Post Doc position in France (Brittany, Lannion), recommendation engines

"frankmf.meyer" <frankmf.meyer-/[email protected]> Fri, 18 Jul 2008 09:23:51 -0000
Newsgroups gmane.comp.information-retrieval.webir
Message-ID <[email protected]>
--rkoepZNBw6egbz5bKsNGVo6M--tNAVtby7P3WOK
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

France T=E9l=E9com R&D
offers a 1 year post doc position at France Telecom in Lannion=20
(Brittany, France)
2008/2009


Field of research
Recommendation engines: hybrid and anytime engines

Key words
Data Mining, Collaborative filtering, Recommendation Engine, Content-
based Recommendation

Context
Recommendation engines are well-known systems used on many=20
commercial websites (see annexe). They offer recommendations such=20
as "people who liked/bought this item also liked/bought these=20
items", or "we predict that you will rate this item 4 stars."

Recommendation engines use 2 different techniques:

o	Techniques called collaborative filtering: these techniques=20
use the purchase histories or user browsing logs with algorithm such=20
as "basket analysis" (association rules) or methods such as k-
nearest neighbours.
o	Techniques called content-based filtering: in such a case=20
one seeks to build a thematic profile (for instance, using key-
words) which enables the system to score each item. The list of the=20
scored items is then presented to the user.

The subject on the recommendation engines has two main axes:

1.	Hybrid engine

We are looking for the best of both worlds: an engine which can work=20
with both collaborative filtering and content-based filtering whilst=20
still providing good predictive and scalability performances.

2.	"AnyTime" engine

We are looking for an algorithm and a framework architecture which=20
enables the recommendation system to work "anytime" (the model=20
updates are in real time or incremental).
Research Work


The research work will have 3 parts:

1. State of the art
First of all, you will have to make a state of the art of the=20
different technologies of hybrid or incremental recommendation=20
engines: types of learning method used, type of data used, bench and=20
performances=85.
=20
2. Algorithms and evaluation
Next you will have to define one or more innovative recommendation=20
engine algorithms suited to the aforementioned constraints.

3. Communication
This work will consist in publishing international articles and/or=20
develop patents.

Type of position and location
-	A one year temporary work contract for FT R&D (Orange Labs)
-	working in a research team of about 15 people
-	department: France Telecom / R&D / Tech / Easy / Tsi
-	location : Lannion (Brittany, C=F4tes d'Armor, France)
-	salary: approx 32 K Euros

Education and skills required
-	knowledge in machine learning techniques
-	PhD in Computer Science, if possible in AI, Statistics,=20
Machine Learning or Signal Processing.
-	skill in programming: knowledge and experience in Java
NB: CVs which not correspond to this profile will not be considered

Useful supplementary skills
-	experience in the field of recommendation systems
-	experience in the field of Information Retrieval
-	Knowledge in Databases (MySQL=85)
-	experience of working in a team

Application deadline

Application for this position will be accepted until November 2008=20
at the latest.

Contact

Frank Meyer
France Telecom R&D/TECH/EASY (LD128)
2 avenue Pierre Marzin 22307 Lannion Cedex=20
E-mail : [email protected]
Telephone : +33 (0)2 96 05 28 89
http://www.francetelecom.com/rd



R=E9f=E9rences
-	http://en.wikipedia.org/wiki/Collaborative_filtering
-	http://en.wikipedia.org/wiki/Recommendation_system
-	NetFlix Challenge :  http://www.netflixprize.com/


(Adomavicius & Tuzhilin, 2005) Adomavicius, G., & Tuzhilin, A.=20
(2005). Toward the next generation of recommender systems: A survey=20
of the state-of-the-art and possible extensions. IEEE Transactions=20
on Knowledge and Data Engineering, 17(6), 734=96749.

(Bell et al., 2007) Bell, R., Koren, Y., & Volinsky, C. (2007).=20
Modeling relationships at multiple scales to improve accuracy of=20
large recommender systems. In 13th ACM SIGKDD International=20
Conference on Knowledge Discovery and Data Mining (pp. 95=96104). New=20
York, NY, USA. ACM.

(Candillier et al., 2007) Candillier, L., Meyer, F., & Boull=E9, M.=20
(2007). Comparing state-of-the-art collaborative filtering systems.=20
In Perner, P. (Ed.), 5th International Conference on Machine=20
Learning and Data Mining in Pattern Recognition (pp. 548=96562),=20
Leipzig, Germany. Springer Verlag.

