Hannah6879
Badges: 8
Rep:
?
#1
Report Thread starter 1 year ago
#1
I don't know what to base my project on. Any ideas? I was thinking about making a 3D first person shooter game, a bandwidth monitor or a games stats thing (records game statistics like total kills, gold earned for players) but I don't know if they will help me get top marks and they will probably be hard to make.
0
reply
Idra
Badges: 5
Rep:
?
#2
Report 1 year ago
#2
Not doing A-levels, but I might be able to provide some ideas. Machine learning always impresses examiners - if you're using python, I would look into tensorflow and try applying it to real-life scenarios, or you could even try building your own machine-learning engine (look into backpropagation if you're a mathsy person). Might not be your sort of thing though, in which case I would maybe think about a tile-based 2D game (which could probably go pretty far) or you could look into 3D graphics rendering and try to build your own game engine (Rasterization is probably the best technique for this). There's also the option of building something more immediately practical - a timetabling application could be interesting, I know that someone in my school is doing a detention scheduling app. Spreadsheet applications for student/teacher/lesson etc information, maybe?

Also, what language are you using? That's often a deciding factor because of the libraries you have available.
0
reply
Hannah6879
Badges: 8
Rep:
?
#3
Report Thread starter 1 year ago
#3
I will probably use python as that is the only language we've been taught. I heard that Monkey X is good for games so I might use that if I make a game. I looked up Tensorflow and it looks quite interesting, maybe I could do something with neural networking, deep learning (maybe a text summariser?) but I have no knowledge of these things at all so I would have to do a lot of research. Machine learning sounds good, I could make a spam filter.
I'm also interested in robotics and artificial intelligence, but I don't know what I could do with those (besides an AI game)
I thought about making an interactive PC builder, but that may be too far-fetched.
I want to do something that I will enjoy, so perhaps not something that relies heavily on databases.
Thanks a lot for the help!!
0
reply
Idra
Badges: 5
Rep:
?
#4
Report 1 year ago
#4
(Original post by HannahA2002)
I will probably use python as that is the only language we've been taught. I heard that Monkey X is good for games so I might use that if I make a game. I looked up Tensorflow and it looks quite interesting, maybe I could do something with neural networking, deep learning (maybe a text summariser?) but I have no knowledge of these things at all so I would have to do a lot of research. Machine learning sounds good, I could make a spam filter.
I'm also interested in robotics and artificial intelligence, but I don't know what I could do with those (besides an AI game)
I thought about making an interactive PC builder, but that may be too far-fetched.
I want to do something that I will enjoy, so perhaps not something that relies heavily on databases.
Thanks a lot for the help!!
If you're using python there are a metric ton of libraries that you can use. If you're making a game, I strongly recommend pygame as it has relatively easy-to-learn functions, and is probably powerful enough for anything you'd need at A-Level. However if you're interested in machine learning there are a couple of things you could do. An AI-based game is an option but probably wouldn't leverage the ability of artificial intelligent too much. Another thing is that machine learning algorithms (let's refer to them as neural networks from now on, as that's what they're typically called) have different types and are suited for different tasks.

For instance a DNN (Deep Neural Network) is suitable for classifying data and predicting results based on inputs (for example predicting a student's test outcome based on the hours they spend sleeping and revising). A CNN (Convolutional Neural Network) is great for classifying images, and is often used in those demonstrations where an AI can tell a dog apart from a cat. A modern architecture called LSTM is used for predicting results based on multiple inputs at different time steps and is used a lot in language processing such as predicting sentences and translating text (this is quite complicated and computationally intensive to train though). There are other structures as well but these are some of the most well-known ones.

What exactly you can do with them depends on how low-level you want to go - if you're interested in mathematics then it can be really interesting to build a neural network from scratch, although it takes a while to get familiar with all the concepts. This is generally a pretty impressive project, especially when you demonstrate it on real-life data. Beware though that python isn't the most optimised language, so code typically needs to run very fast and often in the case of neural networks utilise the GPU to perform the huge numbers of calculations needed for larger datasets, and this is especially frustrating when trying to work out the best way to train your network. If you want to do projects with more data involved, it would probably be more practical to use tensorflow just for the speed increase.

As for the practical applications themselves, I've seen a few projects that make use of webcams and take real-world input - for instance a neural network that can read how many fingers you're holding up. Otherwise you could train a neural network (probably a LSTM as I mentioned above, but other neural networks would also be able to do some of these competently) to auto-complete sentences, try and shorten sentences into smaller phrases, or even teach it to play a videogame. There are a few other options of course, but these are the ideas that come to mind.

If you want to try some low-level programming try watching 3Blue1Brown's neural network videos - they are a fantastic starting point and give you all the information you need to build your own neural network. The subject can be fairly daunting at first but given enough time it's possible to understand most of it. Otherwise I suggest you refer to the inner workings as a 'black box' and instead focus on how you can get the output from the input.

