The paper deals with the topic of the further education of employees in the industrial sector. Even though Europe today has one of its most highly qualified cohorts of young graduates entering the labour market, approximately a quarter of European Union employees have significant professional skill deficits, their professional skills are much lower compared to those needed by an average worker to be fully proficient in their job. Companies must have employees who are able to quickly adapt to an ever-changing world market. That is why they need to invest in on-going employee training and development in order to both keep their employees and be successful. Academics from Constantine the Philosopher University in Nitra have a long-term cooperation with the EATON company, a global leader in electrotechnical technologies. Within the framework of this cooperation, the university was asked by this company to assess courses which it offered to its own employees as well as employees of other companies from industrial production practices, and if appropriate to suggest an improved model of the tailor-made courses provided. The authors present a case study, the aim of which was to evaluate the effectiveness and the meeting of the objectives of the training carried out for the particular target groups of participants. They describe the methodology of the assessment carried out and the main findings resulting from it.
Modeling players in digital games increasingly relies on rich game telemetry, but many studies still provide one-off, game-specific procedures that are difficult to reuse. This paper presents a typologically anchored, game event log-based framework for inferring interpretable and stable player personas from in-game behavior without relying on questionnaires. We use telemetry from a first-person shooter game designed explicitly as a research testbed, and reconstruct player sessions into a set of player-level metrics that capture pacing, exploration, combat performance, risk exposure, and resource management. These metrics are organized into blocks inspired by Bartle’s typology (Achiever, Explorer, Killer) to maintain a clear conceptual link between low-level events and higher-level motivational tendencies. After correlation screening and event log transformation, we apply principal component analysis to obtain an orthogonal latent space and perform k-means clustering. The stability of the clusters is assessed using a core-clustering procedure based on pruning and bootstrap resampling. The resulting personas exhibit distinct and interpretable behavioral profiles, including aggressive, high-risk players, cautious survivors, and exploration-oriented tacticians. We discuss how this methodology can be transferred to other games and how the identified personas can influence adaptive design and training scenarios for serious games.