Difference between revisions of "ConferenceCorpus/statistics"

From BITPlan Wiki
Jump to navigation Jump to search
Line 143: Line 143:
 
<source lang='sql'>
 
<source lang='sql'>
 
SELECT  
 
SELECT  
  series,
+
      series,
  min(ordinal) as minOrdinal,  
+
      min(ordinal) as minOrdinal,  
  max(ordinal) as maxOrdinal,
+
      max(ordinal) as maxOrdinal,
  avg(ordinal) as avgOrdinal,
+
      avg(ordinal) as avgOrdinal,
  max(Ordinal)-min(Ordinal) as available,
+
      max(Ordinal)-min(Ordinal) as ordinalRange,
  (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
+
      (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
 
FROM event_dblp
 
FROM event_dblp
Where ordinal is not null  
+
WHERE ordinal is not null  
group by series
+
GROUP BY series
order by 6 desc
+
ORDER by 6 DESC
               
+
                   
 
</source>
 
</source>
=== histogramm ===
+
=== event series completeness of dblp ===
[[File:dblp_series_completeness.png|600px]]
+
[[File:dblp:dblp:event_dblp_series_completeness.png|600px]]
 
+
== OpenResearch ==
== openresearch ==
 
 
=== sql query ===
 
=== sql query ===
 
<source lang='sql'>
 
<source lang='sql'>
 
SELECT  
 
SELECT  
  inEventSeries,
+
      inEventSeries,
  min(ordinal) as minOrdinal,  
+
      min(ordinal) as minOrdinal,  
  max(ordinal) as maxOrdinal,
+
      max(ordinal) as maxOrdinal,
  avg(ordinal) as avgOrdinal,
+
      avg(ordinal) as avgOrdinal,
  max(Ordinal)-min(Ordinal) as available,
+
      max(Ordinal)-min(Ordinal) as ordinalRange,
  (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
+
      (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
 
FROM event_orclone
 
FROM event_orclone
Where ordinal is not null  
+
WHERE ordinal is not null  
group by inEventSeries
+
GROUP BY inEventSeries
order by 6 desc
+
ORDER by 6 DESC
               
+
                   
 
</source>
 
</source>
=== histogramm ===
+
=== event series completeness of orclone ===
[[File:orclone_series_completeness.png|600px]]
+
[[File:OpenResearch:orclone:event_orclone_series_completeness.png|600px]]
 
+
== WikiCFP ==
== tibkat ==
 
=== sql query ===
 
For tibkat a direct SQL query was not possible since the series info is not available.
 
We therefore filtered by series acronym using python code to get an indication. 3136 series where identified this way and then the histogramm was created from the data
 
=== histogramm ===
 
[[File:tibkat_series_completeness.png|600px]]
 
== wikicfp ==
 
 
=== sql query ===
 
=== sql query ===
 
<source lang='sql'>
 
<source lang='sql'>
 
SELECT  
 
SELECT  
  seriesId,
+
      seriesId,
  min(ordinal) as minOrdinal,  
+
      min(ordinal) as minOrdinal,  
  max(ordinal) as maxOrdinal,
+
      max(ordinal) as maxOrdinal,
  avg(ordinal) as avgOrdinal,
+
      avg(ordinal) as avgOrdinal,
  max(Ordinal)-min(Ordinal) as available,
+
      max(Ordinal)-min(Ordinal) as ordinalRange,
  (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
+
      (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
 
FROM event_wikicfp
 
FROM event_wikicfp
Where ordinal is not null  
+
WHERE ordinal is not null  
group by seriesId
+
GROUP BY seriesId
order by 6 desc
+
ORDER by 6 DESC
               
+
                   
 
</source>
 
</source>
=== histogramm ===
+
=== event series completeness of wikicfp ===
[[File:wikicfp_series_completeness.png|600px]]
+
[[File:WikiCFP:wikicfp:event_wikicfp_series_completeness.png|600px]]
 
+
== Wikidata ==
== wikidata ==
 
 
=== sql query ===
 
=== sql query ===
 
<source lang='sql'>
 
<source lang='sql'>
 
SELECT  
 
SELECT  
  eventInSeriesId,
+
      eventInSeriesId,
  min(ordinal) as minOrdinal,  
+
      min(ordinal) as minOrdinal,  
  max(ordinal) as maxOrdinal,
+
      max(ordinal) as maxOrdinal,
  avg(ordinal) as avgOrdinal,
+
      avg(ordinal) as avgOrdinal,
  max(Ordinal)-min(Ordinal) as available,
+
      max(Ordinal)-min(Ordinal) as ordinalRange,
  (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
+
      (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
 
FROM event_wikidata
 
FROM event_wikidata
Where ordinal is not null  
+
WHERE ordinal is not null  
group by eventInSeriesId
+
GROUP BY eventInSeriesId
order by 6 desc
+
ORDER by 6 DESC
               
+
                   
 
</source>
 
</source>
=== histogramm ===
+
=== event series completeness of wikidata ===
[[File:wikidata_series_completeness.png|600px]]
+
[[File:Wikidata:wikidata:event_wikidata_series_completeness.png|600px]]

Revision as of 09:46, 19 May 2022

Ordinal histogramms

The Zipf digramms show the log frequency (leaving out the first event, since typically the decline begins from the second event)

confref

To few available ordinals for analysis.

