diff --git a/main.ipynb b/main.ipynb
index 18aaee7883f49cddfd1767223ed9274064b2880f..9f77ed8fa13d24cc4678dc2f20cb320cbed7dfc6 100644
--- a/main.ipynb
+++ b/main.ipynb
@@ -5,8 +5,8 @@
"execution_count": 1,
"metadata": {
"ExecuteTime": {
- "end_time": "2023-05-05T16:58:30.285996286Z",
- "start_time": "2023-05-05T16:58:27.520729058Z"
+ "end_time": "2023-07-19T08:57:37.490880538Z",
+ "start_time": "2023-07-19T08:57:35.005710556Z"
}
},
"outputs": [],
@@ -38,8 +38,8 @@
"execution_count": 2,
"metadata": {
"ExecuteTime": {
- "end_time": "2023-05-05T16:58:33.643656486Z",
- "start_time": "2023-05-05T16:58:30.289546804Z"
+ "end_time": "2023-07-19T08:57:46.777266756Z",
+ "start_time": "2023-07-19T08:57:37.493551556Z"
}
},
"outputs": [],
@@ -51,24 +51,22 @@
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": 3,
"metadata": {
"ExecuteTime": {
- "end_time": "2023-05-04T23:11:23.647059226Z",
- "start_time": "2023-05-04T23:08:45.535290656Z"
+ "end_time": "2023-07-19T09:00:31.330783432Z",
+ "start_time": "2023-07-19T08:57:46.780973332Z"
}
},
"outputs": [
{
"data": {
+ "text/plain": " 0%| | 0/21 [00:00<?, ?it/s]",
"application/vnd.jupyter.widget-view+json": {
- "model_id": "dc8966ab2d924576890dd6f0598da481",
"version_major": 2,
- "version_minor": 0
- },
- "text/plain": [
- " 0%| | 0/21 [00:00<?, ?it/s]"
- ]
+ "version_minor": 0,
+ "model_id": "4d5d3d0f0d60477d840af6d66d63615a"
+ }
},
"metadata": {},
"output_type": "display_data"
@@ -88,13 +86,13 @@
"Loading o1\n",
"Loading o2\n",
"0\n",
- "ontology iterations: 1804, (0.0, 0, 0.0), aligns: 0, po1: 22, po2: 41\n",
+ "ontology iterations: 1804, (0, 0, 0.0), aligns: 0, po1: 22, po2: 41\n",
"----------------------------------------------------------------------------------------------------\n",
"cmt.owl Conference.owl\n",
"Loading o1\n",
"Loading o2\n",
"3\n",
- "ontology iterations: 7552, (0.25, 0.3333333333333333, 0.28571428571428575), aligns: 3, po1: 59, po2: 64\n",
+ "ontology iterations: 7552, (0.3333333333333333, 0.3333333333333333, 0.3333333333333333), aligns: 3, po1: 59, po2: 64\n",
"----------------------------------------------------------------------------------------------------\n",
"cmt.owl iasted.owl\n",
"Loading o1\n",
@@ -166,13 +164,13 @@
"Loading o1\n",
"Loading o2\n",
"2\n",
- "ontology iterations: 3068, (0.6666666666666666, 1.0, 0.8), aligns: 2, po1: 59, po2: 26\n",
+ "ontology iterations: 3068, (1.0, 1.0, 1.0), aligns: 2, po1: 59, po2: 26\n",
"----------------------------------------------------------------------------------------------------\n",
"Conference.owl edas.owl\n",
"Loading o1\n",
"Loading o2\n",
"3\n",
- "ontology iterations: 6400, (0.6666666666666666, 0.6666666666666666, 0.6666666666666666), aligns: 3, po1: 64, po2: 50\n",
+ "ontology iterations: 6400, (1.0, 0.6666666666666666, 0.8), aligns: 3, po1: 64, po2: 50\n",
"----------------------------------------------------------------------------------------------------\n",
"cmt.owl edas.owl\n",
"Loading o1\n",
@@ -190,7 +188,7 @@
"Loading o1\n",
"Loading o2\n",
"3\n",
- "ontology iterations: 3328, (0.5, 0.3333333333333333, 0.4), aligns: 3, po1: 64, po2: 26\n",
+ "ontology iterations: 3328, (1.0, 0.3333333333333333, 0.5), aligns: 3, po1: 64, po2: 26\n",
"----------------------------------------------------------------------------------------------------\n",
"confOf.owl edas.owl\n",
"Loading o1\n",
@@ -203,7 +201,7 @@
"Loading o2\n",
"4\n",
