 Let us see application that is a couple of research papers in using NLP. We are doing it for last few weeks. So, let us do a one research paper this week. So, for automated grading, there is a paper from JMC Lester s group. It is accessing elementary students science competency with test antics. Let us look at this paper, it is 2014. So, I just tried to pick up a paper which is bit old so that we can see the automated grading system using bag of words approach. So, let us look at this paper, assessing elementary students science competency with test antics. I request you to go ahead and read the paper. It has very interesting things called write eval. It is just a word they use for this particular grading systems. So, there are nine students answer, how can you make the two bulbs in a serious circuit brighter? So, the student has answered answers here and the grading also then probably correct, incorrect, correct. So, average, poor, better, good, something like that. So, they have a sentences. So, from these sentences, they form the set of dictionary from that they can able to identify the grading systems. So, let us go and see. So, it is not simply that grading system only based on the bag of word approach what they did they out of from combination, for example, they find the similarities scores of answers between the correct answer and the given answer by the student. The student answer and the human graded answer is compared to find the similarities. The correct answers partially correct, incorrect, the statistics is used and also the students answers from the bag of word and the reference answers. Everything is being used to create into a classifier that classifier grades the output. So, it is the basic approach is what we saw that in a bag of words approach with under students writing has you create a big dictionary and use those dictionary to create a feature set and you are grading as a label and you train the system. Similar, but it also use the similarity syntaxes and other indexes what we saw in this week's course. So, check this video and they used VECA, they used Descent Tree, check this video and try to understand this paper. Hope you will able to do that. That is that is all I wanted to tell you about. There is also another paper on a MOOC data analytics. So, it is based on students response to the forums in the MOOC. We were seeing in the last weeks, we constructed features based on the interaction between the forums like how many times they answered, how many times they apported, how many active actions. These particular concepts that plus the content they are writing, what are they writing. So, that is where the NLP is more interesting here, not just the features you can extract also the content you are writing also part of it. So, check this paper, I am not going to look at this paper. It is interesting paper, check this paper. It is from the LAC 2016, check this. Given this knowledge of basic concepts of NLP and the bag of words similarity or Ngram all this information, can you think of a two applications of NLP in learning environments. You might have had a question why we are talking about NLP in learning analytics. That one week course, I just wanted to introduce what is NLP because it is very, very important to understand the content analytics. Can you think of two applications of NLP in learning analytics and write it down. Once you write it down, please watch the video to continue. So, I just listed very few. This is based on my interaction with some of the students and other participants when I go out and present this talk about NLP or just analytics to the students or other participants. And these are the answers, there are a lot of more answers can be possible. Like quality assessment of the content, you can check whether the content is really quality or not. Now, you know part of speech, you know the identifiers, whether the sentence is written correctly or not, whether the sentence has a proper sentence, the quality of writing, the spelling mistakes, everything can be automatically identified. And that is interesting. And intelligent tutoring systems can adapt based on the text input feedback from the students. Till now I was talking about intelligent tutoring systems can adapt feedback and content based on the students interaction, whether clicks or their responses in the assessment. But they are writing a feedback, they are writing in forum that can be also used to provide a content feedback that it is very interesting, but nobody has run it, it is interesting one. Interesting suggestion by one of the participants where I presented this. And assistive systems for report generation, that is very important. And you can also use this to detect plagiarism and exact that was happening in a plagiarism detectives like turn 18, all the existing plagiarism detection softwares use natural language processing. And information extraction for education, it has not been done quietly, not very popular, extracting the information of education context. IE has been done for movie industries, other industries banking, marketing, all these things. But for education, you might be able to create a concept map of concepts you are teaching in your education, your domain or in your topic, all these things can be done automatically. That is also interesting to see, very interesting application areas of NLP, most of them are still in infant level, infancy level. So, go ahead and try if you are interested in NLP and learning environment. So, in this video is what is NLP application, but this week we saw test analytics and why NLP is important for education. Thank you.