D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 12 Projects 2 Releases 2 Wiki Activity

Implemented a first prototype of incremental harvesting and trasformation using readlock

This commit is contained in:
Sandro La Bruzzo 2021-02-01 13:56:05 +01:00
parent b6b835ef49
commit 6ff234d81b
13 changed files with 297 additions and 132 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
dhp-common/pom.xml
View File

@ -98,6 +98,12 @@
<artifactId>dnet-pace-core</artifactId>
</dependency>
<dependency>
<groupId>org.apache.httpcomponents</groupId>
<artifactId>httpclient</artifactId>
</dependency>
<dependency>
<groupId>eu.dnetlib.dhp</groupId>
<artifactId>dhp-schemas</artifactId>

28
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/aggregation/common/AggregationUtility.java Normal file
View File

@ -0,0 +1,28 @@
package eu.dnetlib.dhp.aggregation.common;
import java.io.BufferedOutputStream;
import java.io.IOException;
import java.nio.charset.StandardCharsets;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.spark.sql.SparkSession;
public class AggregationUtility {
public static void writeTotalSizeOnHDFS(final SparkSession spark, final Long total, final String path)
throws IOException {
FileSystem fs = FileSystem.get(spark.sparkContext().hadoopConfiguration());
FSDataOutputStream output = fs.create(new Path(path));
final BufferedOutputStream os = new BufferedOutputStream(output);
os.write(total.toString().getBytes(StandardCharsets.UTF_8));
os.close();
}
}

97
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/collection/GenerateNativeStoreSparkJob.java
View File

@ -5,9 +5,9 @@ import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.io.*;
import java.nio.charset.StandardCharsets;
import java.util.Collections;
import java.util.Objects;
import java.util.Optional;
import java.util.Properties;
import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils;
@ -20,10 +20,9 @@ import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoder;
import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.SaveMode;
import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.sql.*;
import org.apache.spark.sql.expressions.Aggregator;
import org.apache.spark.util.LongAccumulator;
import org.dom4j.Document;
import org.dom4j.Node;
@ -34,19 +33,62 @@ import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.data.mdstore.manager.common.model.MDStoreVersion;
import eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode;
import eu.dnetlib.dhp.aggregation.common.AggregationUtility;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.collection.worker.CollectorWorkerApplication;
import eu.dnetlib.dhp.common.rest.DNetRestClient;
import eu.dnetlib.dhp.model.mdstore.MetadataRecord;
import eu.dnetlib.dhp.model.mdstore.Provenance;
import eu.dnetlib.message.MessageManager;
import scala.Tuple2;
public class GenerateNativeStoreSparkJob {
private static final Logger log = LoggerFactory.getLogger(GenerateNativeStoreSparkJob.class);
private static final String DATASET_NAME = "/store";
public static class MDStoreAggregator extends Aggregator<MetadataRecord, MetadataRecord, MetadataRecord> {
@Override
