Demographics and Personality Discovery on Social Media: A Machine Learning Approach

This research proposes a new feature extraction algorithm using aggregated user engagements on social media in order to achieve demographics and personality discovery tasks. Our proposed framework can discover seven essential attributes, including gender identity, age group, residential area, education level, political affiliation, religious belief, and personality type. Multiple feature sets are developed, including comment text, community activity, and hybrid features. Various machine learning algorithms are explored, such as support vector machines, random forest, multi-layer perceptron, and naïve Bayes. An empirical analysis is performed on various aspects, including correctness, robustness, training time, and the class imbalance problem. We obtained the highest prediction performance by using our proposed feature extraction algorithm. The result on personality type prediction was 87.18%. For the demographic attribute prediction task, our feature sets also outperformed the baseline at 98.1% for residential area, 94.7% for education level, 92.1% for gender identity, 91.5% for political affiliation, 60.6% for religious belief, and 52.0% for the age group. Moreover, this paper provides the guideline for the choice of classifiers with appropriate feature sets.

Sex differences in personality scores on six scales: Many significant, but mostly small, differences

This study examined sex differences in domain and facet scores from six personality tests in various large adult samples. The aim was to document differences in large adult groups which might contribute new data to this highly contentious area. We reported on sex differences on the Myers-Briggs Type Indicator (MBTI); the Five Factor NEO-PI-R; the Hogan Personality Indicator (HPI); the Motives and Values Preferences Indicator (MVPI); the Hogan Development Survey (HDS) and the High Potential Trait Indicator (HPTI). Using multivariate ANOVAs we found that whilst there were many significant differences on these scores, which replicated other studies, the Cohen’s d statistic showed very few (3 out of 130) differences >.50. Results from each test were compared and contrasted, particularly where they are measuring the same trait construct. Implications and limitations for researchers interested in assessment and selection are discussed.

Machine Learning Approach to Personality Type Prediction Based on the Myers–Briggs Type Indicator®

Neuro Linguistic Programming (NLP) is a collection of techniques for personality development. Meta programmes, which are habitual ways of inputting, sorting and filtering the information found in the world around us, are a vital factor in NLP. Differences in meta programmes result in significant differences in behaviour from one person to another. Personality types can be recognized through utilizing and analysing meta programmes. There are different methods to predict personality types based on meta programmes. The Myers–Briggs Type Indicator® (MBTI) is currently considered as one of the most popular and reliable methods. In this study, a new machine learning method has been developed for personality type prediction based on the MBTI. The performance of the new methodology presented in this study has been compared to other existing methods and the results show better accuracy and reliability. The results of this study can assist NLP practitioners and psychologists in regards to identification of personality types and associated cognitive processes.

Personality Assessment for Employee Development: Ivory Tower or Real World?

The acceptance and popularity of personality assessments in organizational contexts has grown enormously over the last 40 years. Although these are used across many applications, such as executive coaching, team building, and hiring and promotion decisions, the focus of most published research on the use of personality assessments at work is biased toward assessment for employee selection. Reviews have therefore tended to use criteria that are appropriate for selection, neglecting the additional and different criteria that are important in relation to employee development. An illustration of the often-discussed scientist-practitioner divide is that the Myers-Briggs Type Indicator is the most widely known and used personality assessment in organizations, despite harsh criticism by the academic community. This article reviews this debate, and draws implications for the appropriate choice of personality assessments for use in individual and team development, and a new direction for scientific research.

The Prevalence of Pseudoscientific Ideas and Neuromyths Among Sports Coaches

There has been an exponential growth in research examining the neurological basis of human cognition and learning. Little is known, however, about the extent to which sports coaches are aware of these advances. Consequently, the aim of the present study was to examine the prevalence of pseudoscientific ideas among British and Irish sports coaches. In total, 545 coaches from the United Kingdom and Ireland completed a measure that included questions about how evidence-based theories of the brain might enhance coaching and learning, how they were exposed to these different theories, and their awareness of neuromyths. Results revealed that the coaches believed that an enhanced understanding of the brain helped with their planning and delivery of sports sessions. Goal-setting was the most frequently used strategy. Interestingly, 41.6% of the coaches agreed with statements that promoted neuromyths. The most prevalent neuromyth was "individuals learn better when they receive information in their preferred learning style (e.g., auditory, visual, or kinesthetic)," which 62% of coaches believed. It is apparent that a relatively large percentage of coaches base aspects of their coaching practice on neuromyths and other pseudoscientific ideas. Strategies for addressing this situation are briefly discussed and include changing the content of coach education programs.

Exploring the use of online video games to detect personality dichotomies

Purpose

Personality trait detection is a problem that has been gaining much attention in the computer science field recently. By leveraging users’ personality knowledge software applications are able to adapt their behaviour accordingly. To detect personality traits automatically users must substantially interact with software applications to gather enough information that describe their behaviour. For addressing this limitation, the authors explore the use of online video games as an alternative approach to detect personality dichotomies. The paper aims to discuss these issues.

Design/methodology/approach

The authors analyse the use of several online video games that exhibit features related with Myers-Briggs sensitive-intuitive personality dichotomy. Then, the authors build a user profile that describes users’ behaviour when interacting with online video games. Finally, the authors identify users’ personality by analysing their profile with different classification algorithms.

Findings

The results show that games that obtained better results in the personality dichotomy detection exhibit features that had better match with the sensitive-intuitive dichotomy preferences. Moreover, the results show that the classification algorithms should satisfactorily deal with unbalanced data sets, since it is natural that the frequencies of the dichotomies types are unbalanced. In addition, in the context of personality trait detection, online video games possess several advantages over other type of software applications. By using games, users do not need to have previous experience, since they learn how to play during gameplay. Furthermore, the information and time needed to predict the sensitive-intuitive dichotomy using games is little.

Originality/value

This study shows that online video games are a promising environment in which the users’ personality dichotomies can be detected.