Commit b6f1e07b authored by Adam Fekete's avatar Adam Fekete
Browse files
parents b90f9d37 363283c8
<!-- <div class="col-xs-8 col-xs-offset-2"> -->
<!-- <div class="col-md-12 text-justify"> -->
The Creedo web application can be used free of charge. In case you
publish research results that have been partially obtained using
Creedo, we kindly request you to acknowledge this project by the
following or similar reference:
<p>@misc{creedo, title = "{Creedo web application}", howpublished=
"\url{https://bitbucket.org/realKD/creedo/}" } Moreover, depending on
the specific contents of Creedo you used to obtain your results,
there are additional specific citation requests outlined in the
following subsections.</p>
<h3>Cumulative Jensen Shannon Divergence</h3>
<p>
<b>Content:</b> Any algorithm for subgroup discovery used with <i>Cumulative
Jensen Shannon Divergence</i> as deviation measure
</p>
<p>
<b>Reference:</b> @inproceedings{pkdd/nguyenV15,
title={Non-Parametric Jensen-Shannon Divergence}, author={Nguyen,
Hoang-Vu and Vreeken, Jilles}, booktitle={ECMLPKDD}, year={2015} }
</p>
<p>
<b>See also:</b> http://eda.mmci.uni-saarland.de/cjs
</p>
<h3>Diverse Subgroup Set Discovery</h3>
<p>
<b>Content:</b> Algorithm <i>Diverse subgroup set discovery</i>
</p>
<p>
<b>Reference:</b> @article{van2012diverse, title={Diverse subgroup
set discovery}, author={van Leeuwen, Matthijs and Knobbe, Arno},
journal={Data Mining and Knowledge Discovery}, volume={25},
number={2}, pages={208--242}, year={2012}, publisher={Springer US} }
</p>
<p>
<b>See also:</b> http://www.patternsthatmatter.org/dssd
</p>
<h3>Direct Pattern Sampling</h3>
<p>
<b>Content</b>: Algorithms <i>2-Step Subgroup Sampling</i> or <i>2-Step
Association Sampling</i>
</p>
<p>
<b>Reference</b>: @inproceedings{kdd/BoleyMG12, author = {Boley,
Mario and Moens, Sandy and G{\"{a}}rtner, Thomas }, title = {Linear
space direct pattern sampling using coupling from the past},
booktitle = {The 18th {ACM} {SIGKDD} International Conference on
Knowledge Discovery and Data Mining, {KDD} '12, Beijing, China,
August 12-16, 2012}, pages = {69--77}, year = {2012} }
</p>
<!-- </div> -->
<!-- </div> -->
<!-- <div class="col-xs-8 col-xs-offset-2"> -->
<div class="col-md-12 text-justify">
<h3>Analytics Dashboards</h3>
<p>When you log in with the admin account that you created on
application setup, you will see a number of demo dashboard links on
this index page. Each of these dashboard showcases different realKD
algorithms on different datasets. In particular, each dashboard
contains (a subset of) the following default realKD algorithms that
can be applied to a different dataset:</p>
<ul>
<li>Association Discovery by Beamsearch</li>
<li>Association Discovery by Direct Pattern Sampling</li>
<li>(Generalized) Subgroup Discovery by Beamsearch</li>
<li>(Generalized) Subgroup Discovery by Direct Pattern Sampling</li>
<li>(Generalized) Subgroup Discovery by <a
href="http://www.patternsthatmatter.org/dssd">Diverse Subgroup
Set Discovery</a>.
</li>
</ul>
<p>
In particular, subgroup discovery can be performed with the novel <a
href="http://eda.mmci.uni-saarland.de/cjs">Jensen-Shannon
divergence</a>.
</p>
<h3>Studies</h3>
<p>
For a demonstration of Creedo user studies, login with your admin
account, open the administration frame, and find under studies the
entry <i>forsied_demo_study</i>. The option <i>State</i> underneath
this entry allows you to toggle between different study states. Choose
the option <i>TRIAL</i> to activate this study.
