Machine Learning Projects for Beginners (Datasets Included)

Key Takeaways

hello everybody and welcome back so in this video what I'm redoing is sharing with you some beginner machine learning projects and explaining to you how to go about solving these projects as well as including the datasets that are with them so for all these projects I have datasets linked down below so you don't have to go searching for the datasets you can just download them use them in practice right away and that's the whole point in this video is to make sure everything is kind of here for you guys and that you can get everything right from this video you don't feel searching for how to find this information in this data which typically is the hardest part when you're trying to practice machine learning so with that being said let's get into the video and talk about some machine learning projects for beginners before we get started I need to thank simply learn for sponsoring this video and providing all of you with 10% off their Python data science course this course was co-developed by IBM and contains 68 hours of high quality content teaching you modules like numpy Syfy pandas scikit-learn and matplotlib you'll learn the essential concepts of Python programming and gain deep knowledge in data analytics machine learning data visualization web scraping and natural language processing you'll apply these skills and for real-life industry-based projects and you'll be eligible to receive an IBM simply learn joint certificate after completing 85% of the course you can use the discount code tech with tim and sign up at the link below alright so the first project I recommend is using linear regression this is a basic algorithm if you don't know this and you can save yourself a beginner machine learning user or whatever it is learn this algorithm it's very straightforward and it's very powerful for doing things like I'm about to talk about here so I have a data set down below that includes information about students and their grades and what I'm tasking you guys with as a project idea here is using that data set try to predict a student's final grade now it has information like absences amount of hour studied I think it's like time watching TV siblings a bunch of information in there and what you're gonna have to do is kind of look through this data set and first determine what information do I need so what is actually gonna make sense for me to predict a student's final grade with this information what should I use then after that you're gonna have to train a linear regression model to make this prediction and you should be looking to predict you know an exact integer value or an exact desc values so I believe the grades are from 1 to 20 where it's like every one point is like a half point of a grade or something I don't know exactly how it works but you guys can look at the data set understand it and try to do that this is pretty straightforward and if you're having trouble with this I actually do have a complete tutorial on how to do this and I will leave a card or like a link in the description to that so you can have a look at that for reference as well alright so project number two so now that we've used linear regression a pretty basic algorithm it's time to kind of move into a different flavor of machine learning which is clustering and classification so for this one I'm gonna recommend you use K nearest neighbors to do this and to recommend or not recommend but to classify cars as either safe or not safe based on data about those cars now I have a data set down below and again you're gonna have to follow the same procedure at first understanding what this data set is and how it works and I'll leave the links to the original data sets you can read that but essentially this data set has information like the safety rating or like the amount of doors just many different things about a car and then it also has you know how safe that car is so using this information you know try to be able to pass a few parameters to your model that can predict based on some certain information if this car is going to be rated as safe or not safe this is a cool one because it doesn't involve you know predicting a number value and some of this information you could probably omit and you need to kind of encode it into integers because the information is not just like 0 1 2 3 it's like safe not safe like accuracy very high like it has words and it so you need to deal with that information which is something that you're gonna have to learn as you get better with machine learning and it's a good way to practice kind of you know encoding this information same thing here I have a tutorial on this I will leave a link down below so you guys can reference that if you're having trouble alright to machine learning project number 3 classifying cancer tumors as benign or malignant I believe that's how you say that so essentially cancerous or non-cancerous now this I think you can probably assume why this might be a cool project first of all this is a pretty good application if we can look at a tumor and based on some certain properties on it be able to use machine learning to determine whether or not this is cancerous or not which is actually what doctors use all the time is a bunch of machine learning an AI to help them do these kind of analysis but anyways there's a very popular data set I believe it's just called like the cancer data set or something like that that I've linked down below and using this data set again it has a ton of different properties I want to say it's like 30 or something pick out which ones might be important to classify tumor information to classified as cancerous or non-cancerous now notice I haven't given you a model to use for this I haven't said you know use this algorithm or use this and that's because you can use different algorithms to do this now I would recommend possibly some kind of unsupervised algorithm maybe a neural network maybe some kind of clustering algorithm like k-means but you can do this in a few different ways and I'd be interested to see what you guys think you're gonna do for this so this is kind of amping up the difficulty a bit because you need to make the decision on which algorithm you should use to classify this data all right so now my last project idea for beginners this is not very original whatsoever because this is kind of you know the hello world of machine learning but it is using neural networks to classify the amnesty digit data set now if you're considering yourself a beginner here chances are you might have actually already done this but what I want you to do is try to do this by yourself so there is a data set it's the M nest digit data set essentially contains pixel data of a bunch of different images that are numbers so like handwritten digits like 0 through 9 and what you need to do is write some kind of ML algorithm using a neural network or whatever it is you want to use to classify these digits based on the pixel data so you're gonna have to figure out if you're using a neural network which is what I would recommend what is your input layer gonna be and then what is your output layer gonna be how are we going to encode this information how are we going to determine if a number is you know 0 through 9 how are we gonna take these pixels that are two-dimensional so you have rows and you have columns and feed them into our neural network and this is a more complicated thing there's tons of tutorials online I have one on how to do this as well that you could follow if you're stuck but really just try to think about these because this is what you need to do to get better at machine learning and artificial intelligence a lot of it is not just writing the code but thinking about what should I use what's my input going to be what's my opiate gonna be how am I going to get this data nice and clean and how is all of this gonna work so anyways those are my kind of four project ideas for machine learning for beginners now if you guys have any other ideas for projects please don't hesitate to leave them down below as it can definitely help other people out with that being said if you guys enjoyed the video make sure you leave a like and subscribe to the channel for more content like