"ontology iterations: 4608, (1.0, 0.5, 0.6666666666666666), aligns: 4, po1: 64, po2: 36\n",
- "iterations: 74590, (0.6857142857142857, 0.5217391304347826, 0.5925925925925927)\n"
+ "iterations: 74590, (0.8275862068965517, 0.5217391304347826, 0.64)\n"
]
}
],
@@ -213,23 +211,23 @@
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": 4,
"metadata": {
"ExecuteTime": {
- "end_time": "2023-05-04T23:13:14.075532740Z",
- "start_time": "2023-05-04T23:13:13.788604059Z"
+ "end_time": "2023-06-01T13:30:47.366350949Z",
+ "start_time": "2023-06-01T13:30:46.430444165Z"
}
},
"outputs": [
{
- "ename": "NameError",
- "evalue": "name 'results' is not defined",
+ "ename": "TypeError",
+ "evalue": "'float' object is not iterable",
"output_type": "error",
"traceback": [
"\u001B[0;31m---------------------------------------------------------------------------\u001B[0m",
- "\u001B[0;31mNameError\u001B[0m Traceback (most recent call last)",
- "Cell \u001B[0;32mIn[1], line 1\u001B[0m\n\u001B[0;32m----> 1\u001B[0m p, r, f \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mzip\u001B[39m(\u001B[38;5;241m*\u001B[39m\u001B[43mresults\u001B[49m)\n\u001B[1;32m 3\u001B[0m x \u001B[38;5;241m=\u001B[39m np\u001B[38;5;241m.\u001B[39marange(\u001B[38;5;241m0.0\u001B[39m, \u001B[38;5;241m1\u001B[39m, \u001B[38;5;241m0.01\u001B[39m)\n\u001B[1;32m 5\u001B[0m plt\u001B[38;5;241m.\u001B[39mplot(x, p, label\u001B[38;5;241m=\u001B[39m\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mprecision\u001B[39m\u001B[38;5;124m\"\u001B[39m)\n",
- "\u001B[0;31mNameError\u001B[0m: name 'results' is not defined"
+ "\u001B[0;31mTypeError\u001B[0m Traceback (most recent call last)",
+ "Cell \u001B[0;32mIn[4], line 1\u001B[0m\n\u001B[0;32m----> 1\u001B[0m p, r, f \u001B[38;5;241m=\u001B[39m \u001B[38;5;28;43mzip\u001B[39;49m\u001B[43m(\u001B[49m\u001B[38;5;241;43m*\u001B[39;49m\u001B[43mresults\u001B[49m\u001B[43m)\u001B[49m\n\u001B[1;32m 3\u001B[0m x \u001B[38;5;241m=\u001B[39m np\u001B[38;5;241m.\u001B[39marange(\u001B[38;5;241m0.0\u001B[39m, \u001B[38;5;241m1\u001B[39m, \u001B[38;5;241m0.01\u001B[39m)\n\u001B[1;32m 5\u001B[0m plt\u001B[38;5;241m.\u001B[39mplot(x, p, label\u001B[38;5;241m=\u001B[39m\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mprecision\u001B[39m\u001B[38;5;124m\"\u001B[39m)\n",
+ "\u001B[0;31mTypeError\u001B[0m: 'float' object is not iterable"
]
}
],
diff --git a/main.py b/main.py
new file mode 100644
index 0000000000000000000000000000000000000000..9c84b8e6de7bee36272c96239ebac37a78255897
--- /dev/null
+++ b/main.py
@@ -0,0 +1,98 @@
+from sentence_transformers import SentenceTransformer
+from models import Finbank
+import random
+import torch
+import numpy as np
+from property_matching import PropertyMatcher
+from tqdm.auto import tqdm
+from property_matching import most_common_pair
+import matplotlib.pyplot as plt
+import argparse
+import rdflib
+import tempfile
+from urllib import parse, request
+from om.ont import get_namespace
+
+
+def parse_arguments():
+ arg_parser = argparse.ArgumentParser(description='LD similarity.')
+
+ arg_parser.add_argument('source', help='Source ontology path.')
+ arg_parser.add_argument('target', help='Target ontology path.')
+ arg_parser.add_argument('--output', dest='output', default='./output', help='Folder to save the results.')
+ arg_parser.add_argument('--format', dest='format', default='align', choices=['align', 'sssom'], help='Output format.')
