Scatter. And so twitter_data.retwc, twitter_data.polarity. So let's just go ahead and plot this. As so you can see, and this is just a good visualization to see how they relate. So a lot of Twitter has, all of these things have zero, not many, no retweets. But you could kind of see that maybe things with higher retweet count tends to be negative, on the negative side, right? So that's kind of interesting. So maybe things with negative opinion expressed tends to be tweeted more. Let's see its subjectivity. So, remember, subjectivity goes from zero to one, so on the y-axis here, the higher the value means more subjective. So that's a mix here, because here's the one that got retweeted a lot, and it's very objective. And here's the one that get retweeted a lot, but it's very subjective, so maybe there was some extreme here. Let's dig a little bit deeper, and again, this is to show you how you can go for further analysis. Because often, just looking at the whole data, things may not be as clear. So let's actually find data that's really more subjective, because here subjectivity goes from zero to one, and we're kind of throwing all of them together, but let's get the real subjective data. So let's get a subset. Here from twitter_data. So we want twitter_data that meets certain condition for subjectivity. Subjectivity being greater than 0.5. So, as I said, between 0 and 0.5, it's more objective than subjective. Okay, so this is the condition we're putting on Twitter data. So we're getting Twitter data where subjectivity has greater than 0.5, okay. And let's go ahead and print this correlation. And for the time being, I'm going to just comment this plotting. I'll just go ahead and run this. So let's see if something or anything changed. So what we're seeing here, we have two tables, the first one with all the data and the second one is with subjectivity being greater than 0.5. And so, maybe there isn't really so much difference between retweet and subjectivity correlation. Actually, I'm sorry, we didn't actually print that, we printed the same table again. So we need to have subjective here, okay, and let's run this. Okay, so now we actually have two different tables. All right, so before, this is the first table with all the data. And so you can see that between retweet and subjectivity, there isn't really much correlation. But once we take only the more subjective tweets and between retweet and subjectivity, now there is a positive correlation. So notice the difference, before, there was a negative correlation, yes, given that it wasn't strong, but it was on the negative side. Now there is a positive correlation with subjectivity. All right, so with this, it shows that with higher subjective tweets, the more subjective they are, the higher retweet they get. Okay, and that kind of maybe explains this that up to some limit, higher subjectivity doesn't guarantee more retweets. But beyond some level of subjectivity, then you start getting more and more retweets. So in other words, once you start saying very outrageous things, then that probably has more chance of retweeting than just a little bit of bias added to your Twitter. Now, again, this is done only with a very, very, very small sample on a very specific topic. So don't make any generalizations from this, but at least now you know how you can start doing some analysis on Twitter data. Okay, so we just added this. To summarize this, sentiment analysis, it's a very useful thing. It helps us do some analysis on all this data being generated by people, and that is sort of richer in context, richer in meaning. But at the same time, just keep in mind that there are limits to what we can explain from this. So there are more sophisticated natural language processing approaches. Also, one needs to look at a larger context in which this data is generated. Positive doesn't always mean good, negative doesn't always mean bad. But at least now we know where you can begin exploring and how easy it is to do that with Python and the package that we'll see. All right, so that concludes our case study with Twitter data and sentiment analysis.