The objective of this project is to identify groups of game players with distinct physiological responses to game events and with analogous in-game behaviors. This will be achieved by analyzing data already collected from players during gameplay. The data includes various sensor data such as Heart rate, Respiration, EDA, and ECG along with in-game parameters and social setting of participants. Using this data, the task is to investigate if it is possible to divide the players into meaningful groups according to how they respond to the various game events. Using statistical Analysis the existence of these groups was verified. The clustering pipeline extracts players characteristics by training a model to predict the game events using the physiological responses of the player. The model parameters are then used as a representation of the player, these model parameters were used for clustering players further in the pipeline along with player survey data. Many challenges were presented while clustering due to the nature of the in-game events and the physiological responses associated with them. We discussed the possible interpretations and conducted tests on simulated data to check the validity of the pipeline. Tests were conducted on generated data that represented ideal physiological reactions to in-game events. The tests showed better results with the generated data proving the validity of the pipeline.