diff --git a/src/absinth.py b/src/absinth.py
index 02fd7ae2a74b79983b76fedb24c2e70cdc22abd5..c9cef83300d3d10011b645351333baaecc539b42 100644
--- a/src/absinth.py
+++ b/src/absinth.py
@@ -1,16 +1,27 @@
-import os # for reading files
import sys
+
print('[A] Loading ' + sys.argv[0] + '.\n')
-import spacy # for nlp
+
+import os # for reading files
import networkx as nx # for visualisation
-import matplotlib.pyplot as plt # for visualisation
from copy import deepcopy
from nltk.corpus import stopwords
import numpy as np # for calculations
+
+import config
+
+import spacy # for nlp
nlp = spacy.load('en') # standard english nlp
-def frequencies(corpus_path, target, stop_words=[], allowed_tags=['NN','NNS','JJ','JJS','JJR','NNP','VBZ','VBG'], min_context_size = 2, max_nodes=100000, max_edges=10000000):
+
+def frequencies(corpus_path, target):
+
+ stop_words = set(stopwords.words('english') + config.stop_words)
+ allowed_tags = config.allowed_tags
+ min_context_size = config.min_context_size
+ max_nodes = config.max_nodes
+ max_edges = config.max_edges
node_freq = dict()
edge_freq = dict()
@@ -89,7 +100,11 @@ def frequencies(corpus_path, target, stop_words=[], allowed_tags=['NN','NNS','JJ
return node_freq, edge_freq
-def build_graph(node_freq, edge_freq, min_node_freq=10, min_edge_freq=5, max_weight= 0.9):
+def build_graph(node_freq, edge_freq):
+
+ min_node_freq = config.min_node_freq
+ min_edge_freq = config.min_edge_freq
+ max_weight = config.max_weight
G = nx.Graph()
@@ -115,6 +130,9 @@ def build_graph(node_freq, edge_freq, min_node_freq=10, min_edge_freq=5, max_wei
def root_hubs(graph, edge_freq, min_neighbors=4, theshold=0.8):
+ min_neighbors = config.min_neighbors
+ threshold = config.threshold
+
G = deepcopy(graph)
V = sorted(G.nodes, key=lambda key: G.degree[key], reverse=True) # -1 to sort descending (...3 -> 2 -> 1...)
H = list()
@@ -240,20 +258,12 @@ def disambiguate(mst, hubs, contexts):
return result
-def backup(contexts):
-
- pass
-
-
if __name__ == '__main__':
- data_path = '/home/students/zimmermann/Courses/ws17/fsem/absinth/WSI-Evaluator/datasets/MORESQUE/'
- #corpus_path = '/home/students/zimmermann/Courses/ws17/fsem/absinth/test'
- corpus_path = '/proj/absinth/wikipedia_reduced/'
- results_path = '/home/students/zimmermann/Courses/ws17/fsem/absinth/clustering/'
-
- stop = set(stopwords.words('english') + ['utc', "'s", 'new', 'other', 'talk', 'wikipedia', 'article', 'topic', 'page', 'editors', 'encyclopedia', 'free'])
+ corpus_path = config.corpus
+ data_path = config.dataset
+ output_path = config.output
results = dict()
@@ -274,35 +284,38 @@ if __name__ == '__main__':
with open(data_path+'topics.txt', 'r') as topics_file:
- already_processed = [f.replace('.absinth', '') for f in os.listdir(results_path)]
-
- for line in topics_file.readlines()[1:5]:
+ for line in topics_file.readlines()[1:]:
l = line.split('\t')
- if l[1] not in already_processed:
- topics[l[0]] = l[1]
+ topics[l[0]] = l[1]
for key, value in topics.items():
o_target = value.strip() #original target
print("[A] Processing '"+o_target+"'.\n")
+
if o_target[:4] == 'the_' and o_target.count('_') >= 2: #hard coded 'the'-protection
+
target = o_target[4:]
+
else:
+
target = o_target
- f = open(results_path+target+'.absinth', 'w')
+ f = open(output_path+target+'.absinth', 'w')
f.write('subTopicID\tresultID\n')
print('[A] Counting Tokens...')
- node_freq, edge_freq = frequencies(corpus_path, target, stop)
+ node_freq, edge_freq = frequencies(corpus_path, target)
print('\n[A] Building Graph.\n')
G = build_graph(node_freq, edge_freq)
print('[A] Collecting Root Hubs...')
H = root_hubs(G, edge_freq)
+
for h in H:
+
mfn = sorted(G.adj[h], key=lambda key: edge_freq[h,key] if h < key else edge_freq[key, h], reverse=True)[:6]
print(' {}: {}'.format(h, mfn))
@@ -314,6 +327,7 @@ if __name__ == '__main__':
print(' Mapping:', D, '\n')
print('[A] Writing to file '+o_target+'.absinth.\n\n')
+
for d in D:
f.write(key+'.'+str(d[0])+'\t'+key+'.'+str(d[1])+'\n')
diff --git a/src/config.py b/src/config.py
new file mode 100644
index 0000000000000000000000000000000000000000..ecda9887ce494107b830108ad75695cc7c15c889
--- /dev/null
+++ b/src/config.py
@@ -0,0 +1,46 @@
+'''
+Configuration file
+'''
+
+'''
+Choose paths for corpus, dataset and output.
+- The output directory should be empty when starting absinth.
+'''
+corpus = "/proj/absinth/wikipedia_reduced/"
+dataset = "../WSI-Evaluator/datasets/MORESQUE/"
+output = "../output/"
+
+'''
+Choose stop words and allowed pos-tags.
+- Stop words will not be considered for nodes.
+- Only tokens with allowed pos-tags will be considered.
+'''
+stop_words = ['utc', "'s", 'new', 'other', 'talk', 'wikipedia', 'article', 'topic', 'page', 'editors', 'encyclopedia', 'free']
+allowed_tags=['NN','NNS','JJ','JJS','JJR','NNP','VBZ','VBG']
+
+'''
+Choose the maximum number of nodes and edges that should be considered before building the graph.
+'''
+max_nodes = 100000
+max_edges = 10000000
+
+'''
+Choose the minimum context size.
+'''
+min_context_size = 4
+
+'''
+Choose filters for building the graph.
+- Only considers occurrences/cooccurrences for nodes/edges, that occur more often than these values.
+- Only considers edges with a weight beneath the maximum weight
+'''
+min_node_freq = 10
+min_edge_freq = 5
+max_weight = 0.9
+
+'''
+Choose minimum number of neighbors and maximum median weight of the most frequent neighbors of a node for root hubs.
+- the threshold is calculated using the media of the same number of neighbors declared in min_neighbors.
+'''
+min_neighbors = 6
+theshold = 0.8