semantica/docs/api.md

API Reference

Complete API documentation for Semantica.

Core Classes

Semantica

Main framework class for building semantic layers and knowledge graphs.

from semantica import Semantica

semantica = Semantica(config=None)

Methods

build_knowledge_base(sources, **kwargs)

Build a knowledge base from one or more data sources.

Parameters:

• sources (List[str] | str): Data source(s) - file paths or URLs
• embeddings (bool): Generate embeddings (default: True)
• graph (bool): Build knowledge graph (default: True)
• normalize (bool): Normalize data (default: True)

Returns:

• Dict[str, Any]: Dictionary containing:
  • knowledge_graph: Knowledge graph data
  • embeddings: Embedding vectors
  • metadata: Processing metadata
  • statistics: Processing statistics

Example:

result = semantica.build_knowledge_base(
    sources=["document.pdf"],
    embeddings=True,
    graph=True
)
kg = result["knowledge_graph"]

process_document(source)

Process a single document.

Parameters:

• source (str): File path or URL

Returns:

• Dict[str, Any]: Processed document data

extract_entities(text)

Extract entities from text.

Parameters:

• text (str): Input text

Returns:

• Dict[str, List]: Dictionary with entities list

extract_relationships(text)

Extract relationships from text.

Parameters:

• text (str): Input text

Returns:

• Dict[str, List]: Dictionary with relationships list

---

Knowledge Graph Module

semantica.kg

Knowledge graph construction and analysis.

Methods

build_graph(sources)

Build a knowledge graph from sources.

kg = semantica.kg.build_graph(["document.pdf"])

analyze(graph)

Analyze a knowledge graph.

analysis = semantica.kg.analyze(kg)
print(analysis["statistics"])

visualize(graph, output_path=None)

Visualize a knowledge graph.

semantica.kg.visualize(kg, output_path="graph.html")

merge(graphs)

Merge multiple knowledge graphs.

merged = semantica.kg.merge([kg1, kg2, kg3])

---

Semantic Extraction Module

semantica.semantic_extract

Entity and relationship extraction.

Methods

extract_entities(text)

Extract named entities from text.

result = semantica.semantic_extract.extract_entities(text)
entities = result["entities"]

extract_relationships(text)

Extract relationships from text.

result = semantica.semantic_extract.extract_relationships(text)
relationships = result["relationships"]

extract_triples(text)

Extract subject-predicate-object triples.

result = semantica.semantic_extract.extract_triples(text)
triples = result["triples"]

---

Embeddings Module

semantica.embeddings

Embedding generation and management.

Methods

generate(text)

Generate embedding for a single text.

embedding = semantica.embeddings.generate("Your text here")

generate_batch(texts)

Generate embeddings for multiple texts.

texts = ["text1", "text2", "text3"]
embeddings = semantica.embeddings.generate_batch(texts)

---

Export Module

semantica.export

Export knowledge graphs to various formats.

Methods

to_rdf(kg, path)

Export to RDF format.

semantica.export.to_rdf(kg, "output.rdf")

to_json(kg, path)

Export to JSON format.

semantica.export.to_json(kg, "output.json")

to_csv(kg, path)

Export to CSV format.

semantica.export.to_csv(kg, "output.csv")

to_owl(kg, path)

Export to OWL ontology format.

semantica.export.to_owl(kg, "output.owl")

to_yaml(kg, path)

Export to YAML format.

semantica.export.to_yaml(kg, "output.yaml")

---

Conflict Resolution Module

semantica.conflicts

Conflict detection and resolution.

Classes

ConflictResolver

from semantica.conflicts import ConflictResolver

resolver = ConflictResolver(default_strategy="voting")

Methods:

• resolve_conflicts(conflicts): Resolve multiple conflicts
• resolve_conflict(conflict, strategy=None): Resolve a single conflict
• set_resolution_rule(property, strategy): Set custom resolution rules

Example:

resolver = ConflictResolver(default_strategy="voting")
resolved = resolver.resolve_conflicts(conflicts)

---

Configuration

Config

Configuration class for Semantica.

from semantica import Config

config = Config(
    embeddings=True,
    graph=True,
    normalize=True,
    conflict_resolution="voting"
)

semantica = Semantica(config=