<s>
GroupLens	B-Application
Research	I-Application
is	O
a	O
human	O
–	O
computer	O
interaction	O
research	O
lab	O
in	O
the	O
Department	O
of	O
Computer	O
Science	O
and	O
Engineering	O
at	O
the	O
University	O
of	O
Minnesota	O
,	O
Twin	O
Cities	O
specializing	O
in	O
recommender	B-Application
systems	I-Application
and	O
online	B-Architecture
communities	I-Architecture
.	O
</s>
<s>
GroupLens	B-Application
also	O
works	O
with	O
mobile	O
and	O
ubiquitous	B-Architecture
technologies	O
,	O
digital	O
libraries	O
,	O
and	O
local	O
geographic	B-Application
information	I-Application
systems	I-Application
.	O
</s>
<s>
The	O
GroupLens	B-Application
lab	O
was	O
one	O
of	O
the	O
first	O
to	O
study	O
automated	O
recommender	B-Application
systems	I-Application
with	O
the	O
construction	O
of	O
the	O
"	O
GroupLens	B-Application
"	O
recommender	B-Application
,	O
a	O
Usenet	B-Application
article	O
recommendation	B-Application
engine	I-Application
,	O
and	O
MovieLens	O
,	O
a	O
popular	O
movie	O
recommendation	O
site	O
used	O
to	O
study	O
recommendation	B-Application
engines	I-Application
,	O
tagging	B-General_Concept
systems	O
,	O
and	O
user	B-Application
interfaces	I-Application
.	O
</s>
<s>
system	O
for	O
Usenet	B-Application
news	I-Application
.	O
</s>
<s>
The	O
system	O
collected	O
ratings	O
from	O
Usenet	B-Application
readers	O
and	O
used	O
those	O
ratings	O
to	O
predict	O
how	O
much	O
other	O
readers	O
would	O
like	O
an	O
article	O
before	O
they	O
read	O
it	O
.	O
</s>
<s>
This	O
recommendation	B-Application
engine	I-Application
was	O
one	O
of	O
the	O
first	O
automated	O
collaborative	B-Algorithm
filtering	I-Algorithm
systems	O
in	O
which	O
algorithms	O
were	O
used	O
to	O
automatically	O
form	O
predictions	O
based	O
on	O
historical	O
patterns	O
of	O
ratings	O
.	O
</s>
<s>
The	O
overall	O
system	O
was	O
called	O
the	O
"	O
GroupLens	B-Application
"	O
recommender	B-Application
,	O
and	O
the	O
servers	O
that	O
collected	O
the	O
ratings	O
and	O
performed	O
the	O
computation	O
were	O
called	O
the	O
"	O
Better	O
Bit	O
Bureau	O
"	O
.	O
</s>
<s>
"	O
GroupLens	B-Application
"	O
is	O
now	O
used	O
as	O
a	O
name	O
both	O
for	O
this	O
recommender	B-Application
system	I-Application
,	O
and	O
for	O
the	O
research	O
lab	O
at	O
the	O
University	O
of	O
Minnesota	O
.	O
</s>
<s>
study	O
,	O
in	O
the	O
CSCW	B-Protocol
conference	O
of	O
1994	O
.	O
</s>
<s>
He	O
realized	O
the	O
commercial	O
potential	O
of	O
collaborative	B-Algorithm
filtering	I-Algorithm
,	O
and	O
encouraged	O
the	O
team	O
to	O
found	O
a	O
company	O
in	O
April	O
1996	O
.	O
</s>
<s>
Based	O
on	O
their	O
experience	O
,	O
Riedl	O
and	O
Konstan	O
wrote	O
a	O
book	O
about	O
the	O
lessons	O
learned	O
from	O
deploying	O
recommenders	B-Application
in	O
practice	O
.	O
</s>
<s>
Recommender	B-Application
systems	I-Application
have	O
since	O
become	O
ubiquitous	B-Architecture
in	O
the	O
online	O
world	O
,	O
with	O
leading	O
vendors	O
such	O
as	O
Amazon	B-Application
