From 74eb760e05d6e4921927c8ff94cf785231f85ea7 Mon Sep 17 00:00:00 2001
From: born <born@cl.uni-heidelberg.de>
Date: Tue, 23 Jul 2019 15:22:52 +0200
Subject: [PATCH] Update README.md
---
README.md | 38 ++++++++++++++++++++++++++++++++++++--
1 file changed, 36 insertions(+), 2 deletions(-)
diff --git a/README.md b/README.md
index 0032ee7..f64ca3d 100644
--- a/README.md
+++ b/README.md
@@ -4,7 +4,7 @@ This repo contains the resources (data and scripts) for the workshop entitled "P
## Repo structure
-* _data/_ contains the untagged -- but annotated -- files of the [Old Bailey Corpus](http://fedora.clarin-d.uni-saarland.de/oldbailey/index.html)
+* _data/_ contains the untagged -- but annotated -- files of the [Old Bailey Corpus](http://fedora.clarin-d.uni-saarland.de/oldbailey/index.html) (OBC)
* _output/_ is used as the default output directory for created knowledge graphs
* _output/gold/_ contains some sample knowledge graphs for the years 1720, 1820, and 1913
* _src/_ contains all scripts
@@ -58,4 +58,38 @@ Install with pip:
**NOTE**: _spacy_ must not be version 2.1.5!
-##
\ No newline at end of file
+## How to create knowledge graphs
+
+The file to interact with to generate knowledge graphs from the OBC is _main.py_.
+
+To re-create the sample knowledge graph for the year 1913, simply type:
+
+ python main.py -year 1913 -output_path "../output/example_graph_1913.json"
+
+To display all available options, type:
+
+ python main.py -h
+
+This gives you:
+
+ usage: main.py [-h] [-year YEAR [YEAR ...]] [-output_path OUTPUT_PATH]
+ [-text_node_simplification_mode {None,classifier,spacy_direct_object}]
+ [-verbose {0,1,2}]
+ [-prune_text_nodes_min_freq PRUNE_TEXT_NODES_MIN_FREQ]
+
+ Building a graph from old bailey corpus.
+
+ optional arguments:
+ -h, --help show this help message and exit
+ -year YEAR [YEAR ...]
+ what year(s) the graph should cover, leave empty to
+ cover all available years
+ -output_path OUTPUT_PATH
+ where to save output graph
+ -text_node_simplification_mode {None,classifier,spacy_direct_object}
+ simplify text nodes, e.g.: "murder of x , on the
+ Monday of(...)" --> "murder"
+ -verbose {0,1,2} logging level
+ -prune_text_nodes_min_freq PRUNE_TEXT_NODES_MIN_FREQ
+ how often a text description node has to occur to be
+ included in the graph
\ No newline at end of file
--
GitLab