User Tools

Site Tools


experiment_corpus
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<?xml-stylesheet type="text/xsl" href="lexUnit.xsl"?>
<lexUnit xmlns="http://framenet.icsi.berkeley.edu" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" frame="Experiment" frameID="652" status="Finished_Initial" POS="N" name="experiment.n" ID="8538" totalAnnotated="9" xsi:schemaLocation="../schema/lexUnit.xsd">
    <header>
        <corpus description="BNC2" name="BNC2" ID="111">
            <document description="bncp" name="bncp" ID="421"/>
        </corpus>
        <frame>
            <FE fgColor="FFFFFF" bgColor="FF69B4" type="Core" abbrev="Experiment" name="ML_Experiment"/>
            <FE fgColor="FFFFFF" bgColor="004C99" type="Peripheral" abbrev="Data" name="Data"/>
            <FE fgColor="FFFFFF" bgColor="0000FF" type="Peripheral" abbrev="Algorithm" name="ML_Algorithm"/>
            <FE fgColor="FFFFFF" bgColor="99004C" type="Peripheral" abbrev="Measure" name="Measure"/>
            <FE fgColor="FFFFFF" bgColor="009900" type="Peripheral" abbrev="Error" name="Error"/>
            <FE fgColor="FFFFFF" bgColor="1088BB" type="Peripheral" abbrev="Value" name="Measure_Error_Value"/>
            <FE fgColor="FFFFFF" bgColor="BB9900" type="Peripheral" abbrev="Indication" name="Loss_Gain_Indication"/>
        </frame>
    </header>
    <definition>This is a frame for representing relations between ML Experiment and Data used in the experiment, an ML Algorithm / Models applied, measure used to assess the results of an experiment or possibly an Error calculated based on the experiment results, measure or error value  and indication of possible loss or gain in a comparison.</definition>
    <lexeme POS="N" name="error"/>
 
 
    <subCorpus name="01-unannotated-1">
        <sentence corpID="111" docID="421" sentNo="1" paragNo="181" aPos="0" ID="1215438">
            <text>An extensive empirical investigation, using 33 publicly available data sets, was undertaken to compare the use of random forests to existing state-of-the-art conformal predictors.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-014-5453-0"/>
                    <label cBy="351" end="35" start="13" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="35" start="3" name="ML_Experiment"/>
                <label end="74" start="44" name="Data"/>
                <label end="101" start="92" name="Loss_Gain_Indication"/>
                <label end="127" start="114" name="ML_Algorithm"/>
                <label end="177" start="141" name="ML_Algorithm"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="2" paragNo="181" aPos="0" ID="1215438">
            <text>Experiments on a large OCR data set have shown CB1 to significantly increase generalization accuracy over SSE or CE optimization, from 97.86% and 98.10%, respectively, to 99.11%.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-006-6266-6"/>
                    <label cBy="351" end="10" start="0" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="34" start="15" name="Data"/>
                <label end="75" start="51" name="Loss_Gain_Indication"/>
                <label end="10" start="0" name="ML_Experiment"/>
                <label end="151" start="135" name="Measure_Error_Value"/>
                <label end="176" start="171" name="Measure_Error_Value"/>
                <label end="99" start="77" name="Measure"/>
                <label end="127" start="106" name="ML_Algorithm"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="3" paragNo="181" aPos="0" ID="1215438">
            <text>Preliminary experiments in which more than 10,000 samples are generated produced no significant improvements in predictive accuracy, which indicates that the runs are sufficiently long to produce accurate approximations to the posterior distribution in these models.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-014-5475-7"/>
                    <label cBy="351" end="22" start="12" name="Target"/>
                    <label cBy="351" end="161" start="158" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="56" start="33" name="Data"/>
                <label end="107" start="72" name="Loss_Gain_Indication"/>
                <label end="22" start="0" name="ML_Experiment"/>
                <label end="130" start="112" name="Measure"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="4" paragNo="181" aPos="0" ID="1215438">
            <text>Empirically, the classifier is evaluated using a benchmark synthetic data from random sampling runs for initial statistical evidence regarding its classification accuracy and computational efficiency.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-006-9455-4"/>
                    <label cBy="351" end="98" start="94" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="98" start="49" name="Data"/>
                <label end="26" start="17" name="ML_Algorithm"/>
                <label end="169" start="147" name="Measure"/>
                <label end="198" start="175" name="Measure"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="5" paragNo="181" aPos="0" ID="1215438">
            <text>The algorithms were then run on the resulting unlabeled data.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-013-5393-0"/>
                    <label cBy="351" end="27" start="25" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="59" start="36" name="Data"/>
                <label end="13" start="4" name="ML_Algorithm"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="6" paragNo="181" aPos="0" ID="1215438">
            <text>Best results were obtained in this study when the weights were computed using mutual information between the features and the output class.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1023/A:1022603022740"/>
                    <label cBy="351" end="39" start="35" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="11" start="0" name="Loss_Gain_Indication"/>
                <label end="137" start="45" name="ML_Algorithm"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="7" paragNo="181" aPos="0" ID="1215438">
            <text>Our empirical evaluation shows that Feating performs significantly better than Boosting, Random Subspace and Bagging in terms of predictive accuracy, when a stable learner SVM is used as the base learner.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-010-5224-5"/>
                    <label cBy="351" end="23" start="14" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="72" start="44" name="Loss_Gain_Indication"/>
                <label end="42" start="36" name="ML_Algorithm"/>
                <label end="86" start="78" name="ML_Algorithm"/>
                <label end="103" start="89" name="ML_Algorithm"/>
                <label end="115" start="109" name="ML_Algorithm"/>
                <label end="174" start="157" name="ML_Algorithm"/>
                <label end="148" start="128" name="Measure"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="8" paragNo="181" aPos="0" ID="1215438">
            <text>In evaluation on four benchmark databases, the maximum pseudo-likelihood estimates approach the true conditional probabilities as observations increase.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-013-5362-7"/>
                    <label cBy="351" end="12" start="3" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="40" start="17" name="Data"/>
                <label end="150" start="43" name="Loss_Gain_Indication"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
        <sentence corpID="111" docID="421" sentNo="9" paragNo="181" aPos="0" ID="1215438">
            <text>We empirically show that the advanced MCMC algorithm performs significantly better than the original MCMC sampling scheme from Grzegorczyk and Husmeier (2012b) in terms of convergence and mixing.</text>
            <annotationSet cDate="05/02/2003 04:00:38 PDT Fri" status="MANUAL" ID="1877967">
                <layer rank="1" name="Target">
                    <label end="0" start="0" name="10.1007/s10994-012-5326-3"/>
                    <label cBy="351" end="12" start="3" name="Target"/>
                </layer>
                <layer rank="1" name="FE"><label end="51" start="29" name="ML_Algorithm"/>
                <label end="150" start="92" name="ML_Algorithm"/>
                <label end="81" start="53" name="Loss_Gain_Indication"/>
                <label end="182" start="172" name="Measure"/>
                <label end="193" start="188" name="Measure"/>
            </layer>
                <layer rank="1" name="GF"/>
                <layer rank="1" name="PT"/>
                <layer rank="1" name="Other"/>
                <layer rank="1" name="Sent"/>
                <layer rank="1" name="Verb"/>
            </annotationSet>
        </sentence>
 
    </subCorpus>
</lexUnit>
experiment_corpus.txt · Last modified: 2016/03/17 13:00 by pj