public MetadataRecord zero() {
return new MetadataRecord();
}
@Override
public MetadataRecord reduce(MetadataRecord b, MetadataRecord a) {
return getLatestRecord(b, a);
}
private MetadataRecord getLatestRecord(MetadataRecord b, MetadataRecord a) {
if (b == null)
return a;
if (a == null)
return b;
return (a.getDateOfCollection() > b.getDateOfCollection()) ? a : b;
}
@Override
public MetadataRecord merge(MetadataRecord b, MetadataRecord a) {
return getLatestRecord(b, a);
}
@Override
public MetadataRecord finish(MetadataRecord j) {
return j;
}
@Override
public Encoder<MetadataRecord> bufferEncoder() {
return Encoders.kryo(MetadataRecord.class);
}
@Override
public Encoder<MetadataRecord> outputEncoder() {
return Encoders.kryo(MetadataRecord.class);
}
}
public static void main(String[] args) throws Exception {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
@ -70,6 +112,12 @@ public class GenerateNativeStoreSparkJob {
final MDStoreVersion currentVersion = jsonMapper.readValue(mdStoreVersion, MDStoreVersion.class);
String readMdStoreVersionParam = parser.get("readMdStoreVersion");
log.info("readMdStoreVersion is {}", readMdStoreVersionParam);
final MDStoreVersion readMdStoreVersion = StringUtils.isBlank(readMdStoreVersionParam) ? null
: jsonMapper.readValue(readMdStoreVersionParam, MDStoreVersion.class);
Boolean isSparkSessionManaged = Optional
.ofNullable(parser.get("isSparkSessionManaged"))
.map(Boolean::valueOf)
@ -77,6 +125,9 @@ public class GenerateNativeStoreSparkJob {
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
SparkConf conf = new SparkConf();
conf.set("spark.serializer", "org.apache.spark.serializer.KryoSerializer");
conf.registerKryoClasses(Collections.singleton(MetadataRecord.class).toArray(new Class[] {}));
runWithSparkSession(
conf,
isSparkSessionManaged,
@ -105,8 +156,27 @@ public class GenerateNativeStoreSparkJob {
.distinct();
final Encoder<MetadataRecord> encoder = Encoders.bean(MetadataRecord.class);
Dataset<MetadataRecord> mdstore = spark.createDataset(nativeStore.rdd(), encoder);
if (readMdStoreVersion != null) {
// INCREMENTAL MODE
Dataset<MetadataRecord> currentMdStoreVersion = spark
.read()
.load(readMdStoreVersion.getHdfsPath() + DATASET_NAME)
.as(encoder);
TypedColumn<MetadataRecord, MetadataRecord> aggregator = new MDStoreAggregator().toColumn();
mdstore = currentMdStoreVersion
.union(mdstore)
.groupByKey(
(MapFunction<MetadataRecord, String>) MetadataRecord::getId,
Encoders.STRING())
.agg(aggregator)
.map((MapFunction<Tuple2<String, MetadataRecord>, MetadataRecord>) Tuple2::_2, encoder);
}
mdstore
.write()
.mode(SaveMode.Overwrite)
@ -116,17 +186,8 @@ public class GenerateNativeStoreSparkJob {
final Long total = mdstore.count();
FileSystem fs = FileSystem.get(spark.sparkContext().hadoopConfiguration());
FSDataOutputStream output = fs.create(new Path(currentVersion.getHdfsPath() + "/size"));
final BufferedOutputStream os = new BufferedOutputStream(output);
os.write(total.toString().getBytes(StandardCharsets.UTF_8));
os.close();
AggregationUtility.writeTotalSizeOnHDFS(spark, total, currentVersion.getHdfsPath() + "/size");
});
}
public static MetadataRecord parseRecord(