+
+ return arg_parser.parse_args()
+
+
+def toAlignFormat(aligns, onto1, onto2, location1, location2):
+ data = ["""<?xml version='1.0' encoding='utf-8' standalone='no'?>
+<rdf:RDF xmlns='http://knowledgeweb.semanticweb.org/heterogeneity/alignment#'
+ xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#'
+ xmlns:xsd='http://www.w3.org/2001/XMLSchema#'
+ xmlns:align='http://knowledgeweb.semanticweb.org/heterogeneity/alignment#'>"""]
+
+ data.append(f""" <Alignment>
+ <xml>yes</xml>
+ <level>0</level>
+ <type>**</type>
+ <onto1>
+ <Ontology rdf:about="{onto1}">
+ <location>{location1}</location>
+ </Ontology>
+ </onto1>
+ <onto2>
+ <Ontology rdf:about="{onto2}">
+ <location>{location2}</location>
+ </Ontology>
+ </onto2>""")
+
+ for (entity1, entity2), confidence in aligns.items():
+ data.append(f""" <map>
+ <Cell>
+ <entity1 rdf:resource="{entity1}"/>
+ <entity2 rdf:resource="{entity2}"/>
+ <relation>=</relation>
+ <measure rdf:datatype="http://www.w3.org/2001/XMLSchema#float">{confidence}</measure>
+ </Cell>
+ </map>""")
+
+ data.append(""" </Alignment>
+</rdf:RDF>""")
+
+ return '\n'.join(data)
+
+def ssom(aligns):
+ lines = ['subject_id\tpredicate_id\tobject_id\tmapping_justification\tconfidence']
+ for (entity1, entity2), confidence in aligns.items():
+ lines.append(f"{entity1}\tskos:exactMatch\t{entity2}\tsemapv:LexicalMatching\t{confidence}")
+
+ return "\n".join(lines)
+
+if __name__ == '__main__':
+ args = parse_arguments()
+ wm = Finbank('/home/guilherme/Documents/kg/fin.bin')
+ model = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
+ property_matcher = PropertyMatcher(wm, model)
+
+ o1 = rdflib.Graph().parse(args.source)
+ o2 = rdflib.Graph().parse(args.target)
+
+ p, it = property_matcher.match_ontologies(o1, o2, 0.65)
+
+
+
+ # Parser
+
+
+ if args.format == 'sssom':
+ result = ssom(p)
+ suffix = '.tsv'
+ else:
+ result = toAlignFormat(p, get_namespace(o1), get_namespace(o2), args.source, args.target)
+ suffix = '.rdf'
+
+ with tempfile.NamedTemporaryFile('w', prefix='alignment_', suffix=suffix, delete=False) as out_file:
+ out_file.write(result)
+
+ print(parse.urljoin("file:", request.pathname2url(out_file.name)))
diff --git a/property_matching.py b/property_matching.py
index 6a0509c594c9c2ca6ae75f74628344e312660b14..f3bf11f81d6f3c309648bfcc095999cc289111fa 100644
--- a/property_matching.py
+++ b/property_matching.py
@@ -14,6 +14,7 @@ from collections import Counter
from tqdm.auto import tqdm
import math
+
def get_type_h(e, g, ml=1):
if type(e) is Literal:
return [e.datatype]
@@ -283,9 +284,6 @@ def get_prop(e, g, p):
return s, objc
-
-
-
def build_tf_models(o1, o2):
a_entities = set(filter(lambda x: is_property(x, o1), o1.subjects()))
b_entities = set(filter(lambda x: is_property(x, o2), o2.subjects()))
@@ -380,7 +378,6 @@ class PropertyMatcher:
sim = 0
label_confidence = sim
-
if sim_weights:
conf = []
if 0 in sim_weights:
@@ -403,7 +400,6 @@ class PropertyMatcher:
if tr is not None:
trm = [[0, 0] for _ in tr]
-
for r, k1, k2 in tqdm(list(onts(base, ref))):
print('-' * 100)
@@ -457,7 +453,8 @@ class PropertyMatcher:
trm[i][0] += len(pa.intersection(cp))
trm[i][1] += len(cp)
- print(f'ontology iterations: {oi}, {metrics(len(pa.intersection(cp)), len(cp), current_total)}, aligns: {current_total}, po1: {len(a_entities)}, po2: {len(b_entities)}')
+ print(
+ f'ontology iterations: {oi}, {metrics(len(pa.intersection(cp)), len(cp), current_total)}, aligns: {current_total}, po1: {len(a_entities)}, po2: {len(b_entities)}')
# for a1, a2 in pa.intersection(p):
# print(colored('✓', 'green'), get_n(a1, o1), get_n(a2, o2))
@@ -471,8 +468,9 @@ class PropertyMatcher:
# print(colored('X', 'red'), get_n(d1, o1), get_n(a1, o1), get_n(r1, o1), colored('<>', 'green'),
# get_n(d2, o2), get_n(a2, o2), get_n(r2, o2))
- # print(
- # f'ontology iterations: {oi}, {metrics(current_correct, current_pred, current_total)}, aligns: {current_total}, po1: {len(a_entities)}, po2: {len(b_entities)}')
+ print(
+ f'ontology iterations: {oi}, {metrics(current_correct, current_pred, current_total)}, aligns: {current_total}, po1: {len(a_entities)}, po2: {len(b_entities)}')
+
print(f'iterations: {iterations}, {metrics(correct, pred, total)}')
if tr is not None:
res = []
@@ -540,6 +538,4 @@ class PropertyMatcher:
pm[iv1] = (iv2, sim)
pm[iv2] = (iv1, sim)
-
return p, iterations
-