More confidently uncertain? Teaching learners to apply Bayesian methods to make sense of scientific phenomena

Rosenberg, Joshua M.

doi:10.35542/osf.io/7rptw

Cited by 2 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One is the one presented by the work of Joshua M. Rosenberg (2020), where the question is to treat the problem directly by the students and skills for it. According to this approach, the aim is for students to develop skills to establish models and patterns in science, of a scientific nature, from large masses of data and, obviously, from Bayes' Theorem.…”

Section: Conclusion For Practicementioning

confidence: 99%

The bayesian thinking, a pervasive computational thinking

Ros¹

2021

Preprint

View full text Add to dashboard Cite

In its simplest sense, computational thinking is considered as a series of specific skills that help programmers to do their homework, but that are also useful to people in their professional life and in their personal life as a way to organize the resolution on their problems, and of representing the reality that surrounds them.In a more elaborate scheme, this complex of skills constitutes a new literacy --- or the most substantial part of it --- and an inculturation to deal with a new culture, the digital culture in the knowledge society.We have seen how Bayesian probability is used in epidemiology models to determine models for the evolution of data on contagion and deaths in COVID and in natural language processing.We could also see it in a multitude of cases in the most varied scientific and process analysis fields. In this way, with the automation of Bayesian methods and the use of probabilistic graphical models, it is possible to identify patterns and anomalies in voluminous data sets in fields as diverse as linguistic corpus, astronomical maps, add functionalities to the practice of the magnetic resonance imaging, or to card, online or smartphone purchasing habits. In this new way of proceeding, big data analysis and Bayesian theory are associated.If we consider that Bayesian thinking, this way of proceeding, as one more and more relevant element of computational thinking, then to what has been said on previous occasions we must now add the idea of generalized computational thinking, which goes beyond education. It is no longer about aspects purely associated with ordinary professional or vital practice to deal with life and the world of work, as has been what we have called computational thinking until now, but as a preparation for basic research and research methodology in almost all disciplines. Because, thus defined, computational thinking is influencing research in almost all areas, both in the sciences and in the humanities. An instruction focused on this component of computational thinking, Bayesian thinking, of including it at an early stage, in Secondary (K-12), including the inverse probability formula, would allow, based on Merrill’s First principles of learning, and in particular in the activation principle, activate these learning as very valuable and very complex components in a later stage of professional or research activity, or in the training passed, undergraduate and postgraduate degrees, of these professions or that train for these activities and professions.

show abstract

Section: Conclusion For Practicementioning

confidence: 99%

The bayesian thinking, a pervasive computational thinking

Ros¹

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Una es la que nos plantea el trabajo de Joshua M. Rosenberg (2020), donde la cuestión es abordar el problema directamente por los alumnos y crearles habilidades para ello. Según este planteamiento, se trata de que los estudiantes desarrollen habilidades para establecer modelos y patrones en la ciencia, de carácter científico, a partir de grandes masas de datos y, obviamente, a partir del Teorema de Bayes.…”

Section: Conclusiones Para La Prácticaunclassified

El pensamiento bayesiano, un pensamiento computacional omnipresente

Ros¹,

Palacios²

2021

Red

View full text Add to dashboard Cite

En su acepción más sencilla se considera el pensamiento computacional como una serie de habilidades específicas que sirven a los programadores para hacer su tarea, pero que también son útiles a la gente en su vida profesional y en su vida personal como una forma de organizar la resolución de sus problemas, y de representar la realidad que hay en torno a ellos. En un esquema más elaborado este complejo de habilidades constituye una nueva alfabetización ---o la parte más sustancial de ella--- y una inculturación para manejarse en una nueva cultura, la cultura digital en la sociedad del conocimiento. Hemos visto cómo se usa la probabilidad bayesiana, en modelos de epidemiología, para determinar modelos de evolución de datos sobre contagio y fallecimientos en el COVID y en el procesamiento del lenguaje natural. Igualmente podríamos verlo en multitud de casos en los más variados campos científicos y de análisis de procesos. De esta forma, con la automatización de los métodos bayesianos y el uso de modelos gráficos probabilísticos es posible identificar patrones y anomalías en voluminosos conjuntos de datos en campos tan diversos como son los corpus lingüísticos, los mapas astronómicos, añadir funcionalidades a la práctica de la resonancia magnética, o a los hábitos de compra con tarjeta, online o smartphones. En esta nueva forma de proceder, se asocian el análisis de grandes datos y la teoría bayesiana. Si consideramos que el pensamiento bayesiano, esta forma de proceder, como un elemento más y relevante del pensamiento computacional, entonces a lo dicho en anteriores ocasiones hay que añadir ahora la idea de pensamiento computacional generalizado , que va más allá de la educación Ya no se trata de aspectos puramente asociados a la práctica profesional o vital ordinaria para manejarse por la vida y el mundo del trabajo, como ha sido lo que hemos llamado pensamiento computacional hasta ahora, sino como una preparación para la investigación básica y para metodología investigadora en casi todas las disciplinas. Porque, así definido, el pensamiento computacional está influyendo en la investigación en casi todas las áreas, tanto en las ciencias como en las humanidades. Una instrucción centrada en esta componente de pensamiento computacional, el pensamiento bayesiano, o que lo incluyese en una fase temprana, en Secundaria (K-12), incluyendo la fórmula de la probabilidad inversa, permitiría, basándonos en los First principles of learning de Merrill, y en particular en el principio de activación, activar estos aprendizajes como componentes muy valiosos y muy complejos en una etapa posterior de la actividad profesional o investigadora, o en la fase de formación, grados y postgrados, de estas profesiones o que capacitan para estas actividades y profesiones. In its simplest sense, computational thinking is considered as a series of specific skills that programmers use to do their homework. They are also useful to people in their professional and personal lives, as a way of organizing the resolution of their problems, and of representing the reality that surrounds them. In a more elaborate scheme, this complex of skills constitutes a new literacy --- or the most substantial part of it --- and an inculturation to deal with a new culture: digital culture in the knowledge society. We have seen how Bayesian Probability is used in epidemiology models to determine models of data evolution on contagion and deaths in COVID. We have also seen it in natural language processing. We could also see it in many cases in the most varied scientific and process analysis fields. In this way, with the automation of Bayesian methods and the use of probabilistic graphical models, it is possible to identify patterns and anomalies in voluminous data sets in fields. Fields as diverse as linguistic corpus, astronomical maps, adding functionalities to the practice of magnetic resonance imaging, or to the habits of buying with cards, online or smartphones. In this new way of proceeding, big data analysis and Bayesian theory are associated.. If we consider that Bayesian thinking (this way of proceeding) as one more element of computational thinking, then, to what has been said on previous occasions, we must now add the idea of generalized computational thinking, which goes beyond education It is no longer about aspects purely associated with professional or vital practice to deal with life and the world of work, but as a preparation for basic research and for a research methodology in almost all disciplines. Thus defined, computational thinking is influencing research in almost all areas, both in the sciences and the humanities. An instruction focused on this component of computational thinking, or including it, at an early stage, in Secondary, would allow to activate these learnings as very valuable and very complex components at a later stage. In professional or research activity, in the training phase, undergraduate and postgraduate degrees, of these professions. Those that train for these activities and professions.

show abstract

More confidently uncertain? Teaching learners to apply Bayesian methods to make sense of scientific phenomena

Cited by 2 publications

References 6 publications

The bayesian thinking, a pervasive computational thinking

The bayesian thinking, a pervasive computational thinking

El pensamiento bayesiano, un pensamiento computacional omnipresente

Contact Info

Product

Resources

About