(Herlocker et al., 2004) Herlocker, J., Konstan, J., Terveen, L., &=20
Riedl, J. (2004). Evaluating collaborative filtering recommender=20
systems. In ACM Transactions on Information Systems, 22(1), 5=9653.

(Melville et al., 2002) Melville, P., Mooney, R., & Nagarajan, R.=20
(2002). Content-boosted collaborative filtering for improved=20
recommendations. In 18th National Conference on Artificial=20
Intelligence (pp. 187-192).

(Polcicova et al., 2000) Polcicova, G., Slovak, R., & Navrat, P.=20
(2000). Combining content-based and collaborative filtering. In=20
ADBIS-DASFAA Symposium (pp. 118-127).





--rkoepZNBw6egbz5bKsNGVo6M--tNAVtby7P3WOK
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable


<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/htm=
l4/strict.dtd">
<html>
<head>
</head>




<body style=3D"background-color: #ffffff;">

<!--~-|**|PrettyHtmlStartT|**|-~-->
<div id=3D"ygrp-mlmsg" style=3D"width:655px; position:relative;">
<div id=3D"ygrp-msg" style=3D"width: 470px; margin:0; padding:0 25px 0 0; f=
loat:left; z-index:1;">
<!--~-|**|PrettyHtmlEndT|**|-~-->