39
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/transformation/TransformSparkJobNode.java
View File

@ -3,14 +3,11 @@ package eu.dnetlib.dhp.transformation;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.io.ByteArrayInputStream;
import java.util.HashMap;
import java.io.IOException;
import java.util.Map;
import java.util.Objects;
import java.util.Optional;
import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.sql.Dataset;
@ -18,25 +15,18 @@ import org.apache.spark.sql.Encoder;
import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.SparkSession;
import org.apache.spark.util.LongAccumulator;
import org.dom4j.Document;
import org.dom4j.DocumentException;
import org.dom4j.Node;
import org.dom4j.io.SAXReader;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.data.mdstore.manager.common.model.MDStoreVersion;
import eu.dnetlib.dhp.aggregation.common.AggregationCounter;
import eu.dnetlib.dhp.aggregation.common.AggregationUtility;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
import eu.dnetlib.dhp.model.mdstore.MetadataRecord;
import eu.dnetlib.dhp.transformation.vocabulary.VocabularyHelper;
import eu.dnetlib.dhp.transformation.xslt.XSLTTransformationFunction;
import eu.dnetlib.dhp.utils.DHPUtils;
import eu.dnetlib.dhp.utils.ISLookupClientFactory;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
import eu.dnetlib.message.Message;
import eu.dnetlib.message.MessageManager;
import eu.dnetlib.message.MessageType;
public class TransformSparkJobNode {
@ -59,10 +49,14 @@ public class TransformSparkJobNode {
.orElse(Boolean.TRUE);
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
final String inputPath = parser.get("mdstoreInputPath");
final String outputPath = parser.get("mdstoreOutputPath");
final String mdstoreInputVersion = parser.get("mdstoreInputVersion");
final String mdstoreOutputVersion = parser.get("mdstoreOutputVersion");
// TODO this variable will be used after implementing Messaging with DNet Aggregator
final ObjectMapper jsonMapper = new ObjectMapper();
final MDStoreVersion nativeMdStoreVersion = jsonMapper.readValue(mdstoreInputVersion, MDStoreVersion.class);
final MDStoreVersion cleanedMdStoreVersion = jsonMapper.readValue(mdstoreOutputVersion, MDStoreVersion.class);
final String isLookupUrl = parser.get("isLookupUrl");
log.info(String.format("isLookupUrl: %s", isLookupUrl));
@ -72,11 +66,14 @@ public class TransformSparkJobNode {
runWithSparkSession(
conf,
isSparkSessionManaged,
spark -> transformRecords(parser.getObjectMap(), isLookupService, spark, inputPath, outputPath));
spark -> transformRecords(
parser.getObjectMap(), isLookupService, spark, nativeMdStoreVersion.getHdfsPath(),
cleanedMdStoreVersion.getHdfsPath()));
}
public static void transformRecords(final Map<String, String> args, final ISLookUpService isLookUpService,
final SparkSession spark, final String inputPath, final String outputPath) throws DnetTransformationException {
final SparkSession spark, final String inputPath, final String outputPath)
throws DnetTransformationException, IOException {
final LongAccumulator totalItems = spark.sparkContext().longAccumulator("TotalItems");
final LongAccumulator errorItems = spark.sparkContext().longAccumulator("errorItems");
@ -86,11 +83,13 @@ public class TransformSparkJobNode {
final Dataset<MetadataRecord> mdstoreInput = spark.read().format("parquet").load(inputPath).as(encoder);
final MapFunction<MetadataRecord, MetadataRecord> XSLTTransformationFunction = TransformationFactory
.getTransformationPlugin(args, ct, isLookUpService);
mdstoreInput.map(XSLTTransformationFunction, encoder).write().save(outputPath);
mdstoreInput.map(XSLTTransformationFunction, encoder).write().save(outputPath + "/store");
log.info("Transformed item " + ct.getProcessedItems().count());
log.info("Total item " + ct.getTotalItems().count());
log.info("Transformation Error item " + ct.getErrorItems().count());
AggregationUtility.writeTotalSizeOnHDFS(spark, ct.getProcessedItems().count(), outputPath + "/size");
}
}

6
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/transformation/TransformationFactory.java
View File

@ -30,13 +30,13 @@ public class TransformationFactory {
log.info("Transformation plugin required " + transformationPlugin);
switch (transformationPlugin) {
case "XSLT_TRANSFORM": {
final String transformationRuleName = jobArgument.get("transformationRuleTitle");
if (StringUtils.isBlank(transformationRuleName))
final String transformationRuleId = jobArgument.get("transformationRuleId");
if (StringUtils.isBlank(transformationRuleId))
throw new DnetTransformationException("Missing Parameter transformationRule");
final VocabularyGroup vocabularies = VocabularyGroup.loadVocsFromIS(isLookupService);
final String transformationRule = queryTransformationRuleFromIS(
transformationRuleName, isLookupService);
transformationRuleId, isLookupService);
final long dateOfTransformation = new Long(jobArgument.get("dateOfTransformation"));
return new XSLTTransformationFunction(counters, transformationRule, dateOfTransformation,

6
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/collection_input_parameters.json
View File

@ -35,6 +35,12 @@
"paramDescription": "the Metadata Store Version Info",
"paramRequired": true
},
{
"paramName": "rmv",
"paramLongName": "readMdStoreVersion",
"paramDescription": "the Read Lock Metadata Store Version bean",
"paramRequired": false
},
{
"paramName": "w",
"paramLongName": "workflowId",

4
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/oozie_app/config-default.xml
View File

@ -15,4 +15,8 @@
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
</configuration>

33
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/oozie_app/workflow.xml
View File

