Education in Florida PK-12

We believe in the impact education has in the community. For this reason, our goal is to understand the factors that contribute to academic success.
We are interested in understanding how students of all backgrounds could benefit from this information in order to help contribute to more educated and impactful citizens in the community.
We hope to make conclusions based on information found on academic performance, financials, crime, and mental health.

Our data provides a wide collection of information such as academic performance, financials (teacher salary, instructional supplies, base funding, and other resource allocation),
student attendance, juvenile crime, mental health among others.
Our sources include but are not limited to: the Florida Department of Education, Florida Department of Juvenile Justice, National Center for Education Statistics.


• Florida Department of Education
• U.S. Census Bureau
• FBI Uniform Crime Report
• Office of Economic & Demographic Research
• Profiles Of Florida School Districts

• We are trying to use decision tree algorithm to solve the classification problem, by using tree representation. Each internal node of the tree corresponds to an attribute, and each leaf node corresponds to a class label.
  
• We use correlation to show whether how strongly pairs of variables are related.
  
• We use PCA to emphasize variation and bring out strong patterns in a dataset. TO use it to make data easy to explore and visualize.
  

• According to the student's grades of 'Mathematics Achievement', 'Science Achievement', 'English Language Arts Achievement' and so on.
  The students' academic performance in Florida districts has been graded from A to C. Here we are using the decision tree to build a classify model from 2 sides(Teacher and Student).
  the accuracy of the student classified model is 73%, and the other one is 82%.
• below is the result that starts at the root of the decision tree and takes the decision at each level based on the appropriate feature measurement until you get to the leaf node.
  The prediction is just the majority class of the instances in that leaf node.