Suppose you are a teacher, and have to convey a set of object-property pairs ('lions eat meat'). A good teacher will convey a lot of information, with little effort on the student side. What is the best and most intuitive way to convey this information to the student, without the student being overwhelmed? A related, harder problem is: how can we assign a numerical score to each lesson plan (i.e., way of conveying information)? Here, we give a formal definition of this problem of forming learning units and we provide a metric for comparing different approaches based on information theory. We also design an algorithm, groupNteach, for this problem. Our proposed groupNteach is scalable (near-linear in the dataset size); it is effective, achieving excellent results on real data, both with respect to our proposed metric, but also with respect to encoding length; and it is intuitive, conforming to well-known educational principles. Experiments on real and synthetic datasets demonstrate the effectiveness of groupNteach.
%0 Book Section
%1 citeulike:14087160
%A Hooi, Bryan
%A Song, HyunAh
%A Papalexakis, Evangelos
%A Agrawal, Rakesh
%A Faloutsos, Christos
%B Advances in Knowledge Discovery and Data Mining
%D 2016
%E Bailey, James
%E Khan, Latifur
%E Washio, Takashi
%E Dobbie, Gill
%E Huang, Joshua Z.
%E Wang, Ruili
%I Springer International Publishing
%K electronic-textbook text-analysis
%P 376--387
%R 10.1007/978-3-319-31750-2_30
%T Matrices, Compression, Learning Curves: Formulation, and the GroupNteach Algorithms
%U http://dx.doi.org/10.1007/978-3-319-31750-2_30
%V 9652
%X Suppose you are a teacher, and have to convey a set of object-property pairs ('lions eat meat'). A good teacher will convey a lot of information, with little effort on the student side. What is the best and most intuitive way to convey this information to the student, without the student being overwhelmed? A related, harder problem is: how can we assign a numerical score to each lesson plan (i.e., way of conveying information)? Here, we give a formal definition of this problem of forming learning units and we provide a metric for comparing different approaches based on information theory. We also design an algorithm, groupNteach, for this problem. Our proposed groupNteach is scalable (near-linear in the dataset size); it is effective, achieving excellent results on real data, both with respect to our proposed metric, but also with respect to encoding length; and it is intuitive, conforming to well-known educational principles. Experiments on real and synthetic datasets demonstrate the effectiveness of groupNteach.
@incollection{citeulike:14087160,
abstract = {{Suppose you are a teacher, and have to convey a set of object-property pairs ('lions eat meat'). A good teacher will convey a lot of information, with little effort on the student side. What is the best and most intuitive way to convey this information to the student, without the student being overwhelmed? A related, harder problem is: how can we assign a numerical score to each lesson plan (i.e., way of conveying information)? Here, we give a formal definition of this problem of forming learning units and we provide a metric for comparing different approaches based on information theory. We also design an algorithm, groupNteach, for this problem. Our proposed groupNteach is scalable (near-linear in the dataset size); it is effective, achieving excellent results on real data, both with respect to our proposed metric, but also with respect to encoding length; and it is intuitive, conforming to well-known educational principles. Experiments on real and synthetic datasets demonstrate the effectiveness of groupNteach.}},
added-at = {2018-03-19T12:24:51.000+0100},
author = {Hooi, Bryan and Song, HyunAh and Papalexakis, Evangelos and Agrawal, Rakesh and Faloutsos, Christos},
biburl = {https://www.bibsonomy.org/bibtex/24bff5ce24dc9dd5db646315351dbb479/aho},
booktitle = {Advances in Knowledge Discovery and Data Mining},
citeulike-article-id = {14087160},
citeulike-linkout-0 = {http://dx.doi.org/10.1007/978-3-319-31750-2_30},
citeulike-linkout-1 = {http://link.springer.com/chapter/10.1007/978-3-319-31750-2_30},
doi = {10.1007/978-3-319-31750-2_30},
editor = {Bailey, James and Khan, Latifur and Washio, Takashi and Dobbie, Gill and Huang, Joshua Z. and Wang, Ruili},
interhash = {56cfd6fd9b4fcd51688565c3d2077088},
intrahash = {4bff5ce24dc9dd5db646315351dbb479},
keywords = {electronic-textbook text-analysis},
pages = {376--387},
posted-at = {2016-06-30 14:54:07},
priority = {2},
publisher = {Springer International Publishing},
series = {Lecture Notes in Computer Science},
timestamp = {2018-03-19T12:24:51.000+0100},
title = {{Matrices, Compression, Learning Curves: Formulation, and the GroupNteach Algorithms}},
url = {http://dx.doi.org/10.1007/978-3-319-31750-2_30},
volume = 9652,
year = 2016
}