Introduction
The Capstone Project for Big Data Specialization in Coursera.

In this culminating project, I built a big data ecosystem using machine learning and data analysis tools to analyze the data sets generated from an imaginary massively multiplayer online game, "Catch the Pink Flamingo". The final results were reported as slides to the imaginary high-level officers of the game developer company. Several solutions were recommended as to increase game revenue.

The whole report includes a series of steps as data acquiring, exploring, preparing, analyzing, and reporting.

Major techniques and softwares used in this project:

• Splunk for data previewing and exploring
• KNIME for Decision Tree classification model
• Spark ML for Clustering model
• Neo4j for relationship analysis and visualization

Data
There are two major parts of data:

1. Game related data,
2. Conversation data between users or teams.

The detailed explanation of each file and every step of data processing and analysis can be found in the file of Technical_appendix.pdf.

Report
The full report can be seen in the file Final_report.pdf.