YT YouTube2Text

YouTube2Text → Transcripts → What is Machine Learning? | 100 Days of Machine Learning

What is Machine Learning? | 100 Days of Machine Learning — Transcript

by CampusX · 2,860 words · 322 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Hey Guys,
0:01Welcome to my YouTube Channel
0:03In this video,
0:03I am going to make an announcement.
0:06that, I'm going to create a new Playlist
0:07for my YouTube Channel.
0:10and I'm going to call that Playlist
0:11"100 days of Machine Learning"
0:14So...
0:15You might be thinking
0:16Sounds interesting
0:17But what is it exactly?
0:19Here's the thing
0:20In the past few days,
0:21Many of you messaged me..
0:24over WhatsApp or through Youtube comments regarding
0:28They weren't able to find..
0:29a end-to-end machine learning playlist on my channel.
0:32and that is true actually.
0:33If you go to my channel,
0:34you would find videos related to Machine Learning algorithms
0:38I've created videos on various machine learning algorithms.
0:41But a complete end-to-end playlist isn't available in my channel.
0:47So,I thought why not create one comprehensive Playlist on Machine Learning?
0:52So, that's the plan.
0:53In the coming 100 days,
0:54I have planned to shoot and upload one video every day.
1:00I've developed a structured curriculum
1:04To the best of my experience and knowledge,
1:06I think this Playlist will be more than sufficient to teach you..
1:09Intermediate-level machine learning.
1:11For advanced-level machine learning, it's completely in your hands.
1:13but if you're a beginner or slightly more experienced,
1:17If you follow the course throughout,
1:21I hope you'll advance to a proficient level in Machine Learning.
1:26Now comes the question about
1:27What are the topics we'll be covering?
1:32So, I will tell you something about myself
1:34When I began learning machine learning,
1:36my primary focus was on mastering ML algorithms.
1:41Gradually, I realized that after learning the algorithms
1:45and working on few projects,
1:48there are two essential aspects to focus on in Machine Learning
1:51One of them is learning algorithms
1:53knowing about algorithms is mandatory.
1:55But along with that,
1:57Simultaneously, it's crucial to understand how to develop an end-to-end machine learning project,
2:00including the complete flow.
2:02We call this as "Machine Learning Life Cycle"
2:04also known as the "Product Life Cycle".
2:08and beginners usually doesn't focus much on this aspect.
2:11Beginners often believe that knowing Machine Learning algorithms alone is sufficient.
2:15but not really.
2:17So, I planned in the following 100 days..
2:20I will start with the basics of Machine Learning
2:22and I'll cover the entire flow of Machine Learning.
2:26If you want to do a Machine Learning project,
2:28I'll cover all the potential challenges you might encounter while working on an Machine Learning project.
2:36Since we're covering 100 topics in 100 days,
2:39no topic will be left untouched.
2:42Keep in mind, we won't be delving into algorithms in this Playlist.
2:46We are not going to cover any of the algorithms.
2:48For algorithms, we already have a separate playlist in our channel.
2:51If you want to learn about any algorithms,
2:54then you can go to the playlist of that particular algorithm.
2:59But in this "100 days of Machine Learning",
3:03Here, we learn the techniques
3:05we learn the flow
3:07we learn about how we deploy.
3:10We'll learn how to perform imputation,
3:12how to perform pre-processing,
3:13how to perform analysis,
3:15model selection, feature selection,
3:18and such important concepts like
3:19What is Bias-Variance Trade Off?
3:21and the weighted and important topics
3:24that differentiates ordinary Machine Learning engineers
3:26from extraordinary Machine Learning engineers.
3:28We'll cover such topics here.
3:30I am still working on the curriculum
3:32it is not completed yet.
3:35I will be uploading that curriculum as well in the coming few days.
3:37In fact, if want any certain topic to be covered,
3:42you can share that with me
3:43and I'll be covering that topic as well.
