Hi Hadi!
Thank you for giving it a try!
The language is calculated from the text in your captions. The system is not bulletproof, but it works well in most cases. It is true that it works best the more text it has to analyze, and in your case I saw that your captions are very short, when not just emojis, so that’s probably why it didn’t get yours quite right. Not an excuse, I’ll try to come up with a way to do it better 
As far as fake followers, it’s a delicate matter, let me reuse an explanation from a previous answer:
All the information you see right now on valiu.me is 100% real, meaning it comes directly from the Instagram API or is the result of some mathematical formula which only uses data that comes directly from the Instagram API.
Now the Instagram API des not give you a “number of fake followers”, that is simply not a metric they provide, yet so many tools out there are offering it. They way they do it is by making an educated guess about it. By analyzing large amounts of data we can figure out what a certain profile should look like based on their number of followers, niche, etc. and from there we can see if the profile we are trying to analyze is above or below average to make a guess about how many of their followers are fake (if a 100K account is getting 20 likes, there’s something weird there, right?). We can also go dipper and try to analyze the profiles that are commenting on the photos of this account (Instagram does not give us access to the list of followers) and try to guess if they are real or not and then average those results…
Anyway, without getting too geeky, the point is, the more data you analyze the more accurate y our prediction will be, but at the end of the day when it comes to fake vs. real followers you are always predicting and guessing, so there is not way to know that that metric is actually right. The same is true for many other metrics available on other platforms.
And to answer your question, I’m a very data driven person, and I’m not sure an “educated guess” is good enough for me, so I’ll have to think carefully if I want to start providing that kind of data or stick to 100% accurate data.
As for the followers/following ration that is already present in the analysis, although not with that title.
The “profile type” is calculated with the followers/following ratio, and in this case, since the main goal of the site is to find and manage influencers I was more strict than you suggested and set it up so that anyone with a ration of 1 or less is tagged as a “follower” and from there on bigger ratios get other tags. Hope this makes sense.
Anyway thank you for taking the time to try out the platform and more importantly to give me your feedback. Any other suggestions you might come up with are very welcome and I’ll try to incorporate them the best I can 
