generate keywords and quick solution to handle stopwords

This commit is contained in:
Fabien Benetou
2011-07-29 12:33:24 +02:00
parent 63eee10ebe
commit 379744239c

View File

@@ -0,0 +1,17 @@
#! /usr/bin/env bash
WIKIPATH=.
if [ $# -gt 0 ]
then
WIKIPATH=$1
fi
# lots of cool usage to explore
# automatic category creation
# annotated images
# to facilitate the creation .stopwords run few times
# head ../pub/keywords_distribution/* | sort -n | tail -30
for P in `ls $WIKIPATH -IPmWiki.* -I*RecentChanges -Itotalcounter.stat -I*,del-*`; do grep ^text= $P | sed "s/^text=//" | sed "s/%0a/\\n/g" | sed "s/^(:.*:)//g" | sed "s/^>>.*//g" | sed "s/http.*|//g" | sed "s/[^a-zA-Z]/\\n/g" | sort | uniq -i -c | grep -i -v -f $WIKIPATH/.stopwords | sort -r -n > $WIKIPATH/../pub/keywords_distribution/$P.txt; done