 When we were kids, we used to solve problems using hard-loop, like this piano playing robot. It uses foam motors to hit a dinosaur, which moves around in the water and pulls a little piece of plastic. It's also good for music and other things. In short, it's a hard loop. We switched to AI, presenting science art. In which you would normally find in a science-sense paper. In case in place, I was able to attract a steel plate without increasing the number of chains of wire around the eyebrow. So that's what things we could do to present it to increase the number of steel paper plates attached to the iron rod. Do you know the answer? It's something. This is why we believe in the future. Take comfort. Now, explain. Explain. Magnificent strength. Do you think this requires a type of secret language? This has different types of keywords and phrases, like you saw in the example application, which automatically marks science open-ended with questions. Students can get unlimited practice to learn outcomes like science. Such as Khan Academy, Cheney Book, MC Online, and assessment books. It gets teachers to market by showing your short demo of science. Sizer and take a test. So scientists are going to give them a series of, a selection of tests, click based on science, and issue them questions from its database. So we're going to answer all these questions. Quickly, please. You'll also see that we implemented a chat forum to Sizer, which is going to be used for this film. In the meantime, if you don't want to say, if you don't mind, you can take the mic. That's too far. We'll have to look. What science does, it just looks at these, it looks at the database it has, it points to that database, it takes these answers, and it marks it. So with you, we don't see that we got correct wrong, wrong and wrong, because we didn't answer the last one. So now students want to know, why am I wrong? So they click on the button, click that. And it's loading. Right again, right again. Refresh, refresh. So click on the wrong. Then they'll be able to see all the bottom answers in the database. And they'll be able to understand why the answer was not wrong and how to improve on the answer. But let's say, I don't think this answer was wrong. I think Sizer is stupid. So what would the students do? They click over here on these little check boxes to dispute the answer. And then it goes on and it will continue. Ah, they mess up. So now, if I launch it as a teacher or an admin, we can validate the feedback and the disputes that the students gave us. Validating that. So over here, we'll see all the disputed answers in the system. You can see that the answer that we just disputed was at the bottom. As a teacher, you can click here and see the bottom answers. Then you can choose to either mark it correct, wrong, or ignore it. So that's pretty much what Sizer does. Next. That's okay. The technology used in real-worlds is the AI, Python, the brain, HTML, the front end, and Flash, which connects Python and HTML. This is a flowchart of what Sizer does. First, the student marks it, receives the test, select the test, receives the question, and answers the question. This is pretty much what I just showed you in the demo, but more thought-out. Then Sizer checks the answer against modern artists for this abusive IBM Watson's natural language recession. Then it gives the student this cause. This student gives feedback, like clicking the checkboxes sign, the answers that feedback for the teachers who validate it, and push it back to IBM Watson's natural language recession. Here are some future technical developments that we can do. We can scan text books to get new questions and answers. This saves the pain of typing out all the new questions and answers. We can use it for solace to get more correct answers, because there are so many different words, and we can't use all of them. Normal, what makes Sizer do the level 20? We are meeting teachers and talking to MOE. We collaborate with people such as front-end developers so that they can make our app look beautiful, and teach our students and characters and give us feedback. We need teachers, especially inside teachers, so we have a bigger database, and Sizer can be more accurate. You can get in touch with us after the presentation. You can email me at CuriousMovie. You can access Sizer at Http.org Yes, 47. Then we use it for more time. Do it. How to? Then no, we don't have time. 37 seconds. You can do it. We don't have time. Go for it. It's going to take some time. Wait, is the internet even connected? It can. It's kind of weird. Can we do it? No. It's all right. You can show the video later. People are hanging out.