    <div id=3D"ygrp-text">
            <p>France T=E9l=E9com R&amp;D<br>
offers a 1 year post doc position at France Telecom in Lannion <br>
(Brittany, France)<br>
2008/2009<br>
<br>
Field of research<br>
Recommendation engines: hybrid and anytime engines<br>
<br>
Key words<br>
Data Mining, Collaborative filtering, Recommendation Engine, Content-<br>
based Recommendation<br>
<br>
Context<br>
Recommendation engines are well-known systems used on many <br>
commercial websites (see annexe). They offer recommendations such <br>
as &quot;people who liked/bought this item also liked/bought these <br>
items&quot;, or &quot;we predict that you will rate this item 4 stars.&quot=
;<br>
<br>
Recommendation engines use 2 different techniques:<br>
<br>
o	Techniques called collaborative filtering: these techniques <br>
use the purchase histories or user browsing logs with algorithm such <br>
as &quot;basket analysis&quot; (association rules) or methods such as k-<br=
>
nearest neighbours.<br>
o	Techniques called content-based filtering: in such a case <br>
one seeks to build a thematic profile (for instance, using key-<br>
words) which enables the system to score each item. The list of the <br>
scored items is then presented to the user.<br>
<br>
The subject on the recommendation engines has two main axes:<br>
<br>
1.	Hybrid engine<br>
<br>
We are looking for the best of both worlds: an engine which can work <br>
with both collaborative filtering and content-based filtering whilst <br>
still providing good predictive and scalability performances.<br>
<br>
2.	&quot;AnyTime&quot; engine<br>
<br>
We are looking for an algorithm and a framework architecture which <br>
enables the recommendation system to work &quot;anytime&quot; (the model <b=
r>
updates are in real time or incremental)<wbr>.<br>
Research Work<br>
<br>
The research work will have 3 parts:<br>
<br>
1. State of the art<br>
First of all, you will have to make a state of the art of the <br>
different technologies of hybrid or incremental recommendation <br>
engines: types of learning method used, type of data used, bench and <br>
performances=85<wbr>.<br>
 <br>
2. Algorithms and evaluation<br>
Next you will have to define one or more innovative recommendation <br>
engine algorithms suited to the aforementioned constraints.<br>
<br>
3. Communication<br>
This work will consist in publishing international articles and/or <br>
develop patents.<br>
<br>
Type of position and location<br>
-	A one year temporary work contract for FT R&amp;D (Orange Labs)<br>
-	working in a research team of about 15 people<br>
-	department: France Telecom / R&amp;D / Tech / Easy / Tsi<br>
-	location : Lannion (Brittany, C=F4tes d'Armor, France)<br>
-	salary: approx 32 K Euros<br>
<br>
Education and skills required<br>
-	knowledge in machine learning techniques<br>
-	PhD in Computer Science, if possible in AI, Statistics, <br>
Machine Learning or Signal Processing.<br>
-	skill in programming: knowledge and experience in Java<br>
NB: CVs which not correspond to this profile will not be considered<br>
<br>
Useful supplementary skills<br>
-	experience in the field of recommendation systems<br>
-	experience in the field of Information Retrieval<br>
-	Knowledge in Databases (MySQL=85)<br>
-	experience of working in a team<br>
<br>
Application deadline<br>
<br>
Application for this position will be accepted until November 2008 <br>
at the latest.<br>
<br>
Contact<br>
<br>
Frank Meyer<br>
France Telecom R&amp;D/TECH/EASY (LD128)<br>
2 avenue Pierre Marzin 22307 Lannion Cedex <br>
E-mail : <a href=3D"mailto:franck.meyer%40orange-ftgroup.com">franck.meyer@=
<wbr>orange-ftgroup.<wbr>com</a><br>
Telephone : &#43;33 (0)2 96 05 28 89<br>
<a href=3D"http://www.francetelecom.com/rd">http://www.francete<wbr>lecom.c=
om/<wbr>rd</a><br>
<br>
R=E9f=E9rences<br>
-	<a href=3D"http://en.wikipedia.org/wiki/Collaborative_filtering">http://e=
n.wikipedia<wbr>.org/wiki/<wbr>Collaborative_<wbr>filtering</a><br>
-	<a href=3D"http://en.wikipedia.org/wiki/Recommendation_system">http://en.=
wikipedia<wbr>.org/wiki/<wbr>Recommendation_<wbr>system</a><br>
-	NetFlix Challenge :  <a href=3D"http://www.netflixprize.com/">http://www.=
netflixp<wbr>rize.com/</a><br>
<br>
(Adomavicius &amp; Tuzhilin, 2005) Adomavicius, G., &amp; Tuzhilin, A. <br>
(2005). Toward the next generation of recommender systems: A survey <br>
of the state-of-the-<wbr>art and possible extensions. IEEE Transactions <br=
>
on Knowledge and Data Engineering, 17(6), 734=96749.<br>
<br>
(Bell et al., 2007) Bell, R., Koren, Y., &amp; Volinsky, C. (2007). <br>
Modeling relationships at multiple scales to improve accuracy of <br>
large recommender systems. In 13th ACM SIGKDD International <br>
Conference on Knowledge Discovery and Data Mining (pp. 95=96104). New <br>
York, NY, USA. ACM.<br>
<br>
(Candillier et al., 2007) Candillier, L., Meyer, F., &amp; Boull=E9, M. <br=
>
(2007). Comparing state-of-the-<wbr>art collaborative filtering systems. <b=
r>
In Perner, P. (Ed.), 5th International Conference on Machine <br>
Learning and Data Mining in Pattern Recognition (pp. 548=96562), <br>
Leipzig, Germany. Springer Verlag.<br>
<br>
(Herlocker et al., 2004) Herlocker, J., Konstan, J., Terveen, L., &amp; <br=
>
Riedl, J. (2004). Evaluating collaborative filtering recommender <br>
systems. In ACM Transactions on Information Systems, 22(1), 5=9653.<br>
<br>
(Melville et al., 2002) Melville, P., Mooney, R., &amp; Nagarajan, R. <br>
(2002). Content-boosted collaborative filtering for improved <br>
recommendations. In 18th National Conference on Artificial <br>
Intelligence (pp. 187-192).<br>
<br>
(Polcicova et al., 2000) Polcicova, G., Slovak, R., &amp; Navrat, P. <br>
(2000). Combining content-based and collaborative filtering. In <br>
ADBIS-DASFAA Symposium (pp. 118-127).<br>
<br>
</p>
    </div>=20=20

