we're going to do is we're going to see one in practice. So on the right hand side I have logged in as Jennifer who if I make the screen bigger you can see that she has only taken one course called how social and the clinical inequity create patients. I'm going to park it on the right hand side. On the left hand side I also have this course open and you can see that here's a list of courses and you will see that there are four courses that have shared three common learners. So I'm going to just click on my courses again to refresh but because we have recommendation turned on there's this new option on the user navigation on the left hand side that says recommend the courses. If I click into it you will see that there's quite a few courses. As a matter of fact all these courses are based on what is listed here at the top four one two three four is basically what these ones are. Obviously like I mentioned earlier if there are the you know the tiebreaker is a lifetime but if there is no tie if everything's really just tied then you know these are just going to be listed. I won't say random but I'm undeterministic right so but as you can see these are the four courses that's listed right there is the diabetes in native hawaiian this is number one on the mm view it ends up being number three on the learner view because technically the top four are all tied right so this is how the user the user view will look like. Now what I will do is let's say as Jennifer Rambo I'm going to do the course that's called sudden rise and fall in telehealth. So let's say I do this course oh it looks like it's a course that doesn't have any content so I can just claim credit. Okay so that's just a silly setup but with that said what I will do now is I'm on the left hand side I'm just going to refresh the screen go to related and it is important to notice that Jennifer Rambo finished the course but the relationship here did not update you expect that this one will become four. The reason is that the recommendation logic is not constantly being updated as the user I would say the recommendation metrics is not constantly updated as the learner finished courses because if we constantly update the system the recommendation logic is just going to be constantly changing so what we do is we actually make we have a daily process that essentially update all the counts once per day just so that the recommendation is not constantly changing and the system is not overwhelmed when a lot of people are doing courses because if we constantly update the recommendation matrix it will just cause performance issue. So just keep that in mind the recommendation is essentially being updated on a daily basis instead of on a real-time basis. Thank you.

course recommendation

learning platform

system efficiency

daily updates

user experience