Analyzing data and provide insights on Financial Fraud Detection using Spark ML.
Payism simulates mobile money transactions based on a sample of real transactions extracted from one month of financial logs from a mobile money service implemented in an African country. The original logs were provided by a multinational company, who is the provider of the mobile financial service which is currently running in more than 14 countries all around the world.
We will take the whole dataset and try three different classification models in spark ML and performed data preparation, model building and validation and finally model evaluation and interpretation.
Script live at Data Science