    <!--~-|**|PrettyHtmlStart|**|-~-->
    <span width=3D"1" style=3D"color: white;">__._,_.___</span>
    <!-- Start the section with Message In topic -->
    <div id=3D"ygrp-actbar">
              <span class=3D"left">
          <a href=3D"http://groups.yahoo.com/group/webir/message/2549;_ylc=
=3DX3oDMTM0N3VicjJsBF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDgz=
MjA2BG1zZ0lkAzI1NDkEc2VjA2Z0cgRzbGsDdnRwYwRzdGltZQMxMjE2NTU0NjI1BHRwY0lkAzI=
1NDk-">
            Messages in this topic          </a> (<span class=3D"bld">1</sp=
an>)
        </span>
        <a href=3D"http://groups.yahoo.com/group/webir/post;_ylc=3DX3oDMTJw=
a2hwMmt2BF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDgzMjA2BG1zZ0l=
kAzI1NDkEc2VjA2Z0cgRzbGsDcnBseQRzdGltZQMxMjE2NTU0NjI1?act=3Dreply&messageNu=
m=3D2549">
          <span class=3D"bld">
            Reply          </span> (via web post)
        </a>  |=20
        <a href=3D"http://groups.yahoo.com/group/webir/post;_ylc=3DX3oDMTJl=
dmo1MjRwBF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDgzMjA2BHNlYwN=
mdHIEc2xrA250cGMEc3RpbWUDMTIxNjU1NDYyNQ--" class=3D"bld">
          Start a new topic        </a>
          </div>=20
    <!-------     Start Nav Bar  ------>
    <!-- |**|begin egp html banner|**| -->
    <div id=3D"ygrp-vitnav">
                <a href=3D"http://groups.yahoo.com/group/webir/messages;_yl=
c=3DX3oDMTJlaDdnZzAzBF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDg=
zMjA2BHNlYwNmdHIEc2xrA21zZ3MEc3RpbWUDMTIxNjU1NDYyNQ--">Messages</a>=20=20
=20=20=20=20=20=20=20=20
=20=20=20=20=20=20=20=20
=20=20=20=20=20=20=20=20
=20=20=20=20=20=20=20=20
=20=20=20=20=20=20=20=20
=20=20=20=20=20=20=20=20
=20=20=20=20=20=20=20=20
    </div>=20=20
    <!-- |**|end egp html banner|**| -->

                        <div id=3D"ygrp-grft">
=20=20=20=20=20=20=20=20=20=20

        </div>
=20=20=20=20=20=20
                  <div id=3D"ygrp-mkp">
      <div id=3D"hd">MARKETPLACE</div>
      <div id=3D"ads">
                  <div class=3D"ad">
            <hr size=3D1 noshade><a href=3D"http://us.ard.yahoo.com/SIG=3D1=
3rt99eo7/M=3D624381.12730922.13032918.10835568/D=3Dgroups/S=3D1705083206:MK=
P1/Y=3DYAHOO/EXP=3D1216561825/L=3D/B=3DYfQxAULaX9k-/J=3D1216554625955908/A=
=3D5379723/R=3D0/SIG=3D14evd63no/*http://media.adrevolver.com/adrevolver/hr=
ef?banner=3D189158&place=3D26143&url_=3Dhttp://tc.deals.yahoo.com/tc/blockb=
uster/display.com?cid=3Dbbi00027">Blockbuster</a> is giving away a FREE tri=
al of - <a href=3D"http://us.ard.yahoo.com/SIG=3D13rt99eo7/M=3D624381.12730=
922.13032918.10835568/D=3Dgroups/S=3D1705083206:MKP1/Y=3DYAHOO/EXP=3D121656=
1825/L=3D/B=3DYfQxAULaX9k-/J=3D1216554625955908/A=3D5379723/R=3D1/SIG=3D14e=
vd63no/*http://media.adrevolver.com/adrevolver/href?banner=3D189158&place=
=3D26143&url_=3Dhttp://tc.deals.yahoo.com/tc/blockbuster/display.com?cid=3D=
bbi00027">Blockbuster Total Access.</a>
<!--AdRevolver code begin-->
<script type=3D"text/javascript">
<!--
var title =3D 'TITLE';
try {
if (title =3D=3D unescape('%u0054%u0049%u0054%u004C%u0045') && document.tit=
le) title =3D document.title;
var https =3D false; try {https =3D document.location.href.indexOf('https')=
=3D=3D0}catch(e){}
document.write('<img width=3D"0" height=3D"0" border=3D"0" src=3D"http' + (=
https ? 's' : '') + '://media.adrevolver.com/adrevolver/trace?adpath=3D1195=
9&title=3D' + escape(title) + '&ref=3D' + escape(document.referrer) + '&rnd=
=3D' + Math.round(Math.random() * 10000000) + '">');
} catch(e){}
//-->
</script>
<noscript>
<img width=3D"0" height=3D"0" border=3D"0" src=3D"https://media.adrevolver.=
com/adrevolver/trace?adpath=3D11959">
</noscript>
<!--AdRevolver code end-->          </div>
                              </div>
    </div>
=20=20=20=20=20=20
    <!-- yahoo logo -->
    <!-- |**|begin egp html banner|**| -->
    <div id=3D"ygrp-ft">
      <a href=3D"http://groups.yahoo.com/;_ylc=3DX3oDMTJkbHNudTRnBF9TAzk3Mz=
U5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDgzMjA2BHNlYwNmdHIEc2xrA2dmcARzd=
GltZQMxMjE2NTU0NjI1">
      <img src=3D"http://us.i1.yimg.com/us.yimg.com/i/yg/img/logo/ma_grp_16=
0.gif" height=3D"15" width=3D"106" border=3D"0" alt=3D"Yahoo! Groups"></a> =
<br>
      <a href=3D"http://groups.yahoo.com/group/webir/join;_ylc=3DX3oDMTJmOG=
xnOGRxBF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDgzMjA2BHNlYwNmd=
HIEc2xrA3N0bmdzBHN0aW1lAzEyMTY1NTQ2MjU-">Change settings via the Web</a> (Y=
ahoo! ID required) <br>
      Change settings via email: <a href=3D"mailto:webir-digest@yahoogroups=
.com?subject=3DEmail Delivery: Digest">Switch delivery to Daily Digest</a> =
| <a href =3D "mailto:[email protected]?subject=3DChange De=
livery Format: Traditional">Switch format to Traditional</a> <br>

