Sentiment Analysis in DataSpell with @JetBrainsTV

hi everyone I'm Jody Burchell and I'm the developer advocate in data science at jetbrains today I'd like to briefly talk about sentiment analysis this is the ability to automatically pick up whether a piece of text has a positive or A negative tone and it's particularly helpful for determining things like how satisfied customers are with your company by analyzing that feedback or those social media posts that they're giving to you so my favorite package for sentiment analysis in Python is called Veda and it uses a bunch of rules to get the overall sentiment of a sentence by seeing how words behave in context so it can pick up not only the use of words such as excellent or terrible which have obvious sentiment it's quite a clever package because it can also pick up on things that change the intensity of that sentiment such as capitalization exclamation marks and even emojis and it can also take into account the use of qualifying words such as really or kind of which give words more or less importance so let's have a quick look at how it works so you can see that we've installed the sentiment intensity analyzer method from the beta sentiment package and we've assigned it to the more readable analyzer variable so let's now pass a sentence through this analyzer what you can see is that we get four scores for each sentence we get a negative a neutral a positive and then an overall compound score the closer to one this compound score is the more positive the sentiment and the closer to negative one the more negative the sentiment let's now have a look at how the context of the word fun changes the sentiment of the sentences so you can see in this second sentence we've added kind of in front of fun which has slightly reduced how positive the sentiment is in contrast adding really slightly increases how positive the sentiment is and then finally adding things like capitalization exclamation marks and this fun little smiley face here drastically increases how positive the sentiment is so I do hope you enjoyed this quick overview of Veda and what I really hope is that you're feeling inspired to go away and try your own sentiment analysis project so if I were to give you a book recommendation to help you in the new year it could only be Gilbert strang's introduction to linear algebra so this book and the corresponding lecture recordings on MIT open courseware are such a friendly and engaging introduction to linear algebra and once you finish them you're going to understand a big chunk of the math behind algorithms like principal components analysis neural Nets and word embeddings so thanks so much for watching if you have any questions or comments I really would love to hear them so please leave them in the comments section below or reach out to me on Twitter or Mastodon and if you'd like to see more of my content you can also check out my writing on both my and jetbrain's blogs have a great day foreign