[Python Project] Sentiment Analysis and Visualization of Stock News

Enroll in The Complete Python Programming Bootcamp! www.udemy.com/course/pythonbo...
Become a Member on TheCodex for FREE and jumpstart your career - thecodex.me
This project is a beginner-friendly Python and Data Science scraper focused on building a script that analyzes the sentiment of news articles of stocks on FinViz. We'll learn how to use the Requests module in Python, get the HTML returned in BeautifulSoup and parse the Article Data, apply Sentiment Analysis on the data with NLTK, and finally visualize the results of our data frame in Pandas with MatPlotLib.
Watch this entire project and more on TheCodex for free! thecodex.me/projects/sentimen...
Blog for Project: blog.thecodex.me/sentiment-an...
GitHub for Source Code: github.com/TheCodex-Me/Projects
00:00 - Part 0: Welcome to the Project
01:25 - Part 1: Getting Finviz Article Data
11:05 - Part 2: Parsing and Manipulating Finviz Data
21:47 - Part 3: Applying Sentiment Analysis
33:27 - Part 4: Visualization of Sentiment Analysis
► SIGN UP ON THECODEX TODAY! www.thecodex.me/
► THECODEX BLOG: blog.thecodex.me/
► TWITTER: / thecodexme
► FACEBOOK: / thecodexme
♦♦♦♦♦♦♦♦♦♦
THECODEX is a company focused on empowering and enabling anyone to learn how to code. Established in 2015 by Avinash Jain, THECODEX has taught over 600,000 students around the world how to code. With over 16 programming courses on a wide variety of topics, and more than 20,000 5-Star reviews, THECODEX is the right choice for you.
We provide engaging and comprehensive videos that break down complex information and cover everything you need to know about the subject. All students get access to our Q/A forums where you can get your doubts and problems cleared up. At the end of every course, you will receive a certificate of completion.
Jumpstart your Coding Journey with THECODEX today!
♦♦♦♦♦♦♦♦♦♦
Music: Flash Funk (Marshmello)
♦♦♦♦♦♦♦♦♦♦
Project inspired by Damian Boh's post: towardsdatascience.com/sentim...

[@TheCodex]

Trying something new with this Project - all the videos in the series are in one complete video! Let me know what you think of this format :) 00:00 - Part 0: Welcome to the Project 01:25 - Part 1: Getting Finviz Article Data 11:05 - Part 2: Parsing and Manipulating Finviz Data 21:47 - Part 3: Applying Sentiment Analysis 33:27 - Part 4: Visualization of Sentiment Analysis Watch the videos as a series on our platform: thecodex.me/projects/sentiment-analysis-of-stock-news-python Follow along the Project on our Blog: blog.thecodex.me/sentiment-analysis-tool-for-stock-trading/

[@Ur.Podcast_Buddy]

1st viewer

[@nolann2382]

@Matthew Khari @SelfPromotionPolice

[@seizuresaladventures]

You have covered so many concepts we’re covering in a data science boot camp. Much props and kudos to you. I value your knowledge sharing - bless

[@skybeat6931]

You have a passion for teaching. Great work mate.

[@jibrilrichter2852]

Dude, you are amazing! I just like the way you explained the entire process. You got yourself a new subscriber!

[@Jam05_]

Thank you so much for this amazing project bro!

[@rabeyaakter1830]

It was really good. Loved your energy!

[@tvsk7838]

Great Work Bro!! Very Well Explained . I followed along and built this program on my laptop.

[@philrenda]

This is fantastic; thank you for sharing. It took a little work to get the modules installed correctly. Pillow had prerequisite libraries that were missing. I would be very interested in learning how to modify the code to layer closing stock price onto the graph....

[@kathiravanmm8791]

Awesome Project yet Explained simply and earned yourself a SUBSCRIBER 😎😎😎

[@raymondpeng2005]

Great project. Thanks for sharing

[@capitalofmotivation6369]

such a great project, proud of you

[@LAME_MMA]

This is an excellent resource, thank you for sharing. And you have earned a subscriber \m/

[@Maximus18.6]

Mate, this video is exceptional,. congratulations and thank you so much for sharing.

[@lancehoward1609]

You are a boss. Excellent project

[@sarthaksurana8328]

Amazing project..loved it

[@TheCodex]

Thank you so much 😊

[@John-dw6jb]

GREAT TUTORIAL THANK YOU

[@prathameshmohite3755]

Thanks for this amazing project. I guess we can get much accurate sentiments if we consider extracting the entire news article instead of just the headlines?

[@69nukeee]

Very well done, great job! Thanks for sharing!

[@sang7081]

excellent project!

[@SameerKapur]

dope project love it

[@TheCodex]

thanks!

[@nicholasn255]

Thank you for this tutorial! In terms of design of this project do you have any pseudocode version? or possible resources which look at the pseudo code?

[@Maximus18.6]

Excellent video

[@randb9378]

Hi there, Great video and really helpful. I am trying to do something very similar using VADER, but do you have any suggestions on how I can do this on real-time by extracting data from a website, and automatically updating the charts? Also, could I apply this on an index instead of a company?

[@jameszhou162]

Hi Avi, great video on introducing NLTK score on text! everything going great except i can extract timestamp from finviz site. instead of simple space split, there are etc. what would be a quick way to get rid of them. thanks a lot.

