The UBC CS Science Education Initiative (CSSEI) has resulted in a number of research projects. New teaching methods and student assessment instruments are introduced to engage student learning and evaluations of their understanding. In this paper, we report four of these recent initiatives and their initial findings.

[pt] A aplicação de princípios de design tem sido reconhecida há muito tempo como benéfica para a compreensão e manutenção em projetos de software tradicionais. Esses benefícios podem ser válidos de forma semelhante para projetos de aprendizado de máquina (ML), que envolvem experimentação iterativa com dados, modelos e algoritmos. No entanto, os componentes de ML são frequentemente desenvolvidos por cientistas de dados com diversas formações educacionais, resultando potencialmente em código que não segue as práticas recomendadas de desenvolvimento de software. Para compreender melhor esse fenômeno, investigamos o impacto dos princípios de design SOLID na compreensão do código de ML. Para tanto, conduzimos um experimento controlado com três trials independentes (replicações exatas), envolvendo no total 100 cientistas de dados. Reestruturamos o código de ML real da indústria que não usava princípios SOLID. Dentro de cada ensaio, um grupo foi apresentado ao código de ML original, enquanto o outro foi apresentado ao código de ML incorporando princípios SOLID. Os participantes de ambos os grupos foram convidados a analisar o código e preencher um questionário que incluía perguntas abertas e fechadas sobre a sua compreensão. Os resultados do estudo fornecem evidências estatisticamente significativas de que a adoção dos princípios de design SOLID pode melhorar a compreensão do código no âmbito dos projetos de ML. Propomos que os princípios de design de engenharia de software devem ser difundidos na comunidade de ciência dedados e considerados para melhorar a capacidade de manutenção do código de ML. / [en] Applying design principles has long been acknowledged as beneficial for understanding and maintainability in traditional software projects. These benefits may similarly hold for machine learning (ML) projects, which involve iterative experimentation with data, models, and algorithms. However, ML components are often developed by data scientists with diverse educational backgrounds, potentially resulting in code that doesn t adhere to software development best practices. In order to better understand this phenomenon, we investigated the impact of the SOLID design principles on ML code understanding. To this end, we conducted a controlled experiment with three independent trials (exact replications), overall involving 100 data scientists. We restructured ML code from a real industrial setting that did not use SOLID principles. Within each trial, one group was presented with the original ML code, while the other one was presented with ML code incorporating SOLID principles. Participants of both groups were asked to analyze the code and fill out a questionnaire that included both open-ended and closed-ended questions on their understanding. The study results provide statistically significant evidence that the adoption of the SOLID design principles can improve code understanding within the realm of ML projects. We put forward that software engineering design principles should be spread within the data science community and considered for enhancing the maintainability of ML code.

7. SUMMARY This diploma thesis The information value and acceptability of forecasting is divided into a theoretical and a practical part. In the theoretical part the forecasting is described as a product of television newscast. I deal with the fact of forecasting entering the media communication. I deal with the information value that forecasts bring and show the way a piece of information is transferred from a producer to an addressee relating to the process of coding and decoding. I describe the acceptability of forecasts and its elements influencing comprehension. In relation to acceptability, textual linguistics is discussed in a separated subchapter. The last theoretical part deals with semiotics and semantics. These disciplines deal mainly with meaning and sings and that is the reason why they are not important only in the theoretical part but also in an analyzing part of this thesis. The analyzing part is based on semiotic survey analysis dealing with individual components of forecasts (visual, audio and audio-visual). The aim of this thesis is to find out which of components of forecasting brings the highest information value to respondents and which is best comprehensible for addressees. The survey results demonstrate that it is the audio forecasting that has a high information value and brings...