 Welcome back to LA Tools course. In this week we will talk about Introduction to Multimodal Learning Analytics. Multimodal learning analytics itself can be a separate course because we collect data from different channels or different modalities then we need to sync them and apply learning analytics on them. But in this week I try to motivate you what is multimodal learning analytics or at least how to collect data from different channels and use them to provide a feedback, affinity feedback or personalized content to the learners. So, you have seen how to collect data from different learning environments like a classroom, online or technology and learning environments in a week 2 and you know you would have used those data for solving some assignments or while reading to apply machine learning models on the data. Now, the question is the data we collected from this log data or any data, is that enough? Can we use that data to you know completely give a holistic approach of learners behavior, is that possible? Because what data we collected in a classroom or online is students interaction with some particular system. For example, in a classroom if you collect data from Moodle you might be collecting data from students interaction with a Moodle or if you are collecting data from students observation or human observation that is perfect, that is good. But is there any more data we can collect to understand the students behavior other than log data? So, assume that you are working on a tele-based search environment and study and you can collect log data in a tele like Moodle we discussed and you can collect a lot of data and what are other sources of data we can collect other than the students interaction with the system like the students clicks and clicks and actions they do. Other than that can we collect some other data to talk about learners, learners learner behavior to model them? If yes, what are the other data we can collect and how do you collect it? For example, I will say that we can collect learners facial expressions by using web camera. So, think of it, think about it other possible data channels and what data you can collect. For this video write down your answers at least write down 3 or 2 channels of data and what data you are collecting then assume the video to continue. So, multi-channel data is like let us say I call the sensors or the data coming to a channel then I call the multi-channel data antics and sensors. So, we can use EEG signal analysis to measure the learners cognition, the brain waves or you can use web cam or the camera to detect students facial expressions. By using facial expressions we can see their emotions or we can capture the discussion with the collaborator or if the student is thinking allowed capture those and the noise the learner makes or the discussion they interact with the collaborator capture those and those can be used to code them and you can do the discourse analysis on that. Or you can use wall mounted biological sensors like if you enter a room there are a lot of sensors in the side panels in the walls that tells the classroom environment like the temperature for all the students or they are they getting the postures different all these things can be observed or we can use eye gaze data for example if the student is interacting with you know computer or laptop or with some other object where are they looking at what are their focus are they looking at a particular place more actually are they looking at a figure like a figures to understand it better are they really reading something or just skimming through it we can get those data from eye trackers log data can provide such as the student is in this particular page and is reading or looking at a figure we are not sure what actually the students is doing with eye gaze data we can tell that the students spend most time on you know trying to understanding from the figure and also reading this sentences and instead of reading a sentences the student might be like a skimming through it going fast and checking it out not really reading. So, these data will add you know additional to the log data this is very important and you can also use GSR to measure the skin conductance like the students attention or cognition can be detected and we can use kinet kind of devices which takes you know camera and multiple two cameras actually so and it tries to model your posture and gesture. So, yeah these are the other ways to do that. Here is the sample eye gaze and GSR data that is collected on a tell environment student is looking at the eye gaze and it is student is looking at a different parts of the environment and you can see the eye gaze is overlaid on the video below the students GSR data. So, in this slide what I try to do is that to motivate that there are data we can collect from different channels and collecting this data from different channels will help you to model the learner better. Thank you.