3:45But I'm thinking in this 100 days,
3:48I have to create a resource that is meaningful for you
3:52and your Machine Learning journey.
3:55and the other question which may pop out in your mind
3:57For whom is this playlist intended?
4:00If you are a beginner, then it is definitely for you.
4:04and even if you know intermediate machine learning,
4:07it could be a great resource for you
4:10as you can come here
4:12and use this as a valuable resource
4:15to watch and learn anything that you could've missed,
4:18or gain a deeper understanding of familiar topics with clarity.
4:24This will be helpful for you in such cases.
4:26In short, this will benefit everyone following the channel,
4:29including students and professionals.
4:32I'll try to be very honest while making these videos
4:37I'll try to put my maximum effort
4:40So, yeah..
4:40This is the announcement I was planning to make.
4:44Now, I am not going waste any time
4:46I will start it today itself.
4:48Today we are going to cover the first topic
4:51which is "What is Machine Learning?"
4:53I know that many of you already knew it.
4:55However, I want to begin everything from scratch.
4:59So let's cover this topic: What is Machine Learning?
5:01So, let's dive into what machine learning is
5:04We'll start with the formal definition
5:06According to definition,
5:07Machine learning is a field of computer science
5:09that uses statistical techniques to give computer systems
5:13the ability to "learn" with data, without being explicitly programmed.
5:18To put it in simpler terms,
5:21Machine Learning is all about learning from data.
5:26There is it a term in this definition called as "Explicit Programming"
5:29Let's explore the concept of explicit programming.
5:31Explicit programming involves writing code for each specific scenario.
5:35To handle that scenario, you write a code.
5:37But in Machine Learning, you don't do that.
5:40what you do is that you've got some data
5:42and you've got an algorithm
5:44You instruct the algorithm to explore the data
5:47and identify patterns between input and output.
5:50Once you've identified the patterns,
5:52we provide new input to the algorithm to derive the output.
5:57If you check out this flow diagram,
5:59you can observe that in conventional programming approach
6:02we write a program
6:05for which logic is written by us.
6:08If you give input to that program, you'll get your output.
6:12But in Machine Learning, things are different.
6:14what you do is you provide some data.
6:17In that data, you give an input as well as an output
6:20But you haven't written any program or logic.
6:23That logic is generated by none other than the Machine Learning algorithm.
6:29The good part is that you don't have to write code for each condition/case.
6:35It is automatically handled by Machine Learning algorithm.
6:39For example, you've written code for adding two numbers
6:44Whenever you give two numbers to that program,
6:47it returns you the value of sum
6:50But in Machine Learning, what you will do is you give data
6:55You'll give an excel file in which the rows contain the numbers and their respective sum.
7:00Whenever Machine learning models train on that data,
7:05The model realises the pattern as addition.
7:09After training, irrespective of giving two or four or ten numbers as an input,
7:12your machine learning model knows that it has to perform addition.
7:16It adds all of them and gives to you.
7:17Whereas in the code written for sum of two numbers,
7:21If you give more than two numbers as input, then the program doesn't function
7:26since it is explicitly coded to perform sum of two numbers.
7:30That's the key difference
7:32I hope from the example
7:34you can understand the reason behind the powerful nature of machine learning in the industry.
7:40Now that we know about Machine Learning,
7:45Let's also discuss when and where Machine Learning is used
7:51Let us know in which type of scenarios, Machine Learning can be used
7:54and is useful than traditional software development
7:59I'll provide you three scenarios
8:00and there are other scenarios as well.
8:02These three scenarios feel significant to me.
8:05First scenario is that you can't perform few things using programming
8:10You can't write cases for everything
8:13In such situations, we use Machine Learning
8:15I will give you one real life scenario
8:16Consider that you are trying to build an e-mail spam classifier
8:20to detect whether the given email is spam or not.
8:25If you were given to write a program for that as a software developer,
8:32Then what would you possibly do?
8:33you will pick a bunch of e-mails
8:35and you'll have the information about that e-mail, whether is spam or not
8:39then you will try to create patterns for them.