Sorry for the long winded answer, had a lot to say :P
0
reply
Hannah6879
Badges: 8
Rep:
?
#5
Report Thread starter 1 year ago
#5
Thank you so much for the advice!! If I were to make a game then I shall use pygame as I agree it looks relatively easy. I'll do more research on neural networking and watch the said videos, even though it does sound pretty challenging. After that I hope I'll come to a final conclusion.
To be safe, I think a game would be a good choice as I could use pygame like you said and I am also considering going into game design in the future, so this would be useful. Then again, I am also considering going into robotics and artifical intelligence, so an AI based project would also be a good idea :/
A game would be easier, but AI, machine learning would be very interesting and well appreciated by the examiners.
I'd better go do my research!
1
reply
bmcd243
Badges: 9
Rep:
?
#6
Report 10 months ago
#6
(Original post by HannahA2002)
Thank you so much for the advice!! If I were to make a game then I shall use pygame as I agree it looks relatively easy. I'll do more research on neural networking and watch the said videos, even though it does sound pretty challenging. After that I hope I'll come to a final conclusion.
To be safe, I think a game would be a good choice as I could use pygame like you said and I am also considering going into game design in the future, so this would be useful. Then again, I am also considering going into robotics and artifical intelligence, so an AI based project would also be a good idea :/
A game would be easier, but AI, machine learning would be very interesting and well appreciated by the examiners.
I'd better go do my research!
Hi, what did you end up doing...very interested in this topic as well and will be starting my NEA soon
0
reply
Hannah6879
Badges: 8
Rep:
?
#7
Report Thread starter 10 months ago
#7
I'm making a 2D fighting game (like street fighter) I'm currently on the analysis stage so I haven't actually started designing or coding it
0
reply
bmcd243
Badges: 9
Rep:
?
#8
Report 10 months ago
#8
(Original post by HannahA2002)
I'm making a 2D fighting game (like street fighter) I'm currently on the analysis stage so I haven't actually started designing or coding it
Ooooo nice, is that with pygame?
0
reply
YasiinsG
Badges: 8
Rep:
?
#9
Report 10 months ago
#9
Hi, im doing the NEA as well. We started earlier so ive already started, im creating a POS system with various other functionalities. Just a word of caution, creating games for the NEA will make it very, very difficult to achieve top marks.
If you are good with python, as suggested above, try machine learning, simulators etc. as the complexity will already be there.
0
reply
bmcd243
Badges: 9
Rep:
?
#10
Report 10 months ago
#10
(Original post by YasiinsG)
Hi, im doing the NEA as well. We started earlier so ive already started, im creating a POS system with various other functionalities. Just a word of caution, creating games for the NEA will make it very, very difficult to achieve top marks.
If you are good with python, as suggested above, try machine learning, simulators etc. as the complexity will already be there.
Thanks for the advice, if you have anymore then please send it my way as I know very little about the NEA. Yeah, I am probably going to avoid games - neural networking is a massive interest for me and definitely something that I want to research further and potentially even study at uni. I've just installed Tensorflow so will definitely have a go with that at some point - POS sounds quite original, that should do well.
0
reply
YasiinsG
Badges: 8
Rep:
?
#11
Report 10 months ago
#11
(Original post by bmcd243)
Thanks for the advice, if you have anymore then please send it my way as I know very little about the NEA. Yeah, I am probably going to avoid games - neural networking is a massive interest for me and definitely something that I want to research further and potentially even study at uni. I've just installed Tensorflow so will definitely have a go with that at some point - POS sounds quite original, that should do well.
Yh sure. These are just my views and my teachers' views, it depends on how much time and effort you are willing to put in (and how good you are at coding).
Im on the coding and testing phase currently
0
reply
Hannah6879
Badges: 8
Rep:
?
#12
Report Thread starter 10 months ago
#12
(Original post by bmcd243)
Ooooo nice, is that with pygame?
Yes
0
reply
bmcd243
Badges: 9
Rep:
?
#13
Report 10 months ago
#13
(Original post by HannahA2002)
Yes
Decent, keep me posted on how it goes
1
reply
X

Quick Reply

Attached files
Write a reply...
Reply
new posts
Back
to top
Latest
My Feed

See more of what you like on
The Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

Personalise

What support do you need with your UCAS application?

I need help researching unis (17)
16.19%
I need help researching courses (8)
7.62%
I need help with filling out the application form (6)
5.71%
I need help with my personal statement (42)
40%
I need help with understanding how to make my application stand out (25)
23.81%
I need help with something else (let us know in the thread!) (2)
1.9%
I'm feeling confident about my application and don't need any help at the moment (5)
4.76%

Watched Threads

View All