We came across the Accenture data analytics track and found the question posed very interesting. Could it be possible to predict success? None of us have any experience with parsing and analyzing data, yet we decided to take up the challenge and build a model for success
What it does
It takes in survey results from 12th graders and predicts whether or not the survey participant will be successful in the future.
How we built it
We first defined "success" using quantifiable values. After much deliberation, we chose to focus on the job satisfaction aspect of success and tried to predict that value.
Challenges we ran into
None of us are very skilled with parsing large abstract data, so finding and putting those techniques into practice took more time and effort than efficient. We ended up using the Pandas library in Python to manipulate the data effectively.
Accomplishments that we're proud of
We were successfully able to make a model with high accuracy and decide which features play a large role in determining success. This is also a team accomplishment as none of us were confident in analyzing large data sets until now.
What we learned
The biggest thing we learned this weekend is the creation and usage of computer made models to find correlations and predict output.
What's next for A Model for Success
We plan to continue to challenge ourselves with data manipulation and finding statistical relations in data.