8:42like any word such as 'discount' or 'sale' or 'awesome' is repeated more often
8:49or it is filled with bunch of pictures,
8:51then you create a long if-else ladder by using if else for each and every condition
8:57That would be the possible program for you Spam classifier.
9:04but let us say that you've written something like
9:07if 'huge' is repeated more than three times,
9:11huge as in discount is used,
9:13you'll label that e-mail to be spam.
9:17and somehow advertising companies got to know that
9:23if code is written to classify the mail as spam if the world is repeated more than three times
9:32then those companies might use 'big' or 'massive' instead of 'huge'
9:39then the program couldn't pick that mail as spam.
9:46then you've to modify the logic of the code
9:52to handle a new scenario
9:54but again, advertising companies can experiment with different words if they get to know about it
9:58In short, you've to keep changing the logic more frequently
10:02to make sure that the code runs effectively.
10:05In machine learning, it doesn't happen that way
10:08since it learns from data, if data changes, then it will automatically get reflected in logic
10:13That's the beauty of 'Machine Learning'
10:15You just have to write just one algorithm,
10:18and everything will be handled by that algorithm itself
10:22and second scenario where Machine Learning is useful over traditional programming
10:30A scenario where you cannot even imagine the number of cases
10:37like 'Image Classification'
10:38Let us say you want to classify dogs, that if dog is present in picture or not
10:44There will be hundreds of breeds which varies in looks,
10:49few large and few short,
10:52vary in colours and other characteristics.
10:58So, if you were to create a program that detects the presence of dog in a picture
11:06Can you imagine the number of cases you've to write inorder to cover characteristics of every breed?
11:13You cannot do so.
11:14You cannot code it
11:17We have use the technique that we, humans, use to identify dogs
11:22We were thought from our childhood to identify that particular animal is dog
11:26That one is not a dog, that is a cat
11:28Our mind mentally keeps tagging the name with the animal
11:32It keeps learning from the data
11:34It is also one of the scenarios where you cannot use conventional software development approach
11:39You'll have to use the Machine Learning approach
11:42One more important use case is Data Mining.
11:46What exactly is Data Mining?
11:49First, let's discuss about what data analysis is.
11:53Data analysis is a process where you extract patterns or search for hidden information
11:59by plotting graphs
12:00That is 'Data Analysis'
12:02but sometimes the information is more hidden which you won't able to get through graphs
12:08I'll give you one scenario
12:10Just by looking at the e-mail content,
12:14if we're unable to detect key words due to which we can the e-mail as spam,
12:22then you perform Data Mining.
12:26In Data Mining, you
12:28apply Machine Learning algorithm on the data
12:31you create a prediction model
12:34just like e-mail spam classifier
12:37you can check the patterns extracted by the Machine Learning model
12:43like if 'huge' is occurring more frequently, then there it is labeled as a spam
12:49if it is not that frequent, it is treated as not spam
12:51After applying Machine Learning,
12:54If you are able to extract important data from the information,
13:00That is known as "Data Mining"
13:01Most of the times in order to extract the hidden patterns,
13:07in order to perform such data analysis, we use Machine Learning
13:11and this is called as Data Mining
13:13Machine learning is a very important tool to perform data mining
13:17You should have understood the importance of machine learning using these 3 to 4 scenarios
13:22And why is it taking over the world
13:27Next you should be knowing a little history about Machine Learning
13:34You should be learning about it's history if you are starting on any new technology
13:43I feel that history of machine learning is more or less like Nawazuddin Siddiqui
13:50I hope you all know about him, he is a great actor
13:53Just like Nawazuddin Siddiqui, machine learning is already existing from a long time in the industry
14:03He was playing a very small role in Munna Bhai MBBS
14:09Similarly, machine learning is there from 40 to 50 years
14:13But it couldn't get into limelight like other important technologies
14:19Until the recent 2010s
14:25Only from then,it raised to the level that machine learning is today
14:31If you talk about Nawazuddin Siddiqui, what could be the success reason behind him?