      <a href=3D"http://groups.yahoo.com/group/webir;_ylc=3DX3oDMTJkaGpyZDM=
yBF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDgzMjA2BHNlYwNmdHIEc2=
xrA2hwZgRzdGltZQMxMjE2NTU0NjI1">
        Visit Your Group=20
      </a> |
      <a href=3D"http://docs.yahoo.com/info/terms/">
        Yahoo! Groups Terms of Use      </a> |
      <a href=3D"mailto:[email protected]?subject=3D">
        Unsubscribe      </a>=20
    </div>     <!-- |**|end egp html banner|**| -->
  </div> <!-- ygrp-msg -->

=20=20
  <!-- Sponsor -->
  <!-- |**|begin egp html banner|**| -->
  <div id=3D"ygrp-sponsor" style=3D"width:160px; float:right; clear:none; m=
argin:0 0 25px 0; background:white;">
    <!-- Network content -->
=20=20=20=20
<!-- Start Recommendations -->
<div id=3D"ygrp-reco">
     </div>
<!-- End Recommendations -->

	    <!-- Start vitality -->
	    <div id=3D"ygrp-vital">
	      		<div id=3D"vithd">Recent Activity</div>
		<ul style=3D"list-style-type:none; padding: 0; margin: 2px 0;">
		        <li style=3D"clear: both;">
      <div class=3D"ct" style=3D"float: right;"><span style=3D"display:none=
">&nbsp;</span>2</div>
      <div class=3D"cat"><a href=3D"http://groups.yahoo.com/group/webir/mem=
bers;_ylc=3DX3oDMTJmNGE0MGFiBF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAM=
xNzA1MDgzMjA2BHNlYwN2dGwEc2xrA3ZtYnJzBHN0aW1lAzEyMTY1NTQ2MjU-">New Members<=
/a></div>
    </li>
=20=20
=09=09=20=20=20=20
=09=09=20=20=20=20
=09=09=20=20=20=20
=09=09=20=20=20=20
=09=09=20=20=20=20
		</ul>
	      	      <a href=3D"http://groups.yahoo.com/group/webir;_ylc=3DX3oDMTJ=
lNWs4a3V0BF9TAzk3MzU5NzE0BGdycElkAzEyMzQ1ODUEZ3Jwc3BJZAMxNzA1MDgzMjA2BHNlYw=
N2dGwEc2xrA3ZnaHAEc3RpbWUDMTIxNjU1NDYyNQ--">
		Visit Your Group	      </a>
	    </div>=20
=09=20=20=20=20=09=20=20=20=20=09=20=20=20=20=20=20
	    <!-- Network content -->
	    	          <div id=3D"nc">
              <div class=3D"ad">
                      <div id=3D"hd1">Yahoo! Groups</div>=20
<p><a href=3D"http://us.ard.yahoo.com/SIG=3D13oadkrsu/M=3D493064.12662709.1=
2980595.8674578/D=3Dgroups/S=3D1705083206:NC/Y=3DYAHOO/EXP=3D1216561825/L=
=3D/B=3DYvQxAULaX9k-/J=3D1216554625955908/A=3D5349286/R=3D0/SIG=3D11ks6tatn=
/*http://advision.webevents.yahoo.com/HealthyLiving/">Discover healthy</a><=
/p>=20
<p>living groups and</p>=20
<p>live a full life.</p>                  </div>
                    <div class=3D"ad">
                      <div id=3D"hd1">Find helpful tips</div>=20
<p><a href=3D"http://us.ard.yahoo.com/SIG=3D13o61v6jm/M=3D493064.12016258.1=
2582637.8674578/D=3Dgroups/S=3D1705083206:NC/Y=3DYAHOO/EXP=3D1216561825/L=
=3D/B=3DY_QxAULaX9k-/J=3D1216554625955908/A=3D5191954/R=3D0/SIG=3D112mhte3e=
/*http://www.ygroupsblog.com/blog/">for Moderators</a></p>=20
<p>on the Yahoo!</p>=20
<p>Groups team blog.</p>                  </div>
