I am making a search engine. On what basic algorithm is it better to base the output of results?

Question

Google

I don’t want to make an analogue of PR, I’m looking for alternatives to
PR (PageRank) - if it simply counts the number of snares per page, the more snares per page, the higher the page in the search results.
The question is, what do you think is better than the PR algorithm? (in the form of the main criterion for generating results)
What are the alternatives? (in the form of the main criterion for generating results)
Non-standard algorithms? (in the form of the main criterion for generating results)
My options:
1. Analyze the number of bookmarks per page in the user's browser [Chrome extension] (the more, the higher the search result)
PR is based on how many sites link to this site. The more, the higher the site in the results.
2. Parse html very hard. Down to the background color of the page.
3. Analyze browsing history in the browser [Chrome extension]
4. Analyze user actions on the page (time spent on the page, clicks, text selections
) i.e. I understand that now there are a lot of factors in ranking, well, it is the main factor that is of interest now that everywhere analogues of PR + are trying to implement machine learning.
It's clear that the analysis of the html itself will also be applied.

Answer 1

Is it too hard to analyze the payment of bills by applicants for the first places in the issue?

Answer 2

1. Analyze the number of bookmarks per page in the user's browser (the more, the higher the search result)

Please tell me how you will check it)

Answer 3

Have you done SEO?
Try to analyze how the search and issuance is going on...
One of my sites has more than 1 million links and it is mediocre in the issuance. And the other one is linked by about 1k links and it is in the top))
Different ways of promotion..

Answer 4

Face the truth.
You most likely won’t make a normal product that someone will be interested in using, the topic is too big for one person.
If you want to make an interesting project for self-development, then first read something about infosearch, for example, this classic book: "Manning, An Introduction to Information Retrieval"
www.ozon.ru/context/detail/id/5497130