@ -51,7 +51,7 @@
</global>
<start to="StartTransaction"/>
<start to="collection_mode"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
@ -61,7 +61,7 @@
<switch>
<case to="StartTransaction">${wf:conf('collectionMode') eq 'REFRESH'}</case>
<case to="BeginRead">${wf:conf('collectionMode') eq 'INCREMENTAL'}</case>
<default to="ImportDatacite"/>
<default to="StartTransaction"/>
</switch>
</decision>
@ -99,7 +99,7 @@
<arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
</java>
<ok to="GenerateNativeStoreSparkJob"/>
<error to="RollBack"/>
<error to="FailCollection"/>
</action>
<action name="GenerateNativeStoreSparkJob">
@ -123,9 +123,10 @@
<arg>--provenance</arg><arg>${dataSourceInfo}</arg>
<arg>--xpath</arg><arg>${identifierPath}</arg>
<arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
<arg>--readMdStoreVersion</arg><arg>${wf:actionData('BeginRead')['mdStoreReadLockVersion']}</arg>
</spark>
<ok to="collection_mode_end"/>
<error to="RollBack"/>
<error to="FailCollection"/>
</action>
@ -133,7 +134,7 @@
<switch>
<case to="CommitVersion">${wf:conf('collectionMode') eq 'REFRESH'}</case>
<case to="EndRead">${wf:conf('collectionMode') eq 'INCREMENTAL'}</case>
<default to="ImportDatacite"/>
<default to="CommitVersion"/>
</switch>
</decision>
@ -161,6 +162,28 @@
<error to="Kill"/>
</action>
<decision name="FailCollection">
<switch>
<case to="RollBack">${wf:conf('collectionMode') eq 'REFRESH'}</case>
<case to="EndReadRollBack">${wf:conf('collectionMode') eq 'INCREMENTAL'}</case>
<default to="RollBack"/>
</switch>
</decision>
<action name="EndReadRollBack">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>READ_UNLOCK</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
<arg>--readMDStoreId</arg><arg>${wf:actionData('BeginRead')['mdStoreReadLockVersion']}</arg>
<capture-output/>
</java>
<ok to="RollBack"/>
<error to="Kill"/>
</action>
<action name="RollBack">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>

4
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/transformation/oozie_app/config-default.xml
View File

@ -15,4 +15,8 @@
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
</configuration>

96
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/transformation/oozie_app/workflow.xml
View File

@ -1,25 +1,25 @@
<workflow-app name="Transformation_Workflow" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>mdstoreInputPath</name>
<description>the path of the native MDStore</description>
<name>mdStoreInputId</name>
<description>the identifier of the native MDStore</description>
</property>
<property>
<name>mdstoreOutputPath</name>
<name>mdStoreOutputId</name>
<description>the identifier of the cleaned MDStore</description>
</property>
<property>
<name>mdStoreManagerURI</name>
<description>the path of the cleaned mdstore</description>
</property>
<property>
<name>transformationRuleTitle</name>
<name>transformationRuleId</name>
<description>The transformation Rule to apply</description>
</property>
<property>
<name>transformationPlugin</name>
<description>The transformation Plugin</description>
</property>
<property>
<name>dateOfTransformation</name>
<description>The timestamp of the transformation date</description>
@ -28,11 +28,34 @@
</parameters>
<start to="TransformJob"/>
<start to="BeginRead"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="BeginRead">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>READ_LOCK</arg>
<arg>--mdStoreID</arg><arg>${mdStoreInputId}</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
<capture-output/>
</java>
<ok to="StartTransaction"/>
<error to="Kill"/>
</action>
<action name="StartTransaction">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>NEW_VERSION</arg>
<arg>--mdStoreID</arg><arg>${mdStoreOutputId}</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
<capture-output/>
</java>
<ok to="TransformJob"/>
<error to="EndReadRollBack"/>
</action>
<action name="TransformJob">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
@ -49,18 +72,63 @@
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--mdstoreInputPath</arg><arg>${mdstoreInputPath}</arg>
<arg>--mdstoreOutputPath</arg><arg>${mdstoreOutputPath}</arg>
<arg>--mdstoreInputVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
<arg>--mdstoreOutputVersion</arg><arg>${wf:actionData('BeginRead')['mdStoreReadLockVersion']}</arg>
<arg>--dateOfTransformation</arg><arg>${dateOfTransformation}</arg>
<arg>--transformationPlugin</arg><arg>${transformationPlugin}</arg>
<arg>--transformationRuleTitle</arg><arg>${transformationRuleTitle}</arg>
<arg>--transformationRuleId</arg><arg>${transformationRuleId}</arg>
</spark>
<ok to="EndRead"/>
<error to="EndReadRollBack"/>
</action>
<action name="EndRead">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>READ_UNLOCK</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
<arg>--readMDStoreId</arg><arg>${wf:actionData('BeginRead')['mdStoreReadLockVersion']}</arg>
<capture-output/>
</java>
<ok to="CommitVersion"/>
<error to="Kill"/>
</action>
<action name="CommitVersion">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>COMMIT</arg>
<arg>--namenode</arg><arg>${nameNode}</arg>
<arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<action name="EndReadRollBack">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>READ_UNLOCK</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
<arg>--readMDStoreId</arg><arg>${wf:actionData('BeginRead')['mdStoreReadLockVersion']}</arg>
<capture-output/>
</java>
<ok to="RollBack"/>
<error to="Kill"/>
</action>
<action name="RollBack">
<java>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>ROLLBACK</arg>
<arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
</java>
<ok to="Kill"/>
<error to="Kill"/>
</action>
<end name="End"/>

