Center for Machine Learning and Intelligent Systems
About  Citation Policy  Donate a Data Set  Contact


Repository Web            Google
View ALL Data Sets

× Check out the beta version of the new UCI Machine Learning Repository we are currently testing! Contact us if you have any issues, questions, or concerns. Click here to try out the new site.

Rocket League Skillshots Data Set Data Set
Download: Data Folder, Data Set Description

Abstract: This dataset contains data of players of the game Rocket League, performing different skillshots.

Data Set Characteristics:  

Multivariate, Time-Series

Number of Instances:

298

Area:

Game

Attribute Characteristics:

Real

Number of Attributes:

N/A

Date Donated

2020-08-25

Associated Tasks:

Classification

Missing Values?

N/A

Number of Web Hits:

5406


Source:

Romain Mathonat, romain.mathonat '@' gmail.com, Université de Lyon, CNRS, INSA Lyon, LIRIS, UMR5205, F-69621, France


Data Set Information:

Each skillshot performed is characterized by 18 features, composed of players inputs and in-game metrics, collected at different time, creating a multivariate time serie.
You can see what skillshots look like in the github of the project:
[Web Link]

There are seven classes, -1 representing noise (composed of failed figures and random moves)
Note that lengthes of those multivariate timeseries varie, and that sample is not collected at regular time iterval (see paper for more details).

Using our pattern mining approach, we obtained an accuracy of 87.6% on a 5-fold stratified cross-validation setup.

Classes meaning:
-1: noise
1: ceiling shot
2: power shot
3: waving dash
5: air dribble
6: front flick
7: musty flick


Attribute Information:

The first line contains the name of each of the 18 features.
The format is the following:

class_number
BallAcceleration_1, Time_1, ..., jump_1
BallAcceleration_2, Time_2, ..., jump_2
...
BallAcceleration_n, Time_n, ..., jump_n
class_number
.
.
.


Relevant Papers:

R Mathonat, JF Boulicaut, M Kaytoue, 'A Behavioral Pattern Mining Approach to Model Player Skills in Rocket League', IEEE Conference on Games 2020.



Citation Request:

@INPROCEEDINGS{mathonat2020,
author={R. {Mathonat} and D. {Nurbakova} and J. {Boulicaut} and M. {Kaytoue}},
booktitle = {{IEEE} Conference on Games, CoG 2020, Online},
title={A Behavioral Pattern Mining Approach to Model Player Skills in Rocket League,
year={2020}
}


Supported By:

 In Collaboration With:

About  ||  Citation Policy  ||  Donation Policy  ||  Contact  ||  CML