Youtube algorithms allow users to browse related videos. Recommendations are based on what other viewers, who have also watched the same videos, have browsed. This nifty feature has become an essential part of Youtube, where 300 hours of video are uploaded every minute. By clicking on a recommended video, the Youtuber can reach new content with expectations that it will fit one's taste.
But what if a Youtuber wanted to learn something new? What if a Youtuber who only watches NBA highlights has an undiscovered desire to watch videos of a cobra fighting a mongoose? Though the sidebar recommendations are great at featuring interesting, related content, there is no way for interesting, unrelated content to show up. And while the Cobra v. Mongoose example is facetious and trivial to some people, the idea of unrelated recommendations can be applied in education.
Picture a curious student who wants to learn about the world. One problem that the curious student faces is that he may not even know what he wants to know. He may have found his classes intellectually stimulating, but he does not know how to proceed once the curriculum ends with graduation. The student may have enjoyed studying math in high school, but beyond Calculus, the linear Mathematics curriculum branches off into Linear Algebra, Real Analysis, Number Theory, Combinatorics, Differential Equations, and Set Theory. In short, there are too many classes and not enough time.
We end up with a Catch-22-esque dilemma, for the curious want to start learning but do not know where to start. They set out to find out more about themselves, which includes finding out their topics of interest. Yet to find out about the topics, one must first know about the topics, which takes time. And time is something that students, especially college students, do not have.
I suspect that this problem plays a big role in students who waste time by changing their majors. (La Verne claims that "50% – 70% of students change their majors at least once, [and] most will change majors at least 3 times before they graduate.") Unfortunately, no big data algorithm will be able to recommend subjects, books, hobbies, or videos without referring to what the Youtuber has sought in the past.
The solution lies in non-artificial intelligence. Teachers are able to look at what the student has, and determine what the student needs. They can recommend a great film to someone who has only been watching cartoons, and an enlightening novel to someone who has only been reading non-fiction.
In an era when the teaching role is being taken over by educational videos, the teacher must take on a greater role of taking responsibility in balancing the student's life. This is something that machines cannot replace, for machines cannot make value-related decisions. Machines, seeking to maximize the number of views over the overall well being of students, are not going to tell the student: "Stop watching history videos, for you have spent too much time in front of a computer. You should head outside and train your physical body, preferably in a team sport so that you could develop social skills and teamwork at the same time." Teachers will.
But what if a Youtuber wanted to learn something new? What if a Youtuber who only watches NBA highlights has an undiscovered desire to watch videos of a cobra fighting a mongoose? Though the sidebar recommendations are great at featuring interesting, related content, there is no way for interesting, unrelated content to show up. And while the Cobra v. Mongoose example is facetious and trivial to some people, the idea of unrelated recommendations can be applied in education.
Picture a curious student who wants to learn about the world. One problem that the curious student faces is that he may not even know what he wants to know. He may have found his classes intellectually stimulating, but he does not know how to proceed once the curriculum ends with graduation. The student may have enjoyed studying math in high school, but beyond Calculus, the linear Mathematics curriculum branches off into Linear Algebra, Real Analysis, Number Theory, Combinatorics, Differential Equations, and Set Theory. In short, there are too many classes and not enough time.
We end up with a Catch-22-esque dilemma, for the curious want to start learning but do not know where to start. They set out to find out more about themselves, which includes finding out their topics of interest. Yet to find out about the topics, one must first know about the topics, which takes time. And time is something that students, especially college students, do not have.
I suspect that this problem plays a big role in students who waste time by changing their majors. (La Verne claims that "50% – 70% of students change their majors at least once, [and] most will change majors at least 3 times before they graduate.") Unfortunately, no big data algorithm will be able to recommend subjects, books, hobbies, or videos without referring to what the Youtuber has sought in the past.
The solution lies in non-artificial intelligence. Teachers are able to look at what the student has, and determine what the student needs. They can recommend a great film to someone who has only been watching cartoons, and an enlightening novel to someone who has only been reading non-fiction.
In an era when the teaching role is being taken over by educational videos, the teacher must take on a greater role of taking responsibility in balancing the student's life. This is something that machines cannot replace, for machines cannot make value-related decisions. Machines, seeking to maximize the number of views over the overall well being of students, are not going to tell the student: "Stop watching history videos, for you have spent too much time in front of a computer. You should head outside and train your physical body, preferably in a team sport so that you could develop social skills and teamwork at the same time." Teachers will.
Comments
Post a Comment