<?xml version='1.0' encoding='utf-8'?>
<eprints xmlns='http://eprints.org/ep2/data/2.0'>
  <eprint id='http://edata.bham.ac.uk/id/eprint/644'>
    <eprintid>644</eprintid>
    <rev_number>35</rev_number>
    <documents>
      <document id='http://edata.bham.ac.uk/id/document/3120'>
        <docid>3120</docid>
        <rev_number>4</rev_number>
        <files>
          <file id='http://edata.bham.ac.uk/id/file/15434'>
            <fileid>15434</fileid>
            <datasetid>document</datasetid>
            <objectid>3120</objectid>
            <filename>NeuralNetwork_Experiment2.zip</filename>
            <mime_type>application/zip</mime_type>
            <hash>29ff4ea8818b24bcb15484af43f4f774</hash>
            <hash_type>MD5</hash_type>
            <filesize>8050815545</filesize>
            <mtime>2021-04-16 15:33:05</mtime>
            <url>http://edata.bham.ac.uk/644/1/NeuralNetwork_Experiment2.zip</url>
          </file>
        </files>
        <eprintid>644</eprintid>
        <pos>1</pos>
        <placement>3</placement>
        <mime_type>application/zip</mime_type>
        <format>archive</format>
        <security>public</security>
        <license>cc_by_4</license>
        <main>NeuralNetwork_Experiment2.zip</main>
        <content>data</content>
      </document>
      <document id='http://edata.bham.ac.uk/id/document/3122'>
        <docid>3122</docid>
        <rev_number>4</rev_number>
        <files>
          <file id='http://edata.bham.ac.uk/id/file/15439'>
            <fileid>15439</fileid>
            <datasetid>document</datasetid>
            <objectid>3122</objectid>
            <filename>NeuralNetwork_Experiment2.z01</filename>
            <mime_type>application/zip</mime_type>
            <hash>9d81ac00cc84fa5b0e05f670ef878b46</hash>
            <hash_type>MD5</hash_type>
            <filesize>8589934592</filesize>
            <mtime>2021-04-16 17:44:16</mtime>
            <url>http://edata.bham.ac.uk/644/2/NeuralNetwork_Experiment2.z01</url>
          </file>
        </files>
        <eprintid>644</eprintid>
        <pos>2</pos>
        <placement>4</placement>
        <mime_type>application/zip</mime_type>
        <format>other</format>
        <security>public</security>
        <license>cc_by_4</license>
        <main>NeuralNetwork_Experiment2.z01</main>
        <content>data</content>
      </document>
      <document id='http://edata.bham.ac.uk/id/document/3127'>
        <docid>3127</docid>
        <rev_number>2</rev_number>
        <files>
          <file id='http://edata.bham.ac.uk/id/file/15455'>
            <fileid>15455</fileid>
            <datasetid>document</datasetid>
            <objectid>3127</objectid>
            <filename>NeuralNetwork_Experiment2.z02</filename>
            <mime_type>application/octet-stream</mime_type>
            <hash>815cc4e8107eaeb87760096566934c48</hash>
            <hash_type>MD5</hash_type>
            <filesize>8589934592</filesize>
            <mtime>2021-04-17 15:22:28</mtime>
            <url>http://edata.bham.ac.uk/644/3/NeuralNetwork_Experiment2.z02</url>
          </file>
        </files>
        <eprintid>644</eprintid>
        <pos>3</pos>
        <placement>3</placement>
        <mime_type>application/octet-stream</mime_type>
        <format>other</format>
        <security>public</security>
        <license>cc_by_4</license>
        <main>NeuralNetwork_Experiment2.z02</main>
        <content>data</content>
      </document>
      <document id='http://edata.bham.ac.uk/id/document/3130'>
        <docid>3130</docid>
        <rev_number>2</rev_number>
        <files>
          <file id='http://edata.bham.ac.uk/id/file/15461'>
            <fileid>15461</fileid>
            <datasetid>document</datasetid>
            <objectid>3130</objectid>
            <filename>NeuralNetwork_Experiment2.z03</filename>
            <mime_type>application/octet-stream</mime_type>
            <hash>521c781f37988563511b7f075c63f076</hash>
            <hash_type>MD5</hash_type>
            <filesize>8589934592</filesize>
            <mtime>2021-04-17 17:08:32</mtime>
            <url>http://edata.bham.ac.uk/644/6/NeuralNetwork_Experiment2.z03</url>
          </file>
        </files>
        <eprintid>644</eprintid>
        <pos>6</pos>
        <placement>6</placement>
        <mime_type>application/octet-stream</mime_type>
        <format>other</format>
        <security>public</security>
        <license>cc_by_4</license>
