Buscador

Datasets used for the empirical evaluation of the privacy-preserving data mashup protocol. Three data sets can be found: Vertical partition of educational provider A (Input_DataPartition_A.csv) Vertical partition of educational provider B (Input_DataPartition_B.csv) Integrated and k-anonymized output dataset by the privacy-preserving data mashup protocol (Output_AnoymizedIntegratedDataset.csv)

Source code is available here: http://drpantera.github.io/gsiege, Data obtained in 10 leagues (different random seeds) of Gades Siege with 2 cohorts of students It suppors the results reported in the paper "Assessment in Software Development for Competitive Environments: an AI strategy development case study"

Serious games are games with an educational purpose. In these games, players develop their skills by facing a number of challenges, and students are assessed according to their game playing behaviour. Assessment of serious game-based learning experiences has to take into account diverse features as game genre, pedagogical aim or game context. This paper analyses how skills are usually assessed in learning experiences based on serious games. To reach this objective, a systematic mapping study of more than 400 papers is undertaken. Papers were identified and classified according to a framework based on four categories: assessment aim, implementation, integration and primary assessment type. The reviewed literature mainly deals with contributions on methods and approaches for serious games. Results have revealed that most assessment methods are applied for a formative purpose more than for a certification purpose. Most frequent implementations such as game scoring and integrations like monitoring states were also uncovered. The main primary type of assessment detected was in-process. In addition, several limitations were found in the assessment methods: regarding the aim of assessment, certification of previous or attained skills was usually implemented out of the game; the scope of some implementations was limited because results were predefined earlier; and most of methods analysed present scalability issues because they rely on manual assessments. Such findings are analysed and discussed to clarify the state of the art and provide recommendations for further work in the area of serious games-based learning