14:36It might be the OTTs or the audience preferring the content-based films
14:43So there is a paradigm shift which led him to be one of the biggest actors of the country.
14:50Similar events occurred in the case of machine learning also
14:54All the theory and Maths existed from very long time
15:00but machine learning is not that famous
15:02because of the reason that machine learning requires significant amount of data
15:09Unfortunately,back then gathering and labelling the data is a quite tedious task
15:17Also, the inefficiency of hardware to run algorithms on such data back then
15:28After 2010, with evolution of internet and smartphone,
15:33These two problems are sorted out.
15:35We are generating data at a heavy pace,
15:38You can consider the example of your own life from morning to evening before going to bed
15:42So much of data is generated by yourself alone
15:44Then imagine the data generated by 4 billion Internet users around the globe
15:50In fact, the amount of digital data created from the starting of mankind till 2015
15:57is generated in 2016 alone
16:02That is the speed at which we are generating data
16:05this data is aiding the growth of machine learning
16:09Second thing is hardware
16:11In the modern day world, we ourselves are carrying up to 12GB of RAM in our pocket in the form of mobile
16:17We are carrying GPUs in our pocket
16:20which were not available to research scientists.
16:23Even 128MB RAM was a big deal then
16:27but we are equipped with good hardware, data and algorithms now
16:32That is the reason why machine learning is enjoying it's fruits
16:36This is not going to stop any time soon
16:38The growth curve will keep on growing exponentially
16:41and that is the reason for jobs in this sector
16:44If we talk about jobs,
16:47I will also have to discuss about few things
16:55Do you think the jobs that are available now, the salaries which we get now in this industry
17:02Will they continue to be same in the future?
17:04The answer is NO
17:05This is just pure economics
17:07When Java entered the market, only a handful of people were familiar with the language.
17:11But companies needed Java because their competitors were implementing it
17:18Therefore, as a company, I also need Java in my software
17:23For that, I have to hire some professionals
17:26But when I reach out to market for hiring process, I realised that there is a lack of talent in this space
17:30So all the companies will be fighting for those few available professionals
17:35So that would obviously lead to more salary of that professional
17:39SIMPLE ECONOMICS
17:40The similar trend is now going on with machine learning
17:44Even in the colleges, machine learning is not being taught
17:46and many of the engineers doesn't know machine learning as of now, which is changing gradually
17:51Every time when a company goes to a college, they only find less number of students who know Machine Learning
18:00So they have to fight for those students resulting in a higher salary
18:04Over time, as salaries increased, more people became interested in learning the technology to secure those jobs.
18:26When everyone in the market learns machine learning in the coming few years
18:34Majority of the population would be knowing machine learning just like Java today
18:38Once most of the people knows machine learning, the salaries would be automatically normalised
18:44as companies will have a lot more options
18:48and that's the reason why they won't be paying high salaries that they're paying today
18:57The good part is that we are at an initial growing phase
19:03The graph of any technology will be like this,
19:05initially increasing and then decreasing
19:08The positive aspect is that for those learning machine learning, we are on an upward trajectory.
19:14and there is still time which I feel
19:17If we learn correctly now,
19:18we can expect to achieve the same level of success that machine learning is currently experiencing.
19:23I understand it's been a lengthy video,
19:25but this was the introduction I wanted to provide on machine learning.
19:31In the next video, we'll delve into an essential topic that often raises doubts among beginners.
19:38What is the difference between AI, ML and DL?
19:42We will be covering this on the next day, which is Monday.
19:49That's a wrap for this video. I hope you like it.
19:52If you plan to follow this series, '100 Days of ML,' please consider subscribing to the channel.
19:58Thank you for watching!

About this transcript

This page contains the full transcript of What is Machine Learning? | 100 Days of Machine Learning by CampusX, generated from the public captions YouTube serves with the video. The transcript has 2,860 words across 322 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.