TF-IDF, term frequency–inverse document frequency, is, explained in short, from rookie to rookie, the relation of term usage frequency in the given document to usage frequency of this term in all documents containing it.
This relation mirrors how relevant is the given document for the given term in the manifold of all documents containing this term.
TF-IDF is successor of keyword density. Some non-demented SEO geriatrics can remind, what means keyword density: number of term's usage in the text divided through the number of all words in the text multiplied with 100. This formula is in reality both most used and wrong.
Well, i'm not your Wiki, and if you're in the full legal age and not banned by Google, you would find fast everything you need to know about keyword density and TF-IDF. I just say: nowadays TF-IDF is nearly the single numeric parameter of content quality. But this article is about something other - i've promised to share some freebies, right?
SEO, know your damn tools inside out! So long as there are such nice free tools, there would be always some bucks for (better things than) buying backlinks.
I've cancelled my OnPage.org account - calculation of TF-IDF was indeed the single option, why i've used to use it. After i've become aware of these free SEO tools i share with you here, i thought i'll save better some bucks, write some quality content and earn some trustful backlinks.
Tconspectus
The first tool is built up on top of the platform https://www.pythonanywhere.com/ by a team of some nameless but extraordinarily smart russians. Wanna guess in what coding language the tool is written up? Not really - is indeed complicated, i was tipping on HTML and failed.A little bit of information about the tool you'll find on their Reddit account: https://www.reddit.com/user/t_conspectus .
Some skills of this tool are: it can deal with text input, uploaded files, and, with paid account from 5$ monthly, with urls too. The tool understands three languages: english, german and ... no, you really want too much from me - make use of your IQ and reveal this secret by your own.
Here are tool functions one by one:
Text summarization (not building an abstract, but summarization - selecting of most important things), with some statistics, like absolute term frequency and their TF-IDF. There is a nice gimmick too: before running the summarization you can adjust how long the text summarization should be, in relation to the original text, like 20%.
Text segmentation. The tool segments the text input into single sentences and filters stop words out.
Term weighting (calculation of TF-IDF) - this is the cause, why i'm loosing my time with writing this article. Its pretty easy: you input the text and the term, select one ore more weighting methods - the tool does the magic:
The fourth tool contains six text summarization methods, listed under https://github.com/miso-belica/sumy/blob/dev/README.rst. What you can do here, is to test how these methods summarize your text and compare them with the text summarization output by the own algo of the tool, which i desribed as first.
Thats all about this tool - go, use it for free and donate some beer to developers. No, was a joke: they are russians, remember? - they drink no beer, only wodka mixed up with gun powder. The tool is btw. available at http://tconspectus.pythonanywhere.com.
URLsMatch
Do you know, what is egg-laying wool-milk-sow? It is german totem animal, looking like
Well, there are some SEO tools like this animal, but all of them are to 99% not useful, because all information they deliver are valueless.
But this tool... it is a kind of egg-laying wool-milk-sow too - it delivers 38 different website and content parameters. But every single parameter has a value for you. You know why? Because this tool delivers all parameters in comparison to websites you want to compare. Take Top-1 and Top-2 competitor and in a second you know, how your website's parameters are located between competitors.
And after all the tool
- understands english, german and italian,
- lets you setup its user agent
- filters out stop words and lets you add your own
- calculates usage amount of all terms
- makes the comparative calculation of all terms
- calculates the usage frequency of all terms, in the main content area and in all website areas
- calculates TF-IDF for all competitor websites and for all terms on them,
- lets you filter the term list for the term you want.
I've seen pretty much SEO tools, but this is the single one of this kind, which gives real added value to delivered data.
Go, wonder, and use it: http://www.urlsmatch.eu/