3 ms·
I'm not sure the "hard way" is the only way that works. If you look at a new field, you can often do some meta-learning that will help you. For instance, you
by lordnacho 10y ago
I'm not sure the "hard way" is the only way that works.
If you look at a new field, you can often do some meta-learning that will help you.
For instance, you can browse the introductory texts for common words. Words that are uncommon in normal English (ie not "the" or unspecific verbs like "do" or "have") will appear frequently. You'll quickly discover the subtopics this way. For instance, you will find the word "inflation" in an econ text and conclude it's something to be learned about.
The organization of the text is also a clue. You can scan headings and suppose that whatever terms are there are considered important.
Once you've done a scan you have some priors that hopefully will put your mind's model nearer to what the experts have, and you can start refining by actually reading.
Basically, apply some ML-like ideas to what you're looking at (clustering, correlations, and so on).