10
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/transformation/transformation_input_parameters.json
View File

@ -13,19 +13,19 @@
},
{
"paramName": "i",
"paramLongName": "mdstoreInputPath",
"paramDescription": "the path of the sequencial file to read",
"paramLongName": "mdstoreInputVersion",
"paramDescription": "the mdStore Version bean of the Input",
"paramRequired": true
},
{
"paramName": "o",
"paramLongName": "mdstoreOutputPath",
"paramDescription": "the path of the result DataFrame on HDFS",
"paramLongName": "mdstoreOutputVersion",
"paramDescription": "the mdStore Version bean of the Output",
"paramRequired": true
},
{
"paramName": "tr",
"paramLongName": "transformationRuleTitle",
"paramLongName": "transformationRuleId",
"paramDescription": "the transformation Rule to apply to the input MDStore",
"paramRequired": true
},

4
dhp-workflows/dhp-aggregation/src/test/resources/eu/dnetlib/dhp/transform/ext_simple.xsl
View File

@ -9,7 +9,9 @@
<oai:record>
<xsl:copy-of select="//oai:header"/>
<metadata>
<xsl:for-each select="//oai:set">
<xsl:copy-of select="//oai:metadata/*"/>
<xsl:for-each select="//oai:setSpec">
<dr:CobjCategory><xsl:value-of select="vocabulary:clean(.,'dnet:publication_resource')"/></dr:CobjCategory>
</xsl:for-each>
</metadata>

96
dhp-workflows/dhp-aggregation/src/test/resources/eu/dnetlib/dhp/transform/input.xml
View File

