I am super excited to announce the beginning of natural language processing in python tutorial playlist today.
Previously I had uploaded machine learning and deep learning tutorial playlist, along with end to end projects on my youtube channel, which received more than 2 million views and not only people were able to understand complex topics using very simple and intuitive explanations but they were able to practice on end-to-end projects and they got lot of them got a job as well. I have received so many testimonials after the success of those two series when I conducted a survey on my channel regarding, which series I should work next; an overwhelming majority of you suggested nlp.
There will be four main highlights of this entire playlist. Number one is very easy intuitive understanding of complex topics. You can go to youtube and search for what is convolution neural network and you will find my video where I used visual representation to explain very complex topic in such a way that even a high school student can understand it easily. By watching that video you will get a glimpse on what kind of presentations you can expect from this particular tutorial series.
Highlight number two will be a lot of hands-on coding and exercises.
Highlight number three would be end-to-end projects. So we will take a real industry problem and build an end-to-end application in nlp along with deployment into the cloud.
And the fourth highlight will be expert talks I know a lot of friends who have who are working as a data scientist or nlp engineer in the industry both in US, and in India and I want to invite them to discuss different topics or how nlp is used in the industry so that youre not just learning academic topics and just practicing on on some dummy toy examples but you also get a feel of what goes on in the industry. I want to say my special thanks to the authors of Practical Nature Language Processing book, because some of the content in this playlist is going to be influenced by this book. I have read this book its an amazing nlp book it has lot of practical tips on how to build nlp system end to end to solve various industry use cases.
Two authors of this book have been on codebasics youtube channel. Anuj Gupta who is a head of machine learning at vahan and Modi Satva, a facebook ai researcher working in California here in USA.
These these guys are experts they know what theyre talking about and I absolutely recommend this book to anyone who is interested in nlp. This is not a sponsored video, by the way this is my own genuine feedback the link of the book is in video description below.
Now what exactly is nlp Im going to show you some real life use cases where nlp is impacting you and by looking at those use cases we will understand what is nlp. The first use case is gmail. When youre typing any sentence in your gmail you will notice that it tries to auto complete see here it says if it changes in the future and this auto completion is done using nlp. The other standard use case is spam filters.
If these emails didnt have spam filters then you will be so much worried and you will get so much headache by this bombarding of all kind of commercial spam but luckily using nlp you can filter them and you can take them out of your inbox and in nlp they use these keywords.
The spam emails will have things like hurry up the offer ends and you want some prize and all of that bull shit. So based on some of those terms and using some machine learning based classification model, you can filter the spam messages.
The other use case is language translation where you can using where using google translate you can pretty much translate a sentence in one language to another language, with very high accuracy. This wasnt possible many years back but nowadays this translators are pretty good.
The other standard use case is customer service chat bot. So nowadays if you are using any service, lets say bank and you go to their chat service you type in a message and many times there is no human on the other hand.
The chat bot can interpret your language the question you are asking and it can it can derive intent out of it and it can respond to your question on its own and sometimes when it doesnt work well then they connect it to human beings. So chatbots is becoming a big use case in nlp.
The other standard one is the voice assistants such as Amazon, Alexa and Google Assistant. I have a Google Pixel phone here, and Im going to show you, Im just going to ask what appointments do I have tomorrow and it will show me so it actually showed me what kind of appointment do I have. So these voice assistants are becoming very powerfu, if you have amazon alex our google assistant at home they can even tell you okay you have a meeting at nine oclock and there is a lot of traffic so instead of leaving at 8 30 you should live at 8 15. It can do do that kind of assistance.
Google search uses nlp or to answer your question and they use this special language model called BERT. So before BERT if you type this question in google saying can you get medicine for someone pharmacy. See before that it wasnt giving you the right answer but now it gives you a precise answer and this is again possible because of the use of natural language processing.
Automated news generation for news companies is another use case. I work for Bloomberg and in Bloomberg we automated the news stories to predict market events. Im going to link this Bloombergs article in the video description below but basically the picture here is of a bloomberg terminal and all the new stories that you are saying they were not written by some human editor basically. They were all generated by computer using nlp, using some ai techniques, they detect some signals and they can auto write these stories and thats pretty cool. There are many other use cases as well but to summarize, nlp is a field in computer science and ai that gives machines an ability to understand human language better and to assist it in language related tasks. If you look at computers traditionally they are designed to work on numbers and they were pretty good at working at tasks which are related to numbers, but now using nlp you can have computer assist you in language related tasks, the use cases that we saw like question and answer the the language translation things like that and in this tutorial series we are going to use python of course as a programming language and we will be using spaCy, Gensim, NLTK. These are like different libraries that allows you to do nlp in python.
Well also use scikit learn for our machine learning problems and then TensorFlow and PyTorch for deep learning related problems in nlp and Hugging Face too.
Now I know there is a lot to learn, what well do is well take a very practical approach well take a concept or a problem and then whatever library can solve the problem in a better way well try to use it. So dont worry that youre going to run or so many things like say one or eight different libraries, in the end these are all tools what you care about is here is the given problem how do you come up with a solution and for coming up with that solution in a better way whatever tool you need to use, you need to use its a syntax you can google it and you can figure things out easily. Now talking about career opportunities in the field of nlp, you have three roles that you can select from, one is data scientist specializing in nlp. The second one is nlp engineer which is basically a machine learning engineer solving nlp problems and the third one is nlp researcher.
All of these roles are very high paying, so if you have expertise in the field of nlp and if you choose any of these roles in US, for example you can make anywhere anywhere from 100000 dollar per year to 650000 dollar per year. Now all of this depends on your experience, uh the company that youre working for, the location and all those factors; but Im just giving you a broader range if you are in India you can make anywhere from 10 lakh rupees a year to one crore rupees a year.
Yes people make that kind of crazy money when they work for big tech company and they are solving cool problems. Thats all we had for today. In the next video we are going to talk about why nlp is booming right now.
In terms of video upload schedule, I will try to upload one video every week but my schedule is very crazy nowadays. So if there is a delay please be with me but I am committed and I will try my best to wrap up this series as soon as I can.