and	O
Netflix	O
deploying	O
highly	O
sophisticated	O
recommender	B-Application
systems	I-Application
.	O
</s>
<s>
Netflix	O
even	O
offered	O
a	O
$1	O
million	O
prize	O
for	O
improvements	O
in	O
recommender	B-Application
technology	O
.	O
</s>
<s>
more	O
general	O
term	O
of	O
recommender	B-Application
systems	I-Application
.	O
</s>
<s>
With	O
Joe	O
Konstan	O
's	O
expertise	O
in	O
user	B-Application
interfaces	I-Application
,	O
</s>
<s>
the	O
team	O
began	O
exploring	O
interface	O
issues	O
in	O
recommenders	B-Application
,	O
such	O
as	O
explanations	O
,	O
and	O
meta-recommendation	O
systems	O
.	O
</s>
<s>
In	O
2002	O
,	O
GroupLens	B-Application
expanded	O
into	O
social	O
computing	O
and	O
online	B-Architecture
communities	I-Architecture
with	O
the	O
addition	O
of	O
Loren	O
Terveen	O
,	O
who	O
was	O
known	O
for	O
his	O
research	O
of	O
social	O
recommender	B-Application
systems	I-Application
such	O
as	O
PHOAKS	O
.	O
</s>
<s>
performance	O
of	O
online	B-Architecture
communities	I-Architecture
.	O
</s>
<s>
In	O
2008	O
GroupLens	B-Application
launched	O
Cyclopath	O
,	O
a	O
computational	O
geo-wiki	O
for	O
bicyclists	O
within	O
a	O
city	O
.	O
</s>
<s>
In	O
2010	O
,	O
GroupLens	B-Application
won	O
the	O
annual	O
ACM	O
software	O
system	O
award	O
.	O
</s>
<s>
Brent	O
Hecht	O
joined	O
the	O
GroupLens	B-Application
faculty	O
in	O
2013	O
,	O
focusing	O
on	O
geographic	O
human-computer	O
interaction	O
.	O
</s>
<s>
Lana	O
Yarosh	O
joined	O
the	O
GroupLens	B-Application
faculty	O
in	O
2014	O
;	O
she	O
works	O
with	O
social	O
computing	O
and	O
child-computer	O
interaction	O
.	O
</s>
<s>
Stevie	O
Chancellor	O
,	O
a	O
human-centered	O
computing	O
and	O
social	O
computing	O
researcher	O
,	O
joined	O
the	O
GroupLens	B-Application
faculty	O
in	O
2020	O
.	O
</s>
<s>
The	O
MovieLens	O
recommender	B-Application
system	I-Application
:	O
MovieLens	O
is	O
a	O
non-commercial	O
movie	O
recommender	B-Application
system	I-Application
that	O
has	O
been	O
running	O
since	O
1997	O
with	O
over	O
164,000	O
unique	O
visitors	O
as	O
of	O
2009	O
,	O
who	O
have	O
provided	O
over	O
15	O
million	O
movie	O
ratings	O
.	O
</s>
<s>
MovieLens	O
ratings	O
datasets	B-General_Concept
:	O
In	O
the	O
early	O
days	O
of	O
recommender	B-Application
systems	I-Application
,	O
research	O
was	O
slowed	O
down	O
by	O
the	O
lack	O
of	O
publicly	O
available	O
datasets	B-General_Concept
.	O
</s>
<s>
In	O
response	O
to	O
requests	O
from	O
other	O
researchers	O
,	O
GroupLens	B-Application
released	O
three	O
datasets	B-General_Concept
:	O
the	O
MovieLens	O
100,000	O
rating	O
dataset	B-General_Concept
,	O
the	O
MovieLens	O
1	O
million	O
rating	O
dataset	B-General_Concept
,	O
and	O
the	O
MovieLens	O
10	O
million	O
rating	O
dataset	B-General_Concept
.	O
</s>
<s>
These	O
datasets	B-General_Concept
became	O
the	O
standard	O
datasets	B-General_Concept
for	O
recommender	B-Application
research	O
,	O
and	O
have	O
been	O
used	O
in	O
over	O
300	O
papers	O
by	O