@ -1,68 +1,32 @@
<?xml version="1.0" encoding="UTF-8"?>
<oai:record xmlns="http://namespace.openaire.eu/"
xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:dr="http://www.driver-repository.eu/namespace/dr"
xmlns:dri="http://www.driver-repository.eu/namespace/dri"
xmlns:oaf="http://namespace.openaire.eu/oaf"
xmlns:oai="http://www.openarchives.org/OAI/2.0/"
xmlns:prov="http://www.openarchives.org/OAI/2.0/provenance" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<oai:header>
<dri:objIdentifier>od______2294::00029b7f0a2a7e090e55b625a9079d83</dri:objIdentifier>
<dri:recordIdentifier>oai:pub.uni-bielefeld.de:2578942</dri:recordIdentifier>
<dri:dateOfCollection>2018-11-23T15:15:33.974+01:00</dri:dateOfCollection>
<oaf:datasourceprefix>od______2294</oaf:datasourceprefix>
<identifier xmlns="http://www.openarchives.org/OAI/2.0/">oai:pub.uni-bielefeld.de:2578942</identifier>
<datestamp xmlns="http://www.openarchives.org/OAI/2.0/">2018-07-24T13:01:16Z</datestamp>
<setSpec xmlns="http://www.openarchives.org/OAI/2.0/">conference</setSpec>
<setSpec xmlns="http://www.openarchives.org/OAI/2.0/">ddc:000</setSpec>
<setSpec xmlns="http://www.openarchives.org/OAI/2.0/">conferenceFtxt</setSpec>
<setSpec xmlns="http://www.openarchives.org/OAI/2.0/">driver</setSpec>
<setSpec xmlns="http://www.openarchives.org/OAI/2.0/">open_access</setSpec>
</oai:header>
<metadata xmlns="http://www.openarchives.org/OAI/2.0/">
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/oai_dc/"
xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
<dc:title>Mobile recommendation agents making online use of visual attention information at the point of sale</dc:title>
<dc:creator>Pfeiffer, Thies</dc:creator>
<dc:creator>Pfeiffer, Jella</dc:creator>
<dc:creator>Meißner, Martin</dc:creator>
<dc:creator>Davis, Fred</dc:creator>
<dc:creator>Riedl, René</dc:creator>
<dc:creator>Jan, vom Brocke</dc:creator>
<dc:creator>Léger, Pierre-Majorique</dc:creator>
<dc:creator>Randolph, Adriane</dc:creator>
<dc:subject>Mobile Cognitive Assistance Systems
Information Systems</dc:subject>
<dc:subject>ddc:000</dc:subject>
<dc:description>We aim to utilize online information about visual attention for developing mobile recommendation agents (RAs) for use at the point of sale. Up to now, most RAs are focussed exclusively at personalization in an e-commerce setting. Very little is known, however, about mobile RAs that offer information and assistance at the point of sale based on individual-level feature based preference models (Murray and Häubl 2009). Current attempts provide information about products at the point of sale by manually scanning barcodes or using RFID (Kowatsch et al. 2011, Heijden 2005), e.g. using specific apps for smartphones. We argue that an online access to the current visual attention of the user offers a much larger potential. Integrating mobile eye tracking into ordinary glasses would yield a direct benefit of applying neuroscience methods in the user’s everyday life. First, learning from consumers’ attentional processes over time and adapting recommendations based on this learning allows us to provide very accurate and relevant recommendations, potentially increasing the perceived usefulness. Second, our proposed system needs little explicit user input (no scanning or navigation on screen) making it easy to use. Thus, instead of learning from click behaviour and past customer ratings, as it is the case in the e-commerce setting, the mobile RA learns from eye movements by participating online in every day decision processes. We argue that mobile RAs should be built based on current research in human judgment and decision making (Murray et al. 2010). In our project, we therefore follow a two-step approach: In the empirical basic research stream, we aim to understand the user’s interaction with the product shelf: the actions and patterns of user’s behaviour (eye movements, gestures, approaching a product closer) and their correspondence to the user’s informational needs. In the empirical system development stream, we create prototypes of mobile RAs and test experimentally the factors that influence the user’s adoption. For example, we suggest that a user’s involvement in the process, such as a need for exact nutritional information or for assistance (e.g., reading support for elderly) will influence the user’s intention to use such as system. The experiments are conducted both in our immersive virtual reality supermarket presented in a CAVE, where