        <main>NeuralNetwork_Experiment2.z03</main>
        <content>data</content>
      </document>
      <document id='http://edata.bham.ac.uk/id/document/3131'>
        <docid>3131</docid>
        <rev_number>2</rev_number>
        <files>
          <file id='http://edata.bham.ac.uk/id/file/15463'>
            <fileid>15463</fileid>
            <datasetid>document</datasetid>
            <objectid>3131</objectid>
            <filename>NeuralNetwork_Experiment2.z04</filename>
            <mime_type>application/octet-stream</mime_type>
            <hash>b8f4b25622a658d908cee235256f6c47</hash>
            <hash_type>MD5</hash_type>
            <filesize>8589934592</filesize>
            <mtime>2021-04-17 18:45:48</mtime>
            <url>http://edata.bham.ac.uk/644/7/NeuralNetwork_Experiment2.z04</url>
          </file>
        </files>
        <eprintid>644</eprintid>
        <pos>7</pos>
        <placement>7</placement>
        <mime_type>application/octet-stream</mime_type>
        <format>other</format>
        <security>public</security>
        <license>cc_by_4</license>
        <main>NeuralNetwork_Experiment2.z04</main>
        <content>data</content>
      </document>
      <document id='http://edata.bham.ac.uk/id/document/3132'>
        <docid>3132</docid>
        <rev_number>2</rev_number>
        <files>
          <file id='http://edata.bham.ac.uk/id/file/15467'>
            <fileid>15467</fileid>
            <datasetid>document</datasetid>
            <objectid>3132</objectid>
            <filename>NeuralNetwork_Experiment2.z05</filename>
            <mime_type>application/octet-stream</mime_type>
            <hash>1d3c2c0c55ad4116d91a8f092fe32d3d</hash>
            <hash_type>MD5</hash_type>
            <filesize>8589934592</filesize>
            <mtime>2021-04-18 04:48:37</mtime>
            <url>http://edata.bham.ac.uk/644/8/NeuralNetwork_Experiment2.z05</url>
          </file>
        </files>
        <eprintid>644</eprintid>
        <pos>8</pos>
        <placement>8</placement>
        <mime_type>application/octet-stream</mime_type>
        <format>other</format>
        <security>public</security>
        <license>cc_by_4</license>
        <main>NeuralNetwork_Experiment2.z05</main>
        <content>data</content>
      </document>
    </documents>
    <eprint_status>archive</eprint_status>
    <userid>267</userid>
    <dir>disk0/00/00/06/44</dir>
    <datestamp>2021-05-05 17:32:16</datestamp>
    <lastmod>2021-05-05 17:38:28</lastmod>
    <status_changed>2021-05-05 17:32:16</status_changed>
    <type>data_collection</type>
    <metadata_visibility>show</metadata_visibility>
    <creators>
      <item>
        <name>
          <family>Wan</family>
          <given>Jizheng</given>
        </name>
        <id>JXW178@bham.ac.uk</id>
        <orcid>0000-0002-1069-4582</orcid>
      </item>
    </creators>
    <title>Research data related to NeuralNetwork-Experiment2</title>
    <divisions>
      <item>10col_ephy</item>
    </divisions>
    <keywords>Neural Network (NN), Neural Complex, Machine Learning</keywords>
    <abstract>The dataset supports the thesis &quot;Semantic Impact - A novel approach for domain concept selection in ontology learning&quot;. The neural network set trained in the second experiment (by the best NN structure identified as part of the experiment, which contains 3 hidden layers and 1500 nodes on each layer).

The format is as follow:
TrainedNet_&lt;Concept_Name&gt;_&lt;Source/Target&gt;
Source is indicated as Y
Target is indicated as X

NN is trained by using DeepLearning4J 1.0.0-beta3 and can be loaded with the following code:

String strPath = &quot;path to the NN file&quot;;
MultiLayerNetwork myNN = ModelSerializer.restoreMultiLayerNetwork(strPath);
myNN.getLayerWiseConfigurations().setTrainingWorkspaceMode(WorkspaceMode.ENABLED);</abstract>
    <date>2021-05-05</date>
    <date_type>published</date_type>
    <publisher>University of Birmingham</publisher>
    <id_number>https://doi.org/10.25500/edata.bham.00000644</id_number>
    <data_type>Dataset</data_type>
    <contact_email>research-data@contacts.bham.ac.uk</contact_email>
    <department>
      <item>School of Computer Science</item>
    </department>
    <data_protection_legislation>no</data_protection_legislation>
    <funder_reqs>no</funder_reqs>
    <contractual_reqs>no</contractual_reqs>
    <university_reqs>TRUE</university_reqs>
  </eprint>
</eprints>
