fixed a typo

README.md

@@ -10,7 +10,7 @@ Lauinger et al. showed that the number of tested websites using a JavaScript lib
 
 Therefore, I focused on reproducing the number of websites with at least one known JavaScript library vulnerability. Lauinger et al. used static and dynamic detection of the libraries. Static detection consists of checking the hash value of a library file and dynamic detection uses an extractor function which is evaluated in the context of the website. I use dynamic detection because it guaranties that a library is loaded correctly, which is not the case if you just check for a hash value of a file. Another reason why I did not use static detection is that I did not find a large source of vulnerable JavaScript library versions represented as hash values. 
 
-Furthermore, there is the blog post [77% of sites use at least one vulnerable JavaScript library](https://snyk.io/blog/77-percent-of-sites-use-vulnerable-js-libraries/) from [snyk](https://snyk.io) which claims that __ 76.6%__ of the Top 5000 Alexa websites use vulnerable JavaScript library versions. They did not release raw data, so it is hard to verify their rather high number. For crawling they used [WebPageTest](https://www.webpagetest.org) and for the source of vulnerable versions they used their own database.
+Furthermore, there is the blog post [77% of sites use at least one vulnerable JavaScript library](https://snyk.io/blog/77-percent-of-sites-use-vulnerable-js-libraries/) from [snyk](https://snyk.io) which claims that __76.6%__ of the Top 5000 Alexa websites use vulnerable JavaScript library versions. They did not release raw data, so it is hard to verify their rather high number. For crawling they used [WebPageTest](https://www.webpagetest.org) and for the source of vulnerable versions they used their own database.
 
 ### Hardware/Software