Second Order Pseudolikelihood Learning in Relational Domain

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <title>Second Order Pseudolikelihood Learning in Relational Domain</title>  <meta charset="UTF-8"> <meta name="viewport" content="width=device-width, initial-scale=1.0"> <meta http-equiv="X-UA-Compatible" content="ie=edge"> <meta name="title" content="Second Order Pseudolikelihood Learning in Relational Domain"> <meta name="description" content="We use composite likelihood for structure learning and parameter estimation in relational dependency networks (RDNs). RDNs currently use pseudolikelihood, to learn parameters, which is a special case of composite likelihood function. Composite likelihood learning is used to give trade-off between computational complexity and performance of the model. Variance of the model is minimum in case of full likelihood and maximum in pseudolikelihood. In particular we focus on modified second order pseudolikelihood function and extend relational Bayesian classifier (RBC) to this setting. Second order RDNs explore pairwise attribute correlation. We evaluate second order learning on synthetic and real world data sets. We observe experimentally second order model has an edge over the pseudolikelihood based model particularly when correlation is high." /> <meta name="keywords" content=""/>   <meta name="dc.title" content="Second Order Pseudolikelihood Learning in Relational Domain"> <meta name="dc.creator" content="Krishna Kumar Tiwari"> <meta name="dc.creator" content="V. Vijaya Saradhi"> <meta name="dc.type" content="Article"> <meta name="dc.source" content="International Conference on Data Mining and Applications (DMAP 2019), Vol.9, No.16"> <meta name="dc.date" content="2019-12-14"> <meta name="dc.identifier" content="10.5121/csit.2019.91603"> <meta name="dc.publisher" content="AIRCC Publishing Corporation"> <meta name="dc.rights" content="http://creativecommons.org/licenses/by/3.0/"> <meta name="dc.format" content="application/pdf"> <meta name="dc.language" content="en"> <meta name="dc.description" content="We use composite likelihood for structure learning and parameter estimation in relational dependency networks (RDNs). RDNs currently use pseudolikelihood, to learn parameters, which is a special case of composite likelihood function. Composite likelihood learning is used to give trade-off between computational complexity and performance of the model. Variance of the model is minimum in case of full likelihood and maximum in pseudolikelihood. In particular we focus on modified second order pseudolikelihood function and extend relational Bayesian classifier (RBC) to this setting. Second order RDNs explore pairwise attribute correlation. We evaluate second order learning on synthetic and real world data sets. We observe experimentally second order model has an edge over the pseudolikelihood based model particularly when correlation is high."> <meta name="dc.subject" content=""> <meta name="dc.subject" content=""> <meta name="dc.subject" content=""> <meta name="dc.subject" content="">   <meta name="prism.publicationName" content="International Conference on Data Mining and Applications (DMAP 2019)"> <meta name="prism.publicationDate" content="2019-12-14"> <meta name="prism.volume" content="9"> <meta name="prism.number" content="16"> <meta name="prism.section" content="Article"> <meta name="prism.startingPage" content="21">   <meta name="citation_journal_title" content="International Conference on Data Mining and Applications (DMAP 2019)"> <meta name="citation_publisher" content="AIRCC Publishing Corporation"> <meta name="citation_authors" content="Krishna Kumar Tiwari, V. Vijaya Saradhi;"> <meta name="citation_title" content="Second Order Pseudolikelihood Learning in Relational Domain"> <meta name="citation_online_date" content="2019-12-14"> <meta name="citation_volume" content="9"> <meta name="citation_issue" content="16"> <meta name="citation_firstpage" content="21"> <meta name="citation_author" content="Krishna Kumar Tiwari"> <meta name="citation_author" content="V. Vijaya Saradhi"> <meta name="citation_doi" content="10.5121/csit.2019.91603"> <meta name="citation_abstract_html_url" content="http://aircconline.com/csit/abstract/v9n16/csit91603.html"> <meta name="citation_pdf_url" content="http://aircconline.com/csit/papers/vol9/csit91603.pdf">   <meta property="og:site_name" content="AIRCC" /> <meta property="og:type" content="article" /> <meta property="og:url" content="http://aircconline.com/csit/abstract/v9n16/csit91603.html"/> <meta property="og:title" content="Second Order Pseudolikelihood Learning in Relational Domain"> <meta property="og:description" content="We use composite likelihood for structure learning and parameter estimation in relational dependency networks (RDNs). RDNs currently use pseudolikelihood, to learn parameters, which is a special case of composite likelihood function. Composite likelihood learning is used to give trade-off between computational complexity and performance of the model. Variance of the model is minimum in case of full likelihood and maximum in pseudolikelihood. In particular we focus on modified second order pseudolikelihood function and extend relational Bayesian