</p>
<p>Then you can logout and login again with one of the following
user accounts (password: creedo):</p>
<ul>
<li>berta@realkd.org</li>
<li>bob@realkd.org</li>
<li>jane@realkd.org</li>
<li>joe@realkd.org</li>
</ul>
<p>
You will now find a new section of dashboards called <i>Study
Assignments</i> with links corresponding to trial assignments in active
studies. After you participated in the trial phase with all demo
users, you can go ahead and switch the study state to <i>EVALUATE</i>
using an admin account. Subsequently, when you log in with demo users,
you will now see evaluation assignments.
</p>
<!-- <a>project wiki</a> to find out more about how to -->
<!-- configure the contents of the server and to define your own user -->
<!-- studies. -->
</div>
<!-- </div> -->
<!-- <div class="col-xs-8 col-xs-offset-2"> -->
<div class="container">
<div class="row">
<!-- content pic left container start-->
<div class="col-md-4">
<h4>Creedo Analytics Dashboards</h4>
<p>
The most characteristic visual element of a Creedo Analytics
Dashboard is the <b>mine button</b> at the upper right corner,
which allows you to run data mining algorithms. When clicking this
button a dialog will pop up where you can choose one of the data
mining algorithms available in this dashboard along with optional
parameter settings.
</p>
<p>Besides the mine button, the browser window is organized in
three distinct areas:
<ol>
<li>Data View</li>
<li>Candidate Area</li>
<li>Result Area</li>
</ol>
</p>
</div>
<div class="col-md-8">
<img src="file.htm?filename=hlp1uidesc.png" width="100%">
</div>
<div class="col-md-12">
<h2>
<small>Data View</small>
</h2>
<p>
In the data view area you can see the input data table, attributes
and binning metadata, and data visualizations. By clicking on one
of the three tabs on the upper right corner of the data view area,
you can switch to one of four alternate <b>Data Views</b>:
</div>
<div class="row">
<div class="col-md-3 portfolio-item">
<h3>
<small><i class="fa fa-table"></i> Data Table</small>
</h3>
<p>Per default, the data view shows the table tab. If you place
the mouse pointer on the column header, you will see summary
statistics describing the column values in a pop up window.</p>
</div>
<div class="col-md-3 portfolio-item">
<h3>
<small><i class="fa fa-cloud"></i> Point Cloud</small>
</h3>
<p>Analyze your data visually to get first insights and to
support the interpretation of knowledge patterns discovered by
ocm.</p>
</div>
<div class="col-md-3 portfolio-item">
<h3>
<small><i class="fa fa-book"></i> Metadata</small>
</h3>
<p>Switch to the Metadata Tab to learn about the attribute
types and bin classification values computed in the data
preprocessing step.</p>
</div>
<div class="col-md-3 portfolio-item">
<h3>
<small><i class="fa fa-book"></i> Dataset Information</small>
</h3>
<p>Open the Dataset Information Tab to read background
information about the dataset and the analysis tasks our demo
datasets intend to support.</p>
</div>
</div>
<div class="row">
<div class="col-md-3 portfolio-item">
<img src="file.htm?filename=hlp1dvoptstable.png" width="100%">
</div>
<div class="col-md-3 portfolio-item">
<img src="file.htm?filename=hlp1dvoptspoints.png" width="100%">
</div>
<div class="col-md-3 portfolio-item">
<img src="file.htm?filename=hlp1dvoptsmeta.png" width="100%">
</div>
<div class="col-md-3 portfolio-item">
<img src="http://placehold.it/750x450" width="100%">
</div>
</div>
<hr>
<!-- content pic left container start-->
<div class="row">
<div class="col-md-6">
<h2>
<small>Candidate Area</small>
</h2>
<p>
Each box appearing below the mine button represents an <b>candidate
knowledge pattern</b>. Different background colors of pattern boxes
indicate different pattern types.
</p>
<p>
Be aware that the number of displayed attributes in the pattern
description section is restricted. If a description contains four
or more attributes, move the mouse pointer onto the pattern box
and the <b>extended pattern description</b> pop up box will open
and enlist all the pattern propositions and some summary
statistics.