                    <div class=3D"ad">
                      <div id=3D"hd1">Yahoo! Groups</div>=20
<p><a href=3D"http://us.ard.yahoo.com/SIG=3D13o83us5a/M=3D493064.12717544.1=
3024012.8674578/D=3Dgroups/S=3D1705083206:NC/Y=3DYAHOO/EXP=3D1216561825/L=
=3D/B=3DZPQxAULaX9k-/J=3D1216554625955908/A=3D5370600/R=3D0/SIG=3D11lnr7bf1=
/*http://advision.webevents.yahoo.com/familyographer/">Familyographer Zone<=
/a></p>=20
<p>Learn how to take</p>=20
<p>great pictures.</p>                  </div>
          </div>
=09=20=20=20=20
	  </div> 	  <!-- |**|end egp html banner|**| -->
	  <div style=3D"clear:both; color: #FFF; font-size:1px;">.</div>
	</div> 		  <img src=3D"http://geo.yahoo.com/serv?s=3D97359714/grpId=3D1234=
585/grpspId=3D1705083206/msgId=3D2549/stime=3D1216554625/nc1=3D5349286/nc2=
=3D5191954/nc3=3D5370600" width=3D"1" height=3D"1"> <br>
=09
	<span  style=3D"color: white;">__,_._,___</span>
	<!--~-|**|PrettyHtmlEnd|**|-~-->
	</body>
	<!--~-|**|PrettyHtmlStart|**|-~-->
	<head>
<style type=3D"text/css">
<!--
#ygrp-mkp{
  border: 1px solid #d8d8d8;
  font-family: Arial;
  margin: 14px 0px;
  padding: 0px 14px;
}
#ygrp-mkp hr{
  border: 1px solid #d8d8d8;
}
#ygrp-mkp #hd{
  color: #628c2a;
  font-size: 85%;
  font-weight: bold;
  line-height: 122%;
  margin: 10px 0px;
}
#ygrp-mkp #ads{
  margin-bottom: 10px;
}
#ygrp-mkp .ad{
  padding: 0 0;
}
#ygrp-mkp .ad a{
  color: #0000ff;
  text-decoration: none;
}
-->
</style>
</head>
	<head>
<style type=3D"text/css">
<!--
#ygrp-sponsor #ygrp-lc{
  font-family: Arial;
}
#ygrp-sponsor #ygrp-lc #hd{
  margin: 10px 0px;
  font-weight: bold;
  font-size: 78%;
  line-height: 122%;
}
#ygrp-sponsor #ygrp-lc .ad{
  margin-bottom: 10px;
  padding: 0 0;
}
-->
</style>
</head>
	<head>
	<style type=3D"text/css">
	<!--
	#ygrp-mlmsg {font-size:13px; font-family: arial,helvetica,clean,sans-serif=
;*font-size:small;*font:x-small;}
	#ygrp-mlmsg table {font-size:inherit;font:100%;}
	#ygrp-mlmsg select, input, textarea {font:99% arial,helvetica,clean,sans-s=
erif;}
	#ygrp-mlmsg pre, code {font:115% monospace;*font-size:100%;}
	#ygrp-mlmsg * {line-height:1.22em;}
	#ygrp-text{
	    font-family: Georgia;=09
	}
	#ygrp-text p{
	    margin: 0 0 1em 0;
	}
	#ygrp-tpmsgs{
	    font-family: Arial;=09
	    clear: both;
	}
	#ygrp-vitnav{
		padding-top: 10px;
		font-family: Verdana;
		font-size: 77%;
		margin: 0;
	}
	#ygrp-vitnav a{
		padding: 0 1px;
	}
	#ygrp-actbar{
		clear: both;
		margin: 25px 0;
		white-space:nowrap;
		color: #666;
		text-align: right;
	}
	#ygrp-actbar .left{
		float: left;
		white-space:nowrap;
	}
	.bld{font-weight:bold;}
	#ygrp-grft{
		font-family: Verdana;
		font-size: 77%;
		padding: 15px 0;
	}
	#ygrp-ft{
	  font-family: verdana;
	  font-size: 77%;
	  border-top: 1px solid #666;=20
	  padding: 5px 0;=20
	}
	#ygrp-mlmsg #logo{
	  padding-bottom: 10px;
	}

