Algorithm: There are a lot of books in the library that I have not read, and I would like to use some method to deduce the hypothetical sort results---suppose I read all the books.
There are a lot of books in the library that I haven't read, and I want to sort out the hypothetical results in some way---suppose I've read all the books (and of course I've got them). So I can choose a book that might be better for me based on the sort results.
I have a vision, but I do not know reasonable unreasonable, how to achieve. If I could find a "different" person (supposed to be the same as my hobbies, tastes, levels, etc.), in short, we both scored the same for any one book, enough of the "different" books I've read
Should be able to include all the books. That is to say that I have read all the books and have scored, then sort the scores to get the desired results.
Of course this is a bit idealistic in the case of not restricting the scope of the book, because there may not be enough "different" for me. For example: Want to find with me in basketball, football, rugby .... People of the same preference (who have the same sort of preference for these teams) can be difficult to find, and it's much easier to get the same people as I do in basketball. Therefore, it is necessary to limit the scope of the book before calculating the hypothetical order.
But how does this idea come true? Or who can provide other ideas and implementations that meet the needs? algorithm???? Sort
Simply put, your idea is to partially deduce the whole, both reliable and not reliable
It means that a lot of people are doing research in this area, which means that you're not wrong about your idea.
The non-reliable refers to the current lack of authoritative papers (multi-factor analysis, single factor analysis)
There are many mathematical models, such as the nervous system, logistic regression, progressive chromatography ... And so on, but it's all a state of contention
My first job was to do something like that--market research.
The development direction of new products (i.e., unknown products) according to the preference status of sample population feedback (product attributes)
If it's a single-factor linear infinity, it's usually in its direction to find a development goal, such as what you say about a sport that you prefer not to like best, that you can find like or dislike in your direction.
The above is just the emotional discussion, to specific ideas I am afraid to find the literature, to establish a mathematical model