python libraries for data analysis and machine learning so I could achieve what I wanted to do for my project. I also learned some more UI skills, since I was able to create a professional-looking website using streamlit.

What's next for Maritime Mortality

I could possibly add more options for graphs and more models for machine learning.

Built With

• matplotlib
• numpy
• pandas
• python
• scikit-learn
• streamlit