SulmanK / PUBG-clustering-player-behavior-for-cheaters

Given a dataset from Kaggle. EDA + ML was utilized to cluster player behavior to detect cheaters. Deliverables include dashboards and a hacker detection application.

PUBG Clustering Player Behavior for Cheaters (Hackers)

Objective

• Complete Exploratory Data Analysis.
• Utilize Machine Learning to cluster.
• Deploy a dashboard to report findings. (The dashboards are offline. I have conserved costs for other vital applications!)
  • Dashboard - PART I
  • Dashboard - PART 2
  • Project Summary
• Deploy an application for detecting hackers.
  • Hacker Detection Application
    • Please be patient in loading the application.

Background Information

• Playerunknown's Battleground (PUBG) is a video game, which set the standard for preceding games in the Battle Royale genre. The main goal is to SURVIVE at all costs.

Process:

• Exploratory Data Analysis conducted utilizing various python packages (Numpy, Matplotlib, Pandas, and Plotly).
• Clustering Algorithms (Sci-Kit Learn)

Table of Contents:

• Part I: Exploratory Data Analysis

  • Jupyter Notebook
  • Dashboard -EDA PART I
  • Dashboard -EDA PART 2

• Part II: Clustering

  • Utilize various clustering algorithms for detecting cheaters or hackers.

    • Jupyter Notebook: K-means
    • Jupyter Notebook: DBSCAN
    • Jupyter Notebook: Anomaly Detection

  • Utilize PCA + previous clustering algorithms (All features).

    • Jupyter Notebook: K-means
    • Jupyter Notebook: DBSCAN
    • Jupyter Notebook: Anomaly Detection

  • Utilize PCA + previous clustering algorithms (Selected features).

    • Jupyter Notebook: K-means
    • Jupyter Notebook: DBSCAN
    • Jupyter Notebook: Anomaly Detection

  • Pertinent Deliverables

    • Project Summary
    • Hacker Detection Application

Demo

• References
  • [1] M. Breunig, H. Kriegel, R. Ng and J. Sander, "LOF", ACM SIGMOD Record, vol. 29, no. 2, pp. 93-104, 2000. Available: 10.1145/335191.335388.
  • [2] "Survey Report on K-Means Clustering Algorithm", International Journal of Modern Trends in Engineering & Research, vol. 4, no. 4, pp. 218-221, 2017. Available: 10.21884/ijmter.2017.4143.lgjzd.
  • [3] A. Drachen, "Introducing Clustering IV: The Case of Tera Online", Gamasutra.com, 2020. [Online]. Available: https://www.gamasutra.com/blogs/AndersDrachen/20140603/218817/Introducing_Clustering_IV_The_Case_of_Tera_Online.php. [Accessed: 02- Mar- 2020].
  • [4] G. News, G. Originals, G. Trailers, G. Guides, M. TV and G. REVIEWS, "Here's How Many PUBG Cheaters Have Been Banned", Game Rant, 2020. [Online]. Available: https://gamerant.com/playerunknowns-battlegrounds-cheater-ban-count/. [Accessed: 02- Mar- 2020].
  • [5] P. Rousseeuw and K. Driessen, "A Fast Algorithm for the Minimum Covariance Determinant Estimator", Technometrics, vol. 41, no. 3, pp. 212-223, 1999. Available: 10.1080/00401706.1999.10485670.
  • [6] S. Ounacer, H. Ait El Bour, Y. Oubrahim, M. Ghoumari and M. Azzouazi, "Using Isolation Forest in anomaly detection: the case of credit card transactions", Periodicals of Engineering and Natural Sciences (PEN), vol. 6, no. 2, p. 394, 2018. Available: 10.21533/pen.v6i2.533.
  • [7] A. Daffertshofer, C. Lamoth, O. Meijer and P. Beek, "PCA in studying coordination and variability: a tutorial", Clinical Biomechanics, vol. 19, no. 4, pp. 415-428, 2004. Available: 10.1016/j.clinbiomech.2004.01.005.