researchers	O
around	O
the	O
world	O
.	O
</s>
<s>
The	O
dataset	B-General_Concept
is	O
also	O
being	O
used	O
for	O
teaching	O
about	O
recommender	B-Application
technology	O
.	O
</s>
<s>
MovieLens	O
tagging	B-General_Concept
dataset	B-General_Concept
:	O
GroupLens	B-Application
added	O
tagging	B-General_Concept
to	O
MovieLens	O
in	O
2006	O
.	O
</s>
<s>
Since	O
then	O
,	O
users	O
have	O
provided	O
over	O
85,000	O
applications	O
of	O
14,000	O
unique	O
tags	B-General_Concept
to	O
movies	O
.	O
</s>
<s>
The	O
MovieLens	O
10	O
million	O
ratings	O
dataset	B-General_Concept
also	O
includes	O
a	O
100,000	O
tag	B-General_Concept
applications	O
dataset	B-General_Concept
for	O
researchers	O
to	O
use	O
.	O
</s>
<s>
Information	O
leakage	O
from	O
recommender	B-Application
datasets	B-General_Concept
:	O
a	O
paper	O
in	O
the	O
information	O
retrieval	O
conference	O
analyzed	O
the	O
privacy	O
risks	O
to	O
users	O
of	O
having	O
large	O
recommender	B-Application
datasets	B-General_Concept
released	O
.	O
</s>
<s>
The	O
basic	O
risk	O
discovered	O
is	O
that	O
an	O
anonymized	O
dataset	B-General_Concept
might	O
be	O
combined	O
with	O
public	O
information	O
to	O
identify	O
a	O
user	O
.	O
</s>
<s>
For	O
instance	O
,	O
a	O
user	O
who	O
has	O
written	O
about	O
his	O
preference	O
for	O
movies	O
on	O
online	B-Application
forums	I-Application
could	O
be	O
associated	O
with	O
a	O
specific	O
row	O
in	O
the	O
MovieLens	O
datasets	B-General_Concept
.	O
</s>
<s>
GroupLens	B-Application
has	O
also	O
explored	O
ways	O
to	O
help	O
editors	O
find	O
pages	O
which	O
they	O
can	O
effectively	O
contribute	O
to	O
with	O
the	O
SuggestBot	O
recommender	B-Application
.	O
</s>
<s>
GroupLens	B-Application
has	O
also	O
explored	O
the	O
functioning	O
of	O
the	O
informal	O
peer	O
review	O
system	O
within	O
Wikipedia	O
to	O
discover	O
ways	O
the	O
decisions	O
being	O
made	O
appear	O
to	O
be	O
influenced	O
inappropriately	O
by	O
ownership	O
,	O
and	O
that	O
experience	O
does	O
not	O
seem	O
to	O
change	O
editor	O
performance	O
very	O
much	O
.	O
</s>
<s>
GroupLens	B-Application
researchers	O
have	O
also	O
explored	O
visualizations	O
of	O
the	O
edit	O
history	O
of	O
Wikipedia	O
articles	O
.	O
</s>
<s>
In	O
2011	O
,	O
the	O
GroupLens	B-Application
researchers	O
completed	O
a	O
scientific	O
exploration	O
of	O
gender	O
imbalance	O
in	O
Wikipedia	O
's	O
popular	O
editors	O
,	O
resulting	O
in	O
finding	O
that	O
there	O
was	O
a	O
large	O
gap	O
between	O
male	O
and	O
female	O
editors	O
.	O
</s>
<s>
Shilling	O
recommender	B-Application
systems	I-Application
:	O
GroupLens	B-Application
has	O
explored	O
ways	O
that	O
users	O
of	O
recommender	B-Application
systems	I-Application
can	O
attempt	O
to	O
inappropriately	O
influence	O
the	O
recommendations	O
given	O
to	O
other	O
users	O
.	O
</s>
<s>
Cyclopath	O
:	O
Beginning	O
in	O
2008	O
,	O
GroupLens	B-Application
launched	O
Cyclopath	O
,	O
a	O
computational	O
geo-wiki	O
for	O
local	O
bicyclists	O
.	O
</s>
