馃攳
Week 1-Lecture 4 : Four Levels of Learning Analytics Overview -II - YouTube
Channel: unknown
[0]
[Music]
[15]
welcome back to learning analytics
[17]
course so in the last video you have
[21]
performed a diagnostic analytics such as
[23]
why a student did not pass the exam
[25]
where the student was not able to answer
[27]
certain questions you collected the data
[31]
such as social performance also the
[34]
attendance now I want you to think how
[37]
will you predict the students
[39]
performance in the future events for
[41]
example you found a correlation that
[44]
students attendance and performance of
[46]
high correlation can you predict what
[48]
will happen in future please pass this
[53]
video write down your answers after
[55]
adding it down resume the video to
[57]
continue so what is predator antics so
[66]
it's predicting what will happen next
[69]
so which course will have a less number
[72]
of registrations its kind of predictions
[75]
because based on the last few stress
[77]
Oracle data we can't read it - what'll
[79]
happen in next year which students will
[82]
not complete the course based on the
[84]
students attendance or students of the
[87]
interaction behavior you can predict
[89]
whether the student can complete the
[90]
course or not and this again using the
[93]
model which you created using the
[94]
historical data what will be the
[97]
performance of the student in the next
[99]
question so you can go much finer level
[102]
like if he is interacting with the
[104]
indigent learning environment or
[106]
technology enhanced learning environment
[108]
what will be the students next set of
[110]
actions will the student able to answer
[112]
the next question correctly all those
[114]
finer level analysis we can do so
[117]
prediction is done based on data from
[121]
past events in the sense if you are
[125]
teaching a class in this year you cannot
[129]
predict create a model using current
[132]
data instead you have to use that
[134]
from last four or five years Khadijah
[135]
like as I mentioned you are last five
[138]
years data from that last historical
[141]
data you might be creating the model
[144]
applied that model in this year that is
[147]
what it says that you have to create the
[149]
model based on past events also in some
[152]
cases we will use the present data also
[155]
because your that will help to improve
[159]
the algorithm to predict it better it's
[162]
most popular in educational data mining
[164]
it's called learning modeling we try to
[167]
model the learner and also it's popular
[170]
in learning analytics there are lot of
[173]
tools available for teachers and other
[175]
stakeholders to perform the predictive
[177]
analytics however we need to understand
[179]
what algorithm to apply and what data
[183]
will suitable for it and you have to
[184]
understand how to interpret the results
[186]
when the algorithm gives so we will talk
[189]
about nabe a station tree in this class
[191]
and the tools like or angelica to use
[196]
this algorithm so on education data so
[201]
in short prior to analytics as extracts
[204]
information from datasets in order to
[207]
detain pattern and predict the future
[209]
events it uses both past and present
[212]
data and offer prediction for the future
[215]
so present data actually so input
[218]
variable to the model and those model
[222]
predict what will happen in the future
[223]
given the present data so let's move on
[227]
to the last activity of this way now
[230]
that you are able to predict whether the
[233]
learner will continue your course or
[235]
drop it assume that what measures will
[238]
you take to ensure your learner's are
[241]
motivated enough to continue a course
[243]
and out for example assume that you have
[249]
created a correlation between attendance
[252]
and performance and you know that
[254]
learner correlation is high with
[257]
attendance using that information you
[260]
can predict whether the learner will
[263]
drop the course or whether the learner
[267]
fail in the exams if you have that
[270]
prediction what measures will you take
[273]
in order to motivate the students and
[276]
continue doing the course please pass
[279]
this video write down your answers after
[281]
writing it down resume this video to
[284]
continue prescriptive in text is if you
[294]
say that I want scaffold to students
[297]
give hints of feedbacks to help students
[301]
to achieve the learning goal that can be
[303]
one of the prescribed antics or you can
[306]
use a personalized or indigent learning
[308]
environments where based on the students
[309]
interaction and performance the system
[312]
gives the feedbacks and ends and adapts
[314]
the content or you can predict the
[317]
learners current state and provide
[320]
feedback and add them achieve the
[322]
learning goal so you can have special
[325]
classes and you can talk to them what is
[327]
the problem it's not that student will
[332]
fail or children will drop out you won't
[334]
understand why also why the student will
[337]
not able to complete because if you know
[340]
why the reasons then you will be able to
[342]
provide a proper inform feedback or
[344]
adaptive content so the prescriptive
[351]
analytics is a process of analyzing the
[353]
data and providing instant
[355]
recommendation to outlook optimized the
[358]
learners learning process example for
[363]
example in instructor can use a
[366]
prescriptive analytics to discover that
[368]
most of the learners needed Priya
[370]
couscous before joining a newly launched
[374]
Advanced Course so based on the previous
[378]
experience or pressed on the teaching
[380]
experience you can so say that further
[383]
newly launched advanced course given the
[386]
learners entry exam you might need a
[390]
previous course so the teacher might
[394]
conduct a test before learner taking the
[397]
course based on the students performance
[399]
a teacher
[400]
can decide whether the learner needs a
[403]
previous tour not so that's best of
[406]
teaching experience or the predatory
[407]
model you are created I want to inform
[411]
you that we saw the four types of
[413]
analytics in last couple of videos so
[417]
it's a if we talk about prescriptive it
[420]
subsumes predict you diagnose you and
[424]
descript antics which means if I doing a
[428]
predictive analytics I should be doing
[430]
the descriptive and diagnostic analytics
[432]
also it's not that I can pick and do
[435]
only the predator analytics I can
[437]
frequently do the prescriptive analytics
[439]
so you can start always with the
[441]
descriptor antics then do the diagnostic
[443]
antics then go for the predator antics
[445]
in this course we'll cover the three
[447]
types of analytics like descriptive
[449]
diagnostic and predictive prescriptive
[452]
analytics is beyond the scope of this
[454]
course well because prescriptive antics
[456]
is designing an intelligent tutoring
[458]
systems so in this video we described
[462]
what is predictive analytics and what is
[465]
prescriptive analytics and you might
[467]
have add some idea on that we will
[469]
detail discuss about this two type of
[472]
analytics in our course thank you
[475]
[Music]
Most Recent Videos:
You can go back to the homepage right here: Homepage