classifier (RBC) to this setting. Second order RDNs explore pairwise attribute correlation. We evaluate second order learning on synthetic and real world data sets. We observe experimentally second order model has an edge over the pseudolikelihood based model particularly when correlation is high." />   <link rel="icon" type="image/ico" href="../img/ico.ico"/> <link rel="stylesheet" type="text/css" href="../main1.css" media="screen" /> <style type="text/css"> a{ color:white; text-decoration:none; line-height:20px; } ul li a{ font-weight:bold; color:#000; list-style:none; text-decoration:none; size:10px;} .imagess { height:90px; text-align:left; margin:0px 5px 2px 8px; float:right; border:none; } #left p { font-family: CALIBRI; font-size: 16px; margin-left: 20px; font-weight: 500; } .right { margin-right: 20px; } #button{ float: left; font-size: 14px; margin-left: 10px; height: 28px; width: auto; background-color: #1e86c6; } </style> </head> <body> <div class="font"> <div id="wap"> <div id="page"> <div id="top"> <form action="https://airccj.org/csecfp/library/Search.php" method="get" target="_blank" > <table width="100%" cellspacing="0" cellpadding="0" > <tr class="search_input"> <td width="665" align="right"> </td> <td width="236" > <input name="title" type="text" value="Enter the paper title" class="search_textbox" onclick="if(this.value=='Enter the paper title'){this.value=''}" onblur="if(this.value==''){this.value='Enter the paper title'}" /> </td> <td width="59"> <input type="image" src="../img/go.gif" /> </td> </tr> <tr> <td colspan="3" valign="top"><img src="../img/top1.gif" alt="Academy & Industry Research Collaboration Center (AIRCC)" /></td> </tr> </table> </form> </div> <div id="font-face"> <div id="menu"> <a href="http://airccse.org">Home</a> <a href="http://airccse.org/journal.html">Journals</a> <a href="http://airccse.org/ethics.html">Ethics</a> <a href="http://airccse.org/conference.html">Conferences</a> <a href="http://airccse.org/past.html">Past Events</a> <a href="http://airccse.org/b.html">Submission</a> </div> <div id="content"> <div id="left"> <h2 class="lighter"><font size="2">Volume 9, Number 16, December 2019</font></h2> <h4 style="text-align:center;height:auto;"><a>Second Order Pseudolikelihood Learning in Relational Domain</a></h4> <h3>  Authors</h3> <p class="#left right" style="text-align:">Krishna Kumar Tiwari, V. Vijaya Saradhi, Indian Institute of Technology Guwahati, India </p> <h3>  Abstract</h3> <p class="#left right" style="text-align:justify">We use composite likelihood for structure learning and parameter estimation in relational dependency networks (RDNs). RDNs currently use pseudolikelihood, to learn parameters, which is a special case of composite likelihood function. Composite likelihood learning is used to give trade-off between computational complexity and performance of the model. Variance of the model is minimum in case of full likelihood and maximum in pseudolikelihood. In particular we focus on modified second order pseudolikelihood function and extend relational Bayesian classifier (RBC) to this setting. Second order RDNs explore pairwise attribute correlation. We evaluate second order learning on synthetic and real world data sets. We observe experimentally second order model has an edge over the pseudolikelihood based model particularly when correlation is high. </p> <br> <button type="button" id="button"><a target="blank" href="/csit/papers/vol9/csit91603.pdf">Full Text</a></button>   <button type="button" id="button"><a href="http://airccse.org/csit/V9N16.html">Volume 9, Number 16</a></button> <br><br><br><br><br> </div> <div id="right"> <div class="menu_right"> <ul> <li id="id"><a href="http://airccse.org/editorial.html">Editorial Board</a></li> <li><a href="http://airccse.org/arch.html">Archives</a></li> <li><a href="http://airccse.org/indexing.html">Indexing</a></li> <li><a href="http://airccse.org/faq.html" target="_blank">FAQ</a></li> </ul> </div> <h2 class="h2" align="center">Conference Proceedings</h2> <a href="http://airccse.org/cscp.html"><img src="cscf.jpg" class="img" /></a> <div class="clear_left"></div> <br> </div> <div class="clear"></div> <div id="footer"> <table width="100%" > <tr> <td width="46%" class="F_menu"><a href="http://airccse.org/subscription.html">Subscription</a> <a href="http://airccse.org/membership.html">Membership</a> <a href="http://airccse.org/cscp.html">AIRCC CSCP</a> <a href="http://airccse.org/acontact.html">Contact Us</a> </td> <td width="54%" align="right"><a href="http://airccse.org/index.php"><img src="/csit/abstract/img/logo.gif" alt="" width="21" height="24" /></a><a href="http://www.facebook.com/AIRCCSE"><img src="/csit/abstract/img/facebook.jpeg" alt="" width="21" height="24" /></a><a href="https://twitter.com/AIRCCFP"><img src="/csit/abstract/img/twitter.jpeg" alt="" width="21" height="24" /></a><a href="http://cfptech.wordpress.com/"><img src="/csit/abstract/img/index1.jpeg" alt="" width="21" height="24" /></a></td> </tr> <tr><td height="25" colspan="2"> <p align="center">All Rights Reserved ® AIRCC</p> </td></tr> </table> </div> </div> </div> </div> </div> </div> </body> </html>

CINXE.COM

Second Order Pseudolikelihood Learning in Relational Domain