NTCIRtop image

    GeoTime

NTCIR GeoTime 2010-11

GEOTIME Results Format

GEOTIME participants should use the following XML format to submit each result.

Tag

Description

TOPIC_SET

Contains a meta data and a list of topics

METADATA

Must include meta information on run id, and system description. If the run used a manual process to create, please specify RUN_TYPE as "Manual"

TOPIC

Each TOPIC is associated with GEOTIME_RESULT.

GEOTIME_RESULT

Contains a ranked list of DOCUMENT

DOCUMENT

Pointer to the document in the corpus. SCORE is optional but you are recommended to produce this value (preferably between 0 and 1).

For the definition of Run ID, refer to RunIDFormat.

XML DTD

<
   !DOCTYPE TOPIC_SET [
   <!ELEMENT TOPIC_SET (METADATA,TOPIC*)>
   <!ELEMENT METADATA (RUNID,DESCRIPTION,RUN_TYPE?)>
   <!ELEMENT RUNID (#PCDATA)>
   <!ELEMENT DESCRIPTION (#PCDATA)>
   <!ELEMENT RUN_TYPE (#PCDATA)>
   <!ELEMENT TOPIC (GEOTIME_RESULT)>
   <!ATTLIST TOPIC ID CDATA #REQUIRED>
   <!ELEMENT GEOTIME_RESULT (DOCUMENT*)>
   <!ELEMENT DOCUMENT EMPTY>
   <!ATTLIST DOCUMENT RANK CDATA #REQUIRED>
   <!ATTLIST DOCUMENT DOCID CDATA #REQUIRED>
   <!ATTLIST DOCUMENT SCORE CDATA #IMPLIED>
   ]>

Sample -- English NYTimes Monolingual

   
<TOPIC_SET>

<METADATA> <RUNID>TEAMX-EN-EN-01-T</RUNID> <DESCRIPTION>Ranked based on logistic regression with blind feedback and no special geographic or temporal retrieval features</DESCRIPTION> <RUN_TYPE>Automatic</RUN_TYPE> </METADATA>

<TOPIC ID="GeoTime-0001"> <GEOTIME_RESULT> <DOCUMENT RANK="1" DOCID="NYT_ENG_20020128.0287" SCORE="0.97" /> <DOCUMENT RANK="2" DOCID="NYT_ENG_20020128.0182" SCORE="0.95" /> <DOCUMENT RANK="3" DOCID="NYT_ENG_20020128.0286" SCORE="0.93" /> <DOCUMENT RANK="4" DOCID="NYT_ENG_20021224.0073" SCORE="0.91" /> </GEOTIME_RESULT> </TOPIC>

<TOPIC ID="GeoTime-0002"> <GEOTIME_RESULT> <DOCUMENT RANK="1" DOCID="NYT_ENG_20041112.0273" SCORE="0.57" /> <DOCUMENT RANK="2" DOCID="NYT_ENG_20050830.0008" SCORE="0.55" /> </GEOTIME_RESULT> </TOPIC>

</TOPIC_SET>

Sample -- Japanese Mainichi Monolingual

<TOPIC_SET> <METADATA> <RUNID>TEAMX-JA-JA-01-T</RUNID> <DESCRIPTION>Ranked based on cosine similarity of tf.idf weighted term vectors.</DESCRIPTION> <RUN_TYPE>Automatic</RUN_TYPE> </METADATA> <TOPIC ID="GeoTime-0001"> <GEOTIME_RESULT> <DOCUMENT RANK="1" DOCID="JA-010101032" SCORE="1.00" /> <DOCUMENT RANK="2" DOCID="JA-001116222" SCORE="0.92" /> <DOCUMENT RANK="3" DOCID="JA-001110059" SCORE="0.91" /> <DOCUMENT RANK="4" DOCID="JA-990825062" SCORE="0.91" /> </GEOTIME_RESULT> </TOPIC> <TOPIC ID="GeoTime-0002"> <GEOTIME_RESULT> <DOCUMENT RANK="1" DOCID="JA-980512181" SCORE="1.00" /> <DOCUMENT RANK="2" DOCID="JA-980531170" SCORE="0.99" /> </GEOTIME_RESULT> </TOPIC> </TOPIC_SET>