Action2Score

1. INTRODUCTION

Multiplayer Online Battle Arena (MOBA) is one of the most successful game genres. MOBA games such as League of Legends have competitive environments where players race for their rank. 

In most MOBA games, a player’s rank is determined by the match result (win or lose). It seems natural because of the nature of team play, but in some sense, it is unfair because the players who put a lot of effort lose their rank just in case of loss and some players even get free-ride on teammates’ efforts in case of a win. 

To reduce the side-effects of the team-based ranking system and evaluate a player’s performance impartially, we propose a novel embedding model that converts a player’s actions into quantitative scores based on the actions’ respective contribution to the team’s victory. 

Our model is built using a sequence-based deep learning model with a novel loss function working on the team match. We showed that our model can evaluate a player’s individual performance fairly and analyze the contributions of the player’s respective actions. 

For more information about this paper, please refer to our paper below. 

2. PUBLICATION

Junho Jang, Ji Young Woo, and Huy Kang Kim. 2022. Action2Score: An Embedding Approach to Score Player Action. Proc. ACM Hum.-Comput. Interact. 6, CHI PLAY, Article 220 (October 2022), 23 pages. https://doi.org/10.1145/3549483

3. DATASET DOWNLOAD

If you want to download the dataset, please visit GitHub to check it out. This site contains the dataset as well as the code used in the paper. 

Download Link: Github

4. CONTACT

We welcome your feedback or comments. Leave a message freely.

Junho Jang (hkonly@korea.ac.kr), Huy Kang Kim (cenda@korea.ac.kr)  

5. SEE ALSO

Please see the page [HCRL/Datasets] to find out more online game datasets or other datasets that we have.

6. ACKNOWLEDGMENT

This was supported by Korea University Grant.