Hadad Shahrazad, Dinu Mihai, Bumbac Robert, Iorgulescu Maria-Cristina, Cantaragiu Ramona
UNESCO Dept. for Business Administration, The Bucharest University of Economic Studies, Calea Grivitei 2-2A, 010731 Bucharest, Romania.
Entropy (Basel). 2020 Aug 21;22(9):918. doi: 10.3390/e22090918.
The paper addresses the dynamics of education by using Markov chains, a powerful probabilistic model able to make predictions on how sources of knowledge either change or stabilize over adulthood. To this end, each student filled in a survey that rated, on a scale from 1 to 5, the utility of five different sources of knowledge. They completed this survey twice, once for their previous and once for their current education. The authors then fitted a Markov chain to these data-essentially, calculating transition probabilities from one ranking of sources of knowledge to another-and inferred the final maximum utility sources of knowledge via the stationary distribution. The overall conclusion is the following: even if the professor used to play a crucial role in early development, students have the tendency to become independent in their learning process, relying more on online materials and less on printed books and libraries.
本文通过使用马尔可夫链来探讨教育的动态变化,马尔可夫链是一种强大的概率模型,能够预测知识来源在成年期如何变化或稳定。为此,每个学生填写了一份调查问卷,该问卷在1到5的量表上对五种不同知识来源的效用进行评分。他们两次完成了这项调查,一次针对之前的教育,一次针对当前的教育。然后,作者将马尔可夫链应用于这些数据——本质上是计算从一种知识来源排名到另一种知识来源排名的转移概率——并通过平稳分布推断出最终的最大效用知识来源。总体结论如下:即使教授在早期发展中曾经发挥关键作用,但学生在学习过程中倾向于变得独立,更多地依赖在线材料,而较少依赖印刷书籍和图书馆。