Vopaaz / big-data-psg-lgd

Dota 2 big data analysis group project for Big Data Engineering and Architecture, CSCI 5751, UMN

Big Data - PSG.LGD

Dota 2 big data analysis group project for Big Data Engineering and Architecture, CSCI 5751, UMN.

Documentations

• Proposal
• Report

Environment Setup

Prerequisites

• Java8
• MongoDB
• Spark
• Neo4j
• sbt

Configuration

Create a config.yml at the project root. You can refer to the example.

Running

ETL

mvn -P FetchStore package
java -jar target/FetchStore.one-jar.jar <start_sequence_num> <total_num> <batch_size>

You can test by setting the sequence number to 4182489531, which is a very famous professional dota2 game.

For more detail, you can refer to the ETL pipeline documentation.

Analysis

sbt
run

It will ask you to choose from

1. BadManner
2. CostTime
3. DamageRate
4. EcoDistributionTrain
5. First15minGain
6. FirstBloodPredict
7. FirstBloodTrain
8. HeroMostStats
9. MostBan
10. MostPick
11. MostPurchasedItem
12. MostUsedItem
13. Neo4j.Mongo2NeoLoader
14. TeamBattleDetector

Each one answers one of our business question, except for 13. Neo4j.Mongo2NeoLoader is used for loading data from MongoDB to Neo4j.