Introduction to Dataset
Learn the role of a dataset in game data science.
Usage of dataset
We chose to use data from Multiplayer Online Battle Arena (MOBA) games in this chapter. These types of games have become very popular, with millions of players playing them at any given time. They have been adopted in esports tournaments. MOBA games are real-time strategy games, where each player assumes a hero type and joins a team competing against another team. Each hero type has different capabilities, which allow players to perform different actions in the game. Each hero will then engage in different types of actions that may include assisting, killing, moving around, etc. The objective of the game is to destroy the other team’s base. The duration of each match varies but mostly averages about an hour of play. Someample games in this genre are League of Legends, developed by Riot and released in 2009, and Dota 2, developed by Valve and released in 2013.
There are several options to get data from MOBA games. One is to use the API provided by the developers to crawl and collect data of different matches. Another option is to use publicly available replays or data from services, of which there are many, including from services such as Kaggle.
Information collected from games
In this chapter, we’ll use the dataset of Dotalicious Gaming, which is a Dota platform with servers geographically distributed over North America and Europe. For each match, information such as the nicknames of the players, the countries from which they are playing, the start and end times, the match results, and friendships between players are included. In this lesson, we’ll discuss the data in sufficient detail to allow us to follow through with the examples used in this chapter.
Attributes in CSV file
Get hands-on with 1400+ tech skills courses.