Another small project developed as a Data Science exercise, adapted from a challenge proposed by Codecademy.

Starting from a CSV file provided by the platform, data was collected, cleaned and organized in dataframes.

Data refers to Jeopardy, a famous american TV show consisted basically in answering questions proposed by the host. In this case, 216,930 questions and their respectively answers, date when the show was aired, category and value (in dollars) of each question can be found within data.

In the first part of this project, a function was created to find the questions that contain all the words of a specific list. Following that, it is possible to discover the average value of questions containing a specific word, typed in by user.

In the second part, a function discovers how many times a specific word can be found in questions of both 90's and 2000's decades, and what it represents in terms of percentage of all questions in those periods. With that, it's possible to understand if the word, in a certain way, has became more or less popular in the show throughout time.

Finally, an interactive adaptation of Jeopardy was developed. The user has 3 chances to correctly answer a random question from the database.

To use the code just uncomment the function's calls.