[@cavingada]

What are your thoughts on us building our own model for determining polarity scores? Perhaps feeding a train dataset with article headers, time and stock movement at that time to help predict the score? Do you believe the results would be more accurate (proportionally?)

[@definitelynatty3934]

For error 403 use: hdr = {'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.64 Safari/537.11', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.3', 'Accept-Encoding': 'none', 'Accept-Language': 'en-US,en;q=0.8', 'Connection': 'keep-alive'} req = Request(url=url, headers=hdr)

[@zora.yt1407]

your a life saver

[@md.shafaatjamilrokon8587]

Thanks

[@binhpham-by6ch]

Thank you for the great tutorial. I want to use the same platform as yours. Can you help me how to set it up ? Thanks

[@krishnatejbhat2643]

Hey this is a great video. I was wondering I could use this for my portfolio?

[@wilsongomes3360]

A good job

[@anashussain4623]

I am getting "ImportError: No module named request" I event tried installing requests module but didn't work

[@hardikbisht7516]

can we use differnt article site?

[@fatimabaig5286]

I am getting this error "RecursionError: maximum recursion depth exceeded" when I type print(news_tables) at the end of the part 1.

[@ryafc16]

Is the code on GitHub?

[@thrqureshi]

Is there a way to run this script in google sheet,

[@kanechulchulikpongse9549]

great

[@amareshwarkaratagi5551]

Which tool (LIke VS, Sublime) is used in this video for compiling code....?

[@shishirbagalkot1068]

Pycharm

[@donbray783]

Sound's good, maybe u to make The parte nine.

[@jeremyedwards4819]

Is there a way to select news from a range of dates? Maybe like just the past month or so?

[@great1xD]

That's exactly what I'm currently looking for

[@hera-12]

req = Request ( url = url, headers ={ 'user_agent': 'my_app'}) it is showing error , can u please tell me what have I done wrong

[@stoner_2474]

Can this be done on google collab ?

[@tarunsureshkanna3437]

What are the algorithms used here..?

[@irisliu4837]

Excuse me, how to get the user agent authentication?

[@noelthomasbejoy3089]

Why does print(parsed_data) only give results of AMZN and not of the other tickers? Isnt that the point of having the outer for loop?

[@franciscobarros6111]

It shouldn't. Might be you kept the break on the first loop.

[@noelthomasbejoy3089]

@@franciscobarros6111 Actually not, it's apparently because print has a limit to how many characters it can display.Just try it yourself (without break)

[@Jam05_]

So have you overcomed with your problem which you asked here?

[@noelthomasbejoy3089]

@@Jam05_ yes

[@Jam05_]

@@noelthomasbejoy3089 can you help me how please? thanks in adv!

[@gerardpuche]

I did, and it works fine, but it gets just a few headlines, how do I fix it? There's no error showing up, but it only gets a few??

[@guruswamy2248]

Hi how to make this application with flask

[@SantoshLLC]

Can we do it on google colab?

[@pratik1762006s]

Do you have the code

[@elghark]

I don't get any response. Maybe I have to register?

[@fatimabaig5286]

File "/Users/letsbegin/PycharmProjects/pythonProject/main.py", line 23, in for ticker1, news_tables1 in news_tables.items(): TypeError: 'NoneType' object is not callable

[@qingboyuan6919]

it would be more interesting to see the sentiment of each article instead of the title only.

[@techpop472]

How long does this project take?

[@TheCodex]

Beginner friendly project & can be completed in an evening!

[@yukunli5396]

Can you share your codes? Thanks.

[@belalkhalil4864]

Question: I get the below error running the request for the URL. Has anyone been able to resolve this? Thank you, extremely useful tutorial and very beginner friendly! thanks for creating this.

[@belalkhalil4864]

I have actually figured out that this was not working for me while using Jupyter on an online learning platform. The reason is that some of these platforms have a list of approved websites that are 'whitelisted' for sending data requests. Otherwise, requests return a 403 forbidden access error. When I ran this locally on Pycharm, it worked fine. I also plan on running it locally on Jupyter for the better visualizations. Thanks again for the tutorial! definitely subscribing and trying out a few other projects!

[@nc4576]

@@belalkhalil4864 hi I’ve run it on pycharm but I am still getting URLError url open error ?? Unable to get local issuer certificate

[@definitelynatty3934]

hdr = {'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.64 Safari/537.11', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.3', 'Accept-Encoding': 'none', 'Accept-Language': 'en-US,en;q=0.8', 'Connection': 'keep-alive'} req = Request(url=url, headers=hdr)

[@roudal-majid3546]

why am i getting error ? AttributeError: 'NoneType' object has no attribute 'findAll'

[@SuperDrunkenmunky]

I found that when you overload the amount of tickers it does this, if you reduce to 5 tickers it should work

[@SuperDrunkenmunky]

correction, i found i get this when the desired ticker has no input articles

[@naziaali642]

why are not we splintting the data into training ad testing

[@TheCodex]

Watch this project and more on TheCodex Platform! thecodex.me/projects/sentiment-analysis-of-stock-news-python Prefer reading over watching? Follow along on our blog: blog.thecodex.me/sentiment-analysis-tool-for-stock-trading/

[@apprendrelafinanceautremen711]

good morning , can I ask you ?

[@abbcc555]

This no longer works due to finviz limiting bots.

[@timchau1954]

AVI U SOUND SO DIFFERENT HERE??

[@dumbfuckjuice9915]

guys i found a solution i need credit so yeah if this yt utber reaches out i will give it to him