we can also easily display information to the user and track the eye movement in great accuracy, as well as in real-world supermarkets (see Figure 1), so that the findings can be better generalized to natural decision situations (Gidlöf et al. 2013). In a first pilot study with five randomly chosen participants in a supermarket, we evaluated which sort of mobile RAs consumers favour in order to get a first impression of the user’s acceptance of the technology. Figure 1 shows an excerpt of one consumer’s eye movements during a decision process. First results show long eye cascades and short fixations on many products in situations where users are uncertain and in need for support. Furthermore, we find a surprising acceptance of the technology itself throughout all ages (23 – 61 years). At the same time, consumers express serious fear of being manipulated by such a technology. For that reason, they strongly prefer the information to be provided by trusted third party or shared with family members and friends (see also Murray and Häubl 2009). Our pilot will be followed by a larger field experiment in March in order to learn more about factors that influence the user’s acceptance as well as the eye movement patterns that reflect typical phases of decision processes and indicate the need for support by a RA.</dc:description>
<dc:date>2013</dc:date>
<dc:type>info:eu-repo/semantics/conferenceObject</dc:type>
<dc:type>doc-type:conferenceObject</dc:type>
<dc:type>text</dc:type>
<dc:identifier>https://pub.uni-bielefeld.de/record/2578942</dc:identifier>
<dc:identifier>https://pub.uni-bielefeld.de/download/2578942/2602478</dc:identifier>
<dc:source>Pfeiffer T, Pfeiffer J, Meißner M. Mobile recommendation agents making online use of visual attention information at the point of sale. In: Davis F, Riedl R, Jan vom B, Léger P-M, Randolph A, eds. <em>Proceedings of the Gmunden Retreat on NeuroIS 2013</em>. 2013: 3-3.</dc:source>
<dc:language>eng</dc:language>
<dc:rights>info:eu-repo/semantics/openAccess</dc:rights>
<record xmlns="http://www.openarchives.org/OAI/2.0/">
<header>
<identifier>oai:lib.psnc.pl:278</identifier>
<datestamp>2011-08-25T15:17:13Z</datestamp>
<setSpec>PSNCRepository:PSNCExternalRepository:exhibitions</setSpec>
<setSpec>PSNCRepository:PSNCExternalRepository:Departments</setSpec>
<setSpec>PSNCRepository:PSNCExternalRepository:Departments:NetworkServices</setSpec>
<setSpec>PSNCRepository:PSNCExternalRepository</setSpec>
<setSpec>PSNCRepository:PSNCExternalRepository:publications</setSpec>
<setSpec>PSNCRepository</setSpec>
</header>
<metadata>
<oai_dc:dc xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/"
xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
<dc:title xml:lang="pl">
<![CDATA[Distributed Search Mechanisms in dLibra Digital Library Framework]]></dc:title>
<dc:creator xml:lang="pl"><![CDATA[Mazurek, Cezary]]></dc:creator>
<dc:creator xml:lang="pl"><![CDATA[Werla, Marcin]]></dc:creator>
<dc:date xml:lang="pl"><![CDATA[2005.10]]></dc:date>
<dc:type xml:lang="pl"><![CDATA[plakat]]></dc:type>
<dc:format xml:lang="pl"><![CDATA[image/jpeg]]></dc:format>
<dc:identifier><![CDATA[https://lib.psnc.pl/dlibra/docmetadata?showContent=true&id=278]]></dc:identifier>
<dc:identifier><![CDATA[oai:lib.psnc.pl:278]]></dc:identifier>
<dc:source xml:lang="pl">
<![CDATA[CERN Workshop on Innovations in Scholarly Communication (OAI4)]]></dc:source>
<dc:language xml:lang="pl"><![CDATA[eng]]></dc:language>
<dc:relation><![CDATA[oai:lib.psnc.pl:publication:339]]></dc:relation>
</oai_dc:dc>
</metadata>
<about xmlns="">
<provenance xmlns="http://www.openarchives.org/OAI/2.0/provenance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/provenance http://www.openarchives.org/OAI/2.0/provenance.xsd">
<originDescription altered="true" harvestDate="2018-11-23T15:15:33.974+01:00">
<baseURL>http://pub.uni-bielefeld.de/oai</baseURL>
<identifier>oai:pub.uni-bielefeld.de:2578942</identifier>
<datestamp>2018-07-24T13:01:16Z</datestamp>
<metadataNamespace>http://www.openarchives.org/OAI/2.0/oai_dc/</metadataNamespace>
</originDescription>
</provenance>
<oaf:datainfo>
<oaf:inferred>false</oaf:inferred>
<oaf:deletedbyinference>false</oaf:deletedbyinference>
<oaf:trust>0.9</oaf:trust>
<oaf:inferenceprovenance/>
<oaf:provenanceaction classid="sysimport:crosswalk:repository"
classname="sysimport:crosswalk:repository"
schemeid="dnet:provenanceActions" schemename="dnet:provenanceActions"/>
</oaf:datainfo>
</about>
</oai:record>
</record>