Profiling Players Using Real-World Datasets
Computer games provide an ideal test bed to collect and study data related to human behavior using a virtual environment having real-world-like features. Studies regarding individual players’ actions in a gaming session and how this correlates with their real-life personality have the potential to reveal great insights in the field of affective computing. This project profiles players using data collected from strategy games. This is done by taking into account the gameplay and the associations between the personality traits and the subjects playing the game. This study uses two benchmark strategy game datasets, namely, StarCraft and World of Warcraft. In addition, the study also uses the Age of Empire-II game data, collected using 50 participants. The IPIP-NEO-120 personality test is conducted using these participants to evaluate them on the Big-Five personality traits. The three datasets are profiled using four clustering techniques. The results identify two clusters in each of these datasets. The quality of cluster formation is also evaluated through the cluster evaluation indices. Using the clustering results, the classifiers are then trained to classify a player, after a gameplay, into one of the two profiles. Results show that the gameplay can be used to predict various personality features using strategy game data.
Research in the area of human behavior pertaining to human emotion recognition and other related affective phenomena is gaining momentum. Profiling of humans is a computational technique used to categorize individuals with common behavior in the same group and those having diverse traits in different groups. Personality tests are an established mean of behavioral research in many fields. They are also used in scientific settings to support analytic decisions. Minnesota Multiphasic Personality Inventory (MMPI) and the Eysenck Personality Questionnaire (EPQ) are the pioneer tools to study personality. The study of human personality finds its utility in a diverse set of fields including, but not limited to, employment test, individual and relationship counseling, trauma treatment and career counseling. However, there are two major criticism for deploying personality tests: criterion-related validity of personality measure and faking. Criterion-related validity of personality measures suggests that often personality tests are deployed to either measure the wrong outcome or the correlations developed are flawed. Personality tests are also susceptible of being faked by respondents, because more than often the desirable answer is apparent and obvious. Personality tests mainly rely on the honesty of the respondent and her/his self-awareness. Personality tests are conducted under controlled conditions similar to a laboratory experiment. This has the disadvantage of not providing the subjects an environment where they could manipulate the environment’s variables and based on this, make a choice using their preferences. Subjects are asked to imagine a scenario, and based on the hypothetical situation or a previous similar experience, they opt for an answer. All this adds to the imprecision that these direct personality tests may have. There are many indirect personality assessment methods but they also comprise of questions. However, there are few other factors that make these personality tests useful, such as the maturity of various models over the period of decades, time bounded, and convenience.
It will be interesting to see how a group of subjects is classified in one of the personality profiles using their actions in the real-world. This will require long durations for observation and making an assessment. For convenience, in such experiments a virtual environment may be useful, however, this will have the same disadvantage mentioned earlier, i.e., the subject being aware of the experiment. Computer games provide a rich mapping of the real-world scenarios to their virtual equivalent. This is evident from the fact that these have been used for in an assortment of fields to gain better performance. One may be able to predict player’s personality from her/his style of playing the game. Subjects may be asked to play RTS gaming session and their actions be recorded during this time. Later, these recorded features can be mapped to specific personality factors, ultimately classifying the subjects in various personality profiles. This will help in a real assessment of the subject’s personality using a test bed that keeps the system peculiarities hidden from the subjects and allows them to exhibit their true behavior in real-world-like situations. However, mapping of the features recorded during the gaming session to the various personality traits will need an in-depth study with logical reasoning. This may be done by adding customized features to the RTS game.