File size: 9,680 Bytes
c58df45 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
#semantic_analysis.py
import streamlit as st
import spacy
import networkx as nx
import matplotlib.pyplot as plt
import io
import base64
from collections import Counter, defaultdict
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import logging
logger = logging.getLogger(__name__)
# Define colors for grammatical categories
POS_COLORS = {
'ADJ': '#FFA07A', 'ADP': '#98FB98', 'ADV': '#87CEFA', 'AUX': '#DDA0DD',
'CCONJ': '#F0E68C', 'DET': '#FFB6C1', 'INTJ': '#FF6347', 'NOUN': '#90EE90',
'NUM': '#FAFAD2', 'PART': '#D3D3D3', 'PRON': '#FFA500', 'PROPN': '#20B2AA',
'SCONJ': '#DEB887', 'SYM': '#7B68EE', 'VERB': '#FF69B4', 'X': '#A9A9A9',
}
POS_TRANSLATIONS = {
'es': {
'ADJ': 'Adjetivo', 'ADP': 'Preposición', 'ADV': 'Adverbio', 'AUX': 'Auxiliar',
'CCONJ': 'Conjunción Coordinante', 'DET': 'Determinante', 'INTJ': 'Interjección',
'NOUN': 'Sustantivo', 'NUM': 'Número', 'PART': 'Partícula', 'PRON': 'Pronombre',
'PROPN': 'Nombre Propio', 'SCONJ': 'Conjunción Subordinante', 'SYM': 'Símbolo',
'VERB': 'Verbo', 'X': 'Otro',
},
'en': {
'ADJ': 'Adjective', 'ADP': 'Preposition', 'ADV': 'Adverb', 'AUX': 'Auxiliary',
'CCONJ': 'Coordinating Conjunction', 'DET': 'Determiner', 'INTJ': 'Interjection',
'NOUN': 'Noun', 'NUM': 'Number', 'PART': 'Particle', 'PRON': 'Pronoun',
'PROPN': 'Proper Noun', 'SCONJ': 'Subordinating Conjunction', 'SYM': 'Symbol',
'VERB': 'Verb', 'X': 'Other',
},
'fr': {
'ADJ': 'Adjectif', 'ADP': 'Préposition', 'ADV': 'Adverbe', 'AUX': 'Auxiliaire',
'CCONJ': 'Conjonction de Coordination', 'DET': 'Déterminant', 'INTJ': 'Interjection',
'NOUN': 'Nom', 'NUM': 'Nombre', 'PART': 'Particule', 'PRON': 'Pronom',
'PROPN': 'Nom Propre', 'SCONJ': 'Conjonction de Subordination', 'SYM': 'Symbole',
'VERB': 'Verbe', 'X': 'Autre',
}
}
ENTITY_LABELS = {
'es': {
"Personas": "lightblue",
"Lugares": "lightcoral",
"Inventos": "lightgreen",
"Fechas": "lightyellow",
"Conceptos": "lightpink"
},
'en': {
"People": "lightblue",
"Places": "lightcoral",
"Inventions": "lightgreen",
"Dates": "lightyellow",
"Concepts": "lightpink"
},
'fr': {
"Personnes": "lightblue",
"Lieux": "lightcoral",
"Inventions": "lightgreen",
"Dates": "lightyellow",
"Concepts": "lightpink"
}
}
##############################################################################################################
def perform_semantic_analysis(text, nlp, lang_code):
logger.info(f"Starting semantic analysis for language: {lang_code}")
try:
doc = nlp(text)
# Conceptos clave y grafo de conceptos
key_concepts = identify_key_concepts(doc)
concept_graph = create_concept_graph(doc, key_concepts)
concept_graph_fig = visualize_concept_graph(concept_graph, lang_code)
#concept_graph_html = fig_to_html(concept_graph_fig)
# Entidades y grafo de entidades
entities = extract_entities(doc, lang_code)
entity_graph = create_entity_graph(entities)
entity_graph_fig = visualize_entity_graph(entity_graph, lang_code)
#entity_graph_html = fig_to_html(entity_graph_fig)
logger.info("Semantic analysis completed successfully")
return {
'doc': doc,
'key_concepts': key_concepts,
'concept_graph': concept_graph_fig,
'entities': entities,
'entity_graph': entity_graph_fig
}
except Exception as e:
logger.error(f"Error in perform_semantic_analysis: {str(e)}")
raise
'''
def fig_to_html(fig):
buf = io.BytesIO()
fig.savefig(buf, format='png')
buf.seek(0)
img_str = base64.b64encode(buf.getvalue()).decode()
return f'<img src="data:image/png;base64,{img_str}" />'
'''
def identify_key_concepts(doc):
logger.info("Identifying key concepts")
word_freq = Counter([token.lemma_.lower() for token in doc if token.pos_ in ['NOUN', 'VERB'] and not token.is_stop])
key_concepts = word_freq.most_common(10)
return [(concept, float(freq)) for concept, freq in key_concepts]
def create_concept_graph(doc, key_concepts):
G = nx.Graph()
for concept, freq in key_concepts:
G.add_node(concept, weight=freq)
for sent in doc.sents:
sent_concepts = [token.lemma_.lower() for token in sent if token.lemma_.lower() in dict(key_concepts)]
for i, concept1 in enumerate(sent_concepts):
for concept2 in sent_concepts[i+1:]:
if G.has_edge(concept1, concept2):
G[concept1][concept2]['weight'] += 1
else:
G.add_edge(concept1, concept2, weight=1)
return G
def visualize_concept_graph(G, lang_code):
fig, ax = plt.subplots(figsize=(12, 8))
pos = nx.spring_layout(G, k=0.5, iterations=50)
node_sizes = [G.nodes[node]['weight'] * 100 for node in G.nodes()]
nx.draw_networkx_nodes(G, pos, node_size=node_sizes, node_color='lightblue', alpha=0.8, ax=ax)
nx.draw_networkx_labels(G, pos, font_size=10, font_weight="bold", ax=ax)
edge_weights = [G[u][v]['weight'] for u, v in G.edges()]
nx.draw_networkx_edges(G, pos, width=edge_weights, alpha=0.5, ax=ax)
title = {
'es': "Relaciones entre Conceptos Clave",
'en': "Key Concept Relations",
'fr': "Relations entre Concepts Clés"
}
ax.set_title(title[lang_code], fontsize=16)
ax.axis('off')
plt.tight_layout()
return fig
def create_entity_graph(entities):
G = nx.Graph()
for entity_type, entity_list in entities.items():
for entity in entity_list:
G.add_node(entity, type=entity_type)
for i, entity1 in enumerate(entity_list):
for entity2 in entity_list[i+1:]:
G.add_edge(entity1, entity2)
return G
def visualize_entity_graph(G, lang_code):
fig, ax = plt.subplots(figsize=(12, 8))
pos = nx.spring_layout(G)
for entity_type, color in ENTITY_LABELS[lang_code].items():
node_list = [node for node, data in G.nodes(data=True) if data['type'] == entity_type]
nx.draw_networkx_nodes(G, pos, nodelist=node_list, node_color=color, node_size=500, alpha=0.8, ax=ax)
nx.draw_networkx_edges(G, pos, width=1, alpha=0.5, ax=ax)
nx.draw_networkx_labels(G, pos, font_size=8, font_weight="bold", ax=ax)
ax.set_title(f"Relaciones entre Entidades ({lang_code})", fontsize=16)
ax.axis('off')
plt.tight_layout()
return fig
#################################################################################
def create_topic_graph(topics, doc):
G = nx.Graph()
for topic in topics:
G.add_node(topic, weight=doc.text.count(topic))
for i, topic1 in enumerate(topics):
for topic2 in topics[i+1:]:
weight = sum(1 for sent in doc.sents if topic1 in sent.text and topic2 in sent.text)
if weight > 0:
G.add_edge(topic1, topic2, weight=weight)
return G
def visualize_topic_graph(G, lang_code):
fig, ax = plt.subplots(figsize=(12, 8))
pos = nx.spring_layout(G)
node_sizes = [G.nodes[node]['weight'] * 100 for node in G.nodes()]
nx.draw_networkx_nodes(G, pos, node_size=node_sizes, node_color='lightgreen', alpha=0.8, ax=ax)
nx.draw_networkx_labels(G, pos, font_size=10, font_weight="bold", ax=ax)
edge_weights = [G[u][v]['weight'] for u, v in G.edges()]
nx.draw_networkx_edges(G, pos, width=edge_weights, alpha=0.5, ax=ax)
ax.set_title(f"Relaciones entre Temas ({lang_code})", fontsize=16)
ax.axis('off')
plt.tight_layout()
return fig
###########################################################################################
def generate_summary(doc, lang_code):
sentences = list(doc.sents)
summary = sentences[:3] # Toma las primeras 3 oraciones como resumen
return " ".join([sent.text for sent in summary])
def extract_entities(doc, lang_code):
entities = defaultdict(list)
for ent in doc.ents:
if ent.label_ in ENTITY_LABELS[lang_code]:
entities[ent.label_].append(ent.text)
return dict(entities)
def analyze_sentiment(doc, lang_code):
positive_words = sum(1 for token in doc if token.sentiment > 0)
negative_words = sum(1 for token in doc if token.sentiment < 0)
total_words = len(doc)
if positive_words > negative_words:
return "Positivo"
elif negative_words > positive_words:
return "Negativo"
else:
return "Neutral"
def extract_topics(doc, lang_code):
vectorizer = TfidfVectorizer(stop_words='english', max_features=5)
tfidf_matrix = vectorizer.fit_transform([doc.text])
feature_names = vectorizer.get_feature_names_out()
return list(feature_names)
# Asegúrate de que todas las funciones necesarias estén exportadas
__all__ = [
'perform_semantic_analysis',
'identify_key_concepts',
'create_concept_graph',
'visualize_concept_graph',
'create_entity_graph',
'visualize_entity_graph',
'generate_summary',
'extract_entities',
'analyze_sentiment',
'create_topic_graph',
'visualize_topic_graph',
'extract_topics',
'ENTITY_LABELS',
'POS_COLORS',
'POS_TRANSLATIONS'
] |