</p>
<p>
<b>Pattern Options:</b><br> <i class="fa fa-search"></i>:
Open detailed pattern view in a new browser tab<br> <i
class="fa fa-screenshot"></i>: Filter data table rows to see only
objects that belong to a pattern<br> <i class="fa fa-remove"></i>:
Remove useless patterns<br>
</p>
</div>
<div class="col-md-6">
<img src="file.htm?filename=hlp1dvcand.png" width="100%">
</div>
</div>
<hr>
<!-- content pic left container end-->
<!-- content pic right container start-->
<div class="row">
<div class="col-md-6">
<img src="file.htm?filename=hlp1dvanab.png" width="100%">
</div>
<div class="col-md-6">
<h2>
<small>Result Area</small>
</h2>
<p>
If you drag a candidate pattern to the analysis board, the content
of the pattern box changes slightly. The previous pattern
description is replaced with the extended pattern description and
a <b>pattern annotation</b> field is added. You can type notes in
the annotation field and they will be added to the detailed
pattern view.
</p>
<p>To remove a pattern from the analysis board, click on the
"Remove Pattern" Button.</p>
</div>
</div>
<!-- content pic right container end-->
</div>
</div>
<!-- </div> -->
\ No newline at end of file
<div class="container">
<!-- <div class="row"> -->
<!-- <div class="col-lg-6 col-md-12 col-sm-12"> -->
<!-- <img class="img-responsive img-home-portfolio" -->
<!-- src="file.htm?filename=creedo_logo_square_margin.png"> -->
<!-- </div> -->
<div class="col-md-12 text-justify">
<p>Creedo is an open source Java server application for data
mining researchers who want to showcase and test their ideas. The
two main features of Creedo are:</p>
<ol>
<li>Creedo allows to define interactive data analytics
dashboards, in which novel data mining algorithm can easily be
embedded if they follow the framework of the realKD library. Users
can access these dashboards through any of the standard web
browsers. Hence, Creedo provides an effective way of defining
demonstrators for new ideas.</li>
<li>Furthermore, Creedo allows the design and execution of
online user studies based on these analytics dashboards. The
application automatically assigns tasks to study participants and
evaluates analytics dashboards in terms of how well they support
users in solving their task. This enables a low-cost extrinsic
evaluation of novel data mining contributions with real users.</li>
</ol>
<p>The development of Creedo was supported by the German Science
Foundation (DFG) under grant number 'GA 1615/2-1'.</p>
<!-- </div> -->
</div>
</div>
This diff is collapsed.
make;categoric;descr
fuel-type;categoric;descr
aspiration;categoric;descr
num-of-doors;categoric;descr
body-style;categoric;descr
drive-wheel;categoric;descr
engine-location;categoric;descr
wheel-base;numeric;descr
length;numeric;descr
width;numeric;descr
height;numeric;descr
curb-weight;numeric;descr
engine-type;categoric;descr
num-of-cylinders;categoric;descr
engine-size;numeric;descr
fuel-system;categoric;descr
bore;numeric;descr
stroke;numeric;descr
compression-ratio;numeric;descr
horsepower;numeric;descr
peak-rpm;numeric;descr
city-mpg;numeric;descr
highway-mpg;numeric;descr
price;numeric;descr
This diff is collapsed.
date;name;date in dd/mm/yyyy format
season;categoric;season (1:springer, 2:summer, 3:fall, 4:winter)
year;categoric;year (0: 2011, 1:2012)
month;ordinal;month ( 1 to 12);1,2,3,4,5,6,7,8,9,10,11,12
holiday;categoric;whether day is holiday or not (extracted from http://dchr.dc.gov/page/holiday-schedule)
weekday;categoric;day of the week
workingday;categoric;if day is neither weekend nor holiday is 1, otherwise is 0
weathersit;categoric;1: Clear, Few clouds, Partly cloudy, Partly cloudy. 2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist. 3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds. 4: Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog
temp;numeric;normalized temperature
atemp;numeric;feeling temperature
hum;numeric;humidity
windspeed;numeric;wind speed
registered;numeric;number of registered bikers
casual;numeric;number of casual bikers
This diff is collapsed.
