Abstract:

Educational games have demonstrated their capability to capture the attention of students by providing an alternative learning vector. These games are usually presented as video games where the interaction between participants is limited. Thus, the development of traditional board games where the interactions are physical appears as a relevant solution. This fact is very relevant in collaborative games, as they are especially focused on generating an important team-building feeling. On the other hand, Data Science is a discipline that is at the crest of the wave, as most IT companies eagerly demand specialists. For this reason, we have created the DSGame kids board game to train computer science students in the Data Science world. It covers in a simplified way the lifecycle of a Data Science project, illustrating the different steps of development. These steps are four and are designed as a mission: data mission, processing mission, modeling mission y visualization mission. The first is focused on the selection of the relevant data to be analyzed. The second represents the data filtering, cleaning, and processing tasks. The third addresses the design and implementation of Machine Learning models. Finally, the last one simulates the visualization of the results and the deployment of a system able to solve the proposed problem. This problem is initially presented at the beginning of the game, and it depends on the selected scenario. Three different scenarios with heterogeneous difficulty levels have been included. Each player has a role inside the scenarios: expert in the domain (i.e., the individual who presents the problem), the computer science expert (i.e., the individual who processes the data), the math expert (i.e., the individual who creates the model) and the deployment expert (i.e., the individual who completes the deployment of the solution). Players must collaborate to solve the scenario problem by buying cards that represent a circuit in their mission spaces. These mission spaces are matrices of 16 cells where they must establish a connection between a predefined starting point and an endpoint. The game works in a cycling way as follows. Initially, four resource cards are provided to each one of the players, representing math, computer science, and domain resources. Four circuit cards are also presented in a market with specific resource costs. Then, following an established order, players can obtain the circuit cards by paying them with their resource cards (one per turn) and the cycle starts again. Players can interact between them negotiating resource cards to promote the purchase of circuit cards. Moreover, circuit cards have different and some of them provide extra points to the players. These features promote the elaboration of plans to create complex circuits. The game is completed when all the players have connected their matrix missions or no more resource cards are available. Then, if players accomplish a prefixed minimum amount of points they win the game.

Keywords:

Educational game, Data Science, Collaborative work, Storytelling, Project development.