Founded in 2000 by a Bronx history teacher, DonorsChoose.org has raised $685 million for America’s classrooms. Teachers at three-quarters of all the public schools in the U.S. have come to DonorsChoose.org to request what their students need, making DonorsChoose.org the leading platform for supporting public education.

To date, 3 million people and partners have funded 1.1 million DonorsChoose.org projects. But teachers still spend more than a billion dollars of their own money on classroom materials. To get students what they need to learn, the team at DonorsChoose.org needs to be able to connect donors with the projects that most inspire them.

In the second Kaggle Data Science for Good challenge, DonorsChoose.org, in partnership with Google.org, is inviting the community to help them pair up donors to the classroom requests that will most motivate them to make an additional gift. To support this challenge, DonorsChoose.org has supplied anonymized data on donor giving from the past five years. The winning methods will be implemented in DonorsChoose.org email marketing campaigns.

Two Prizes were won in this competition.

• Stage 1 Winner is the Exploratory Data Analysis kernel. This has the extensive exploratory data analysis of this huge dataset

• Stage 2 Winner is the Recommendation kernel. Item Item Recommendations were done using

  • Text Features Only using the Project Title and Projects Essays
  • Text and Non Text Features only.The Non Text features include Category, SubCategory, Grade,ResourceCategory, SchoolState, TeacherPrefix.
  • Non Text Features only

The accuracy is highest for the recommendations using the Non Text Features only.We also did the User User Recommnedations using the Users and the Amount.