Described as:
Symbolab is a search engine for students, mathematicians, scientists and anyone else looking for answers in the mathematical and scientific realm. Other search engines that do equation search use LaTex, the document mark up language for mathematical symbols which is the same as keywords, which unfortunately gives poor results.
Symbolab uses proprietary machine learning algorithms to provide the most relevant search results that are theoretically and semantically similar, rather than visually similar. In other words, it does a semantic search, understanding the query behind the symbols, to get results.
The nice thing about math and science is that it’s universal – there’s no need for translation in order to understand an equation. This means scale can come much quicker than other search engines that are limited by language.
From: The guys from The Big Bang Theory will love mathematical search engine Symbolab by Shira Abel. (includes an interview with Michael Avny, the CEO of Symbolab.
Limited to web content at the moment but a “scholar” option is in the works. I assume that will extend into academic journals.
Focused now on mathematics, physics and chemistry, but in principle should be extensible to related areas. I am particularly anxious to hear they are indexing CS publications!
Would be really nice if Springer, Elsevier, the AMS and others would permit indexing of their equations.
That presumes publishers would realize that shutting out users not at institutions is a bad marketing plan. With a marginal delivery cost of near zero and sunk costs from publication already fixed, every user a publisher gains at $200/year for their entire collection is $200 they did not have before.
Not to mention the citation and use of their publication, which just drives more people to publish there. A virtuous circle if you will.
The only concern I have is the comment:
The nice thing about math and science is that it’s universal – there’s no need for translation in order to understand an equation.
Which is directly contrary to what Michael is quoted as saying in the interview:
You say “Each symbol can mean different things within and across disciplines, order and position of elements matter, priority of features, etc.” Can you give an example of this?
The authors of the Foundations of Rule Learning spent five years attempting to reconcile notations used in rule making. Some symbols had different meanings. They resorted to inventing yet another notation as a solution.
Why the popular press perpetuates the myth of a universal language isn’t clear.
It isn’t useful and in some cases, such as national security, it leads to waste of time and resources on attempts to invent a universal language.
The phrase “myth of a universal language” should be a clue. Universal languages don’t exist. They are myths, by definition.
Anyone who says differently is trying to sell you something, Something that is in their interest and perhaps not yours.
I first saw this at Introducing Symbolab: Search for Equations by Angela Guess.