Combination,name,...,
Delta E,numeric,energy E(RS) - energy E(ZB),
sign(Delta E),categoric,Sign of energy difference,
IP(A),numeric,Ionization potential of element A,
EA(A),numeric,Electronic affinity of element A,
H(A),numeric,HOMO energy level of A,
L(A),numeric,LUMO enegry level of element A,
rs(A),numeric,radius of s-orbital of element A,
rp(A),numeric,rp value of element A,
rd(A),numeric,rd value of element A,
IP(B),numeric,IP value of element B,
EA(B),numeric,EA value of element B,
H(B),numeric,H value of element B,
L(B),numeric,L value of element B,
rs(B),numeric,rs value of element B,
rp(B),numeric,rp value of element B,
rd(B),numeric,rd value of element B,
IP(A)-IP(B),numeric,derived,
EA(A)-EA(B),numeric,derived,
H(A)-H(B),numeric,derived,
L(A)-L(B),numeric,derived,
|rs(A)-rs(B)|,numeric,derived,
|rp(A)-rp(B)|,numeric,derived,
|rd(A)-rd(B)|,numeric,derived,
(IP(A)-IP(B))/IP(A),numeric,derived,
(EA(A)-EA(B))/EA(A),numeric,derived,
(H(A)-H(B))/H(A),numeric,derived,
(L(A)-L(B))/L(A),numeric,derived,
(rs(A)-rs(B))/rs(A),numeric,derived,
(rp(A)-rp(B))/rp(A),numeric,derived,
(rd(A)-rd(B))/rd(A),numeric,derived,
EN(A),numeric,electronegativity of A, --- (-1/2)(EA(A)+IP(A))
EN(B),numeric,electronegativity of B, --- (-1/2)(EA(B)+IP(B))
H(A)-L(A),numeric,HOMO-LUMO gap of A,
H(B)-L(B),numeric,HOMO-LUMO gap of B,
IP(A)-EA(A),numeric,derived,
IP(B)-EA(B),numeric,derived,
H(A)-L(B),numeric,derived,
IP(A)-EA(B),numeric,derived,
(IP(A)-EA(A))/rs(A),numeric,derived,
(IP(B)-EA(B))/rs(A),numeric,derived,
(H(A)-L(B))/rs(A),numeric,derived,
(IP(A)-EA(B))/rs(A),numeric,derived,
(IP(A)-EA(A))/rp(A),numeric,derived,
(IP(B)-EA(B))/rp,numeric,derived,
(H(A)-L(B))/rp(A),numeric,derived,
(IP(A)-EA(B))/rp(A),numeric,derived,
(IP(A)-EA(A))/sd(A),numeric,derived,
(IP(B)-EA(B))/sd(A),numeric,derived,
(H(A)-L(B))/sd(A),numeric,derived,
(IP(A)-EA(B))/sd(A),numeric,derived,
(EA(B)-IP(B))/rp(A)^2,numeric,Feature 1 from Ghiringhelli et al. ,
|rs(A)-rp(B)|/exp(rs(A)),numeric,Feature 2 from Ghiringhelli et al. ,
|rp(B)-rs(B)|/exp(rd(A)+rs(B)),numeric,Feature 3 from Ghiringhelli et al. ,
r_sigma,numeric,|(rp(A)+rs(A))-(rp(B)+rs(B))|,
r_pi,numeric,|rp(A)-rs(A)|+|rp(B)-rs(B)|,
\ No newline at end of file
This diff is collapsed.
1;election results 2005;joint_macro_attribute;5,6,7,8,9
2;election results 2009;joint_macro_attribute;11,12,13,14,15
3;CDU;sequence;5:2005,11:2009
4;SPD;sequence;6:2005,12:2009
5;FDP;sequence;7:2005,13:2009
6;GREEN;sequence;8:2005,14:2009
7;LEFT;sequence;9:2005,15:2009
8;Non-voters;sequence;10:2005,16:2009
9;age distribution;joint_macro_attribute;18,19,20,21,22
10;unemployment;joint_macro_attribute;34,35,36
11;workforce distribution;joint_macro_attribute;37,38,39,40
\ No newline at end of file
Area Name;name;Abbreviated name of district
Type;categoric;Character of district
State;categoric;Federal State
Region;categoric;Geographical region of district
CDU 2005;numeric;Percentage of votes for conservative party 2005 (second vote)
SPD 2005;numeric;Percentage of votes for centre-left party 2005 (second vote)
FDP 2005;numeric;Percentage of votes for liberal party 2005 (second vote)
GREEN 2005;numeric;Percentage of votes for green party 2005 (second vote)
LEFT 2005;numeric;Percentage of votes for left party 2005 (second vote)
Non-Voters 2005;numeric;Proportion of non-voters in Bundestag elections 2005 (w.r.t registered votes)
CDU 2009;numeric;Percentage of votes for conservative party 2009 (second vote)
SPD 2009;numeric;Percentage of votes for centre-left party 2009 (second vote)
FDP 2009;numeric;Percentage of votes for liberal party 2009 (second vote)
GREEN 2009;numeric;Percentage of votes for green party 2009 (second vote)
LEFT 2009;numeric;Percentage of votes for left party 2009 (second vote)
Non-Voters 2009;numeric;Proportion of non-voters in Bundestag elections 2009 (w.r.t registered votes)
Pop. density;numeric;Inhabitants per sq km
Elderly pop.;numeric;Percentage of Inhabitants older than 64 years
Old pop.;numeric;Percentage of Inhabitants between 45 and 64 years
Mid-aged pop.;numeric;Percentage of Inhabitants between 25 and 44 years
Young pop.;numeric;Percentage of Inhabitants between 18 and 24 years
Children pop.;numeric;Percentage of Inhabitants below 18 years
Migration balance;numeric;Population gain per 1000 inhabitants
Social welfare;numeric;Percentage of Inhabitants who receive social welfare benefits
Poor elderly inh.;numeric;Percentage of welfare recipients older than 65 years
Poor child.;numeric;Percentage of welfare recipients younger than 7 years
High-school degree;numeric;Percentage of student with highest school degree among all students that leave school
No school degree;numeric;Percentage of student without school degree among all students that leave school
Child care;numeric;Percentage of children under 3 years w.r.t to all children in day-care
Business Regstr.;numeric;Number of business registrations per 10000 inhabitants
Income;numeric;Available annual income per inhabitant
Employees;numeric;Percentage of employed inhabitants
GPD;numeric;GDP per capita in Euro 2009
Unemployed;numeric;Percentage of unemployed inhabitants
Unemployed Youth;numeric;Percentage of unemployed inhabitants younger than 25 years
Permanently unemployed ;numeric;Percentage of permanently unemployed inhabitants
Agricult. workf.;numeric;Number of employees working in agriculture, foresting, and fishing per 1000 inhabitants (Primary econ. sector)
Ind. workf.;numeric;Number of employees working in producing or manufacturing industry or construction per 1000 inhabitants (Secondary econ. sector)
Service workf.;numeric;Number of employees working in service, trade, gastronomy, financing, rental, and business, or traffic per 1000 inhabitants (Tertiary econ. sector)
Pub. serv. workf.;numeric;Number of employees working in public (and equivalent private) services per 1000 inhabitants
Cars;numeric;Number of cars per 1000 inhabitants
Tractors;numeric;Percentage of tractors w.r.t number of motor vehicles
Registrd. Web Doms.;numeric;Number de-Domains registered per 1000 inhabitants
Bankruptcies;numeric;Number of business bankruptcies per 10000 business tax paying business
\ No newline at end of file
This diff is collapsed.
surgery;categoric;Whether horse was treated with surgery
age;categoric;Whether horse is young (less than 6 month) or adult
rect. temp.;numeric;Rectal temperature of horse (�C), normally 37.8, elevated/reduced may occur due to infection/late shock
pulse;numeric;Heart rate (beats p. min.), 30-40 normal for adults, 20-25 in rare cases of athletic horses, may be elevated with painful lesions or circulatory shock
resp. rate;numeric;"Respiratory rate; normally 8 to 10; usefulness is doubtful due to the great fluctuations"
extremity temp.;categoric;Subjective indication of peripheral circulation, cool to cold indicate possible shock, hot should correlate with elevated rectal temp.
periph. Pulse;categoric;Subjective estimation of peripheral pulse, normal or increased indicative of adequate circulation, reduced or absent indicate poor perfusion
mucous membr.;categoric;Subjective color of mucous membranes, may indicate normal or slightly increased circulation (normal pink, bright pink) or early shock (pale pink) or serious circulatory compromise (cyanotic) or septicemia (bright red / injected)
cap. refill time;categoric;Whether capillary refill time is less than 3 seconds or more, long refill corresponds to poor circulation
pain;categoric;Subjective judgement of pain level, the more painful the more likely surgery is required, prior pain treatment may mask level
peristalsis;categoric;Indication of activity in gut, activity decreases as horse more distended or becomes more toxic
abdom. disten.;categoric;Abdominal distension, important parameter that indicates pain and reduced gut motility, when severe then likely to require surgery to relieve pressure
nasogastric tube;categoric;Whether any gas coming out of tube, large is likely to give horse discomfort
nasogast. reflux;categoric;Nasogastric reflux, the greater the more likely serious obstruction of fluid passage from rest of intestine
nasogast. reflux PH;numeric;PH value of nasogastric reflux, normally between 3-4, range from 0-14 with 7 neutral
rect. examin.;categoric;Rectal examination of feces, absent may indicate obstruction
abdomen;categoric;Can indicate obstruction caused by mechanical impaction (firm feces in the large intestine) or surgical lesion (distended)
packed cell vol.;numeric;No. of red cells by volume in blood, normally 30-50, rises as circulation compromised or animal dehydrated
total protein;numeric;Normally 6-7.5 (gms/dL), the higher value the greater dehydration
abdominoc. appear.;categoric;Abdominocentesis appearance obtained by putting needle in the horse's abdomen and inspecting fluid from abdominal cavity, normally clear while cloudy or serosanguinous indicates compromised gut
abdomcent. protein;numeric;Abdominocentesis appearance (gms/dL), higher is more likely to have compromised gut
outcome;categoric;What happened to horse finally
surgical_lesion;categoric;Whether the problem (lesion) was surgical, determined retrospectively by authopsie if not previously operated
This diff is collapsed.
Month;ordinal;Month;10,11,12,1,2,3,4
HomeAway;categoric;Home or away game
Opponent;categoric;Opposing team
Period;ordinal;Game period;1,2,3,4,5,6
SecondsLeftPeriod;numeric;Seconds left to play in the period
SecondsLeftGame;numeric;Seconds left to play in the game
SecondsPassedPeriod;numeric;Seconds passed in the period
SecondsPassedGame;numeric;Seconds passed in the game
Adam_Morrison;categoric;Adam Morrison on the court?
Andrew_Bynum;categoric;Andrew Bynum on the court?
DJ_Mbenga;categoric;DJ Mbenga on the court?
Derek_Fisher;categoric;Derek Fisher on the court?
Jordan_Farmar;categoric;Jordan Farmar on the court?
Josh_Powell;categoric;Josh Powell on the court?
Lamar_Odom;categoric;Lamar Odom on the court?
Luke_Walton;categoric;Luke Walton on the court?
Pau_Gasol;categoric;Pau Gasol on the court?
Ron_Artest;categoric;Ron Artest on the court?
Sasha_Vujacic;categoric;Sasha Vujacic on the court?
Shannon_Brown;categoric;Shannon Brown on the court?
Type;categoric;Shot type
Assisted;categoric;Shot assisted by a teammate?
AssistBy;categoric;Who assisted a shot
Blocked;categoric;Shot blocked by an opponent?
ReboundBy;categoric;Team that rebounded a missed shot
Fouled;categoric;Extra free-throw due to a foul? (“and-one”)
X;numeric;X-coordinate of the shot location
Y;numeric;Y-coordinate of the shot location
Result;categoric;Made or missed?
\ No newline at end of file
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment