basic reputation monitoring system by extending the sentiment classifier that you just saw. So here's what I hope you do. For each of the four code cells below, to click on the cell, then hit shift-enter or the play button, if on a mobile device, to run the code. Let's go to the first code cell. I'm going to hit shift-enter. This is the same code as the previous activity, where it sets up the program environment to send prompts to OpenAI. The second code cell creates a list of reviews. Here we have five reviews, the mochi is excellent, best soup dumplings I've ever eaten, and so on. I'm going to hit Shift-Enter. Now, code cell 3 contains some code to classify each of the reviews as positive or negative. There's a bunch of code here, but you don't need to worry about the details. The one part to pay attention to is this part, We're here, we're using the prompt, classify the following review as having either positive or negative sentiment, say your answer is a single word, either positive or negative, and then the software will put in each of the five reviews into this prompt in turn. So when we run it, it'll take a few moments to run because it's going through all five reviews with this prompt. We end up with these classifications, positive, positive, negative, positive, negative, which I think is correct, corresponding to the five reviews we saw earlier. And then finally, here's another snippet of code. Don't worry about the details, but I'm just going to run it, that will count how many positive and how many negatives there are. And in this case, it says that three positive and two negative reviews, which I think is actually correct. If you know a little bit of coding, feel free to read through this code to see if you can understand what it's doing. But it's not important for the purposes of this exercise to read the details of the code. But you can imagine if instead of five reviews, we had a 100 reviews. The same code can go through a 100 reviews, and classify all of them, and tell you how many of them are positive or negative. In fact, let me change one review. Instead of the parsers code, I'm going to say the parser was amazing. Then, if I rerun code cell 2, just hit Shift Enter, code cell 3. Now it generates the new classifications, positive, positive, negative, positive, positive, and rerun code cell four. Now it says that four positive and one negative review. So please run the four code cells yourself. And after that, go ahead and change these reviews to something else if you wish, and rerun the code and see what results you get. So if you were able to run the code in this exercise, I want to say congratulations. You just ran some code to run a pretty non-trivial AI function for the pretty complex task, frankly, of looking at words to classify restaurant reviews. I hope you had fun with that. Feel free to play around with the code more, and when you're done, let's go back to the Coursera platform to continue the rest of this course. I look forward to seeing you there.