Topic for diploma (specialist) Computer science?

Question

IT education

Hello, I switched to the 3rd year, at the university they are asked to choose a topic for the diploma (yes, yes, 2 more years before passing), and therefore I am faced with a difficult choice (the breadth of my "IT view" is not very).
At the moment I'm going through the book "Algorithms in Java" (Robert Sedgwick) and after that I planned to turn my eyes to AI. Perhaps you can offer some advice in this area. I want not a banal task, but one that will captivate and, perhaps, become my future field of activity. Of the languages ​​​​in which I programmed: C ++ (a little) and C # (a year of permanent work)
PS
Yes, I know that there are a lot of such topics, but I looked at everything and could not find what I like. Thanks to all.

Answer 1

Being an appendage of your program for many years to come is not an enviable fate. It is better to choose a project that can be done in a small amount of time, even in a week, but think it over carefully until you get a clear vision of the result. (Mathematicians have this - before solving the integral, "see" the answer, and then check it).
A good program is a good interface, a good help, a good installer, and a good support. It is unlikely that all this is required from a diploma. Asking a business is a so-so idea, because. business needs solutions, not your diploma, and not all business tasks are drawn to a diploma. At work, for example, I spent two weeks writing a file server directory synchronizer (several tens of thousands of documents and this is only for one project) between the file server and Sharepoint. For my company, a very urgent task, but as a thesis - not so much. Utility and more.
But on the weekend I came across an interesting problem - to take values ​​from paper graphs, nomograms (empirical graphs). It is necessary to take values ​​from them according to the given parameters. Digitization does not roll, It was necessary to write a graphic editor so that the drawing could be scaled (I took a leaflet), added a polyline editor (leaflet.editor) to it to outline the lines of the nomogram and wrote a small mater to complete the graphs by intermediate values ​​(I applied morphing) . It turned out to be a good utility for old calculation manuals, so as not to use a ruler when you take readings both from given lines and from intermediate ones. And shoots quite accurately. Well, here's a way to bring scientific legacy data into the present day. Well, such a task could well go for a diploma, or at least for a blank.
If you're interested, I can give you a list of some projects at my job, in case you have another idea for your diploma.

Answer 2

What topic of the diploma on ... to choose?
You go to a business - you say that you want to work for free and do everything he needs
. If you haven’t found a business - you go to a teacher, he finds you a task or a business that has tasks.
If not, you go to Google and take any task
https: //www.reddit.com/r/AppIdeas/

Answer 3

AI will not reach you soon, but moving to the cloud is already a very popular topic - first-hand info and from direct developers for the cloud