Source code of MERCKX (Multilingual Entity/Resource Combiner & Knowledge eXtractor) and related material for the Historische Kranten project (http://historischekranten.be)
Demo available at http://mastic.ulb.ac.be/ypres/
Run ./merckx-init.sh
to download all resources from DBpedia. This will automatically call merckx-init.py
to generate labels and entities.
Then use python3 merckx.py <type> <filename>
where <type>
is the entity type (EVE, LOC, PER, or ORG) and <filename>
is the name of text file to analyze.
For instance, python3 merckx.py LOC example.txt
(first part of the Wikipedia article about Belgium) will give the following output:
example.txt 35 41 dbr:Belgium
example.txt 72 80 dbr:Belgium
example.txt 208 214 dbr:Europe
example.txt 227 233 dbr:France
example.txt 239 250 dbr:Netherlands
example.txt 506 512 dbr:Europe
example.txt 847 856 dbr:High_Fens
example.txt 914 925 dbr:Netherlands
example.txt 971 977 dbr:France
example.txt 1429 1440 dbr:Netherlands
example.txt 1563 1569 dbr:Europe
example.txt 1861 1869 dbr:Flanders
example.txt 1926 1934 dbr:Wallonia
example.txt 1940 1948 dbr:Brussels
example.txt 2029 2043 dbr:Flemish_Region
example.txt 2094 2102 dbr:Wallonia
example.txt 2695 2703 dbr:Flanders
example.txt 2708 2716 dbr:Wallonia
example.txt 3495 3503 dbr:Brussels
example.txt 3633 3641 dbr:Brussels
example.txt 3804 3812 dbr:Schengen,_Luxembourg