CEUR-WS

sql query

SELECT ordinal
    FROM event_ceurws
    where ordinal is not null
    and ordinal < 50

ceurws ordinals

Ordinalhistogramm event ceurws.pngZipf event ceurws.png

confref

sql query

SELECT ordinal
    FROM event_confref
    where ordinal is not null
    and ordinal < 50

confref ordinals

Ordinalhistogramm event confref.pngZipf event confref.png

Crossref

sql query

SELECT ordinal
    FROM event_crossref
    where ordinal is not null
    and ordinal < 50

crossref ordinals

Ordinalhistogramm event crossref.pngZipf event crossref.png

dblp

sql query

SELECT ordinal
    FROM event_dblp
    where ordinal is not null
    and ordinal < 50

dblp ordinals

Ordinalhistogramm event dblp.pngZipf event dblp.png

GND

sql query

SELECT ordinal
    FROM event_gnd
    where ordinal is not null
    and ordinal < 50

gnd ordinals

Ordinalhistogramm event gnd.pngZipf event gnd.png

OpenResearch

sql query

SELECT ordinal
    FROM event_or
    where ordinal is not null
    and ordinal < 50

or ordinals

Ordinalhistogramm event or.pngZipf event or.png

OpenResearch

sql query

SELECT ordinal
    FROM event_orbackup
    where ordinal is not null
    and ordinal < 50

orbackup ordinals

Ordinalhistogramm event orbackup.pngZipf event orbackup.png

OpenResearch

sql query

SELECT ordinal
    FROM event_orclone
    where ordinal is not null
    and ordinal < 50

orclone ordinals

Ordinalhistogramm event orclone.pngZipf event orclone.png

OpenResearch

sql query

SELECT ordinal
    FROM event_orclonebackup
    where ordinal is not null
    and ordinal < 50

orclonebackup ordinals

Ordinalhistogramm event orclonebackup.pngZipf event orclonebackup.png

TIBKAT

sql query

SELECT ordinal
    FROM event_tibkat
    where ordinal is not null
    and ordinal < 50

tibkat ordinals

Ordinalhistogramm event tibkat.pngZipf event tibkat.png

WikiCFP

sql query

SELECT ordinal
    FROM event_wikicfp
    where ordinal is not null
    and ordinal < 50

wikicfp ordinals

Ordinalhistogramm event wikicfp.pngZipf event wikicfp.png

Wikidata

sql query

SELECT ordinal
    FROM event_wikidata
    where ordinal is not null
    and ordinal < 50

wikidata ordinals

Ordinalhistogramm event wikidata.pngZipf event wikidata.png

Eventseries completeness

dblp

sql query

SELECT 
       series,
       min(ordinal) as minOrdinal, 
       max(ordinal) as maxOrdinal,
       avg(ordinal) as avgOrdinal,
       max(Ordinal)-min(Ordinal) as ordinalRange,
       (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
FROM event_dblp
WHERE ordinal is not null 
GROUP BY series
ORDER by 6 DESC

event series completeness of dblp

600px

OpenResearch

sql query

SELECT 
       inEventSeries,
       min(ordinal) as minOrdinal, 
       max(ordinal) as maxOrdinal,
       avg(ordinal) as avgOrdinal,
       max(Ordinal)-min(Ordinal) as ordinalRange,
       (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
FROM event_orclone
WHERE ordinal is not null 
GROUP BY inEventSeries
ORDER by 6 DESC

event series completeness of orclone

600px

WikiCFP

sql query

SELECT 
       seriesId,
       min(ordinal) as minOrdinal, 
       max(ordinal) as maxOrdinal,
       avg(ordinal) as avgOrdinal,
       max(Ordinal)-min(Ordinal) as ordinalRange,
       (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
FROM event_wikicfp
WHERE ordinal is not null 
GROUP BY seriesId
ORDER by 6 DESC

event series completeness of wikicfp

600px

Wikidata

sql query

SELECT 
       eventInSeriesId,
       min(ordinal) as minOrdinal, 
       max(ordinal) as maxOrdinal,
       avg(ordinal) as avgOrdinal,
       max(Ordinal)-min(Ordinal) as ordinalRange,
       (max(Ordinal)-min(Ordinal)) /(max(Ordinal)-1.0) as completeness
FROM event_wikidata
WHERE ordinal is not null 
GROUP BY eventInSeriesId
ORDER by 6 DESC

event series completeness of wikidata

600px