	#ygrp-reco {
	margin-bottom: 20px;
	padding: 0px;
	}
	#ygrp-reco #reco-head {
		font-weight: bold;
		color: #ff7900;
	}

	#reco-grpname{
        font-weight: bold;
        margin-top: 10px;
  	}
	#reco-category{
        	font-size: 77%;
	}
	#reco-desc{
        	font-size: 77%;
	}

	#ygrp-vital{
		background-color: #e0ecee;
		margin-bottom: 20px;
		padding: 2px 0 8px 8px;
	}
	#ygrp-vital #vithd{
		font-size: 77%;
		font-family: Verdana;
		font-weight: bold;
		color: #333;
		text-transform: uppercase;
	}
	#ygrp-vital ul{
		padding: 0;
		margin: 2px 0;
	}
	#ygrp-vital ul li{
	  list-style-type: none;
	  clear: both;
	  border: 1px solid #e0ecee;=20=20
	}
	#ygrp-vital ul li .ct{
	  font-weight: bold;
	  color: #ff7900;
	  float: right;
	  width: 2em;
	  text-align:right;
	  padding-right: .5em;
	}
	#ygrp-vital ul li .cat{
	  font-weight: bold;
	}
	#ygrp-vital a{
		text-decoration: none;
	}

	#ygrp-vital a:hover{
	  text-decoration: underline;
	}

	#ygrp-sponsor #hd{
		color: #999;
		font-size: 77%;
	}
	#ygrp-sponsor #ov{
		padding: 6px 13px;
		background-color: #e0ecee;
		margin-bottom: 20px;
	}
	#ygrp-sponsor #ov ul{
		padding: 0 0 0 8px;
		margin: 0;
	}
	#ygrp-sponsor #ov li{
		list-style-type: square;
		padding: 6px 0;
		font-size: 77%;
	}
	#ygrp-sponsor #ov li a{
		text-decoration: none;
		font-size: 130%;
	}
	#ygrp-sponsor #nc{
	  background-color: #eee;
	  margin-bottom: 20px;
	  padding: 0 8px;
	}
	#ygrp-sponsor .ad{
		padding: 8px 0;
	}
	#ygrp-sponsor .ad #hd1{
		font-family: Arial;
		font-weight: bold;
		color: #628c2a;
		font-size: 100%;
		line-height: 122%;
	}
	#ygrp-sponsor .ad a{
		text-decoration: none;
	}
	#ygrp-sponsor .ad a:hover{
		text-decoration: underline;
	}
	#ygrp-sponsor .ad p{
		margin: 0;
	}
	o{font-size: 0; }
	.MsoNormal{
	   margin: 0 0 0 0;
	}
	#ygrp-text tt{
	  font-size: 120%;
	}
	blockquote{margin: 0 0 0 4px;}
	.replbq{margin:4}
	-->
	</style>
	</head>
	<!--~-|**|PrettyHtmlEnd|**|-~-->
	</html><!--End group email -->


--rkoepZNBw6egbz5bKsNGVo6M--tNAVtby7P3WOK--