Kaggle 분석

배틀그라운드 등수 예측하기 - Competition / 데이터 요약

데이터사이엔티스트 2019. 10. 5. 18:03

게임을 좋아하는 저에게 아주 흥미로운 Competition이 눈에 들어왔습니다.

 

다름이 아니라 배틀그라운드 유저의 정보를 분석해 경기에서 몇 등을 할지를 예측하는 게임입니다!

정말 재밌을 거 같지 않나요 ㅎㅎ 내가 하면 왠만한 경기 꼴등 찍을거 같지만 간단한 요약 설명 적어놓겠습니다!

 

일단 학습용 데이터와 테스트용 데이터로 나누어져 있습니다.

그리고 정확도 체크를 하기 위한 서브미션 파일이 있습니다.

 

데이터 필드를 살펴보면 플레이어들에 대한 간단한 정보에 대한 데이터가 있습니다.

경기 당 킬 / 어시스트 / 데미지량 등 다양한 데이터가 있습니다.

 

링크: https://www.kaggle.com/c/pubg-finish-placement-prediction/data

 

PUBG Finish Placement Prediction (Kernels Only)

Can you predict the battle royale finish of PUBG Players?

www.kaggle.com

File descriptions

  • train_V2.csv - the training set
  • test_V2.csv - the test set
  • sample_submission_V2.csv - a sample submission file in the correct format

Data fields

  • DBNOs - Number of enemy players knocked.
  • assists - Number of enemy players this player damaged that were killed by teammates.
  • boosts - Number of boost items used.
  • damageDealt - Total damage dealt. Note: Self inflicted damage is subtracted.
  • headshotKills - Number of enemy players killed with headshots.
  • heals - Number of healing items used.
  • Id - Player’s Id
  • killPlace - Ranking in match of number of enemy players killed.
  • killPoints - Kills-based external ranking of player. (Think of this as an Elo ranking where only kills matter.) If there is a value other than -1 in rankPoints, then any 0 in killPoints should be treated as a “None”.
  • killStreaks - Max number of enemy players killed in a short amount of time.
  • kills - Number of enemy players killed.
  • longestKill - Longest distance between player and player killed at time of death. This may be misleading, as downing a player and driving away may lead to a large longestKill stat.
  • matchDuration - Duration of match in seconds.
  • matchId - ID to identify match. There are no matches that are in both the training and testing set.
  • matchType - String identifying the game mode that the data comes from. The standard modes are “solo”, “duo”, “squad”, “solo-fpp”, “duo-fpp”, and “squad-fpp”; other modes are from events or custom matches.
  • rankPoints - Elo-like ranking of player. This ranking is inconsistent and is being deprecated in the API’s next version, so use with caution. Value of -1 takes place of “None”.
  • revives - Number of times this player revived teammates.
  • rideDistance - Total distance traveled in vehicles measured in meters.
  • roadKills - Number of kills while in a vehicle.
  • swimDistance - Total distance traveled by swimming measured in meters.
  • teamKills - Number of times this player killed a teammate.
  • vehicleDestroys - Number of vehicles destroyed.
  • walkDistance - Total distance traveled on foot measured in meters.
  • weaponsAcquired - Number of weapons picked up.
  • winPoints - Win-based external ranking of player. (Think of this as an Elo ranking where only winning matters.) If there is a value other than -1 in rankPoints, then any 0 in winPoints should be treated as a “None”.
  • groupId - ID to identify a group within a match. If the same group of players plays in different matches, they will have a different groupId each time.
  • numGroups - Number of groups we have data for in the match.
  • maxPlace - Worst placement we have data for in the match. This may not match with numGroups, as sometimes the data skips over placements.
  • winPlacePerc - The target of prediction. This is a percentile winning placement, where 1 corresponds to 1st place, and 0 corresponds to last place in the match. It is calculated off of maxPlace, not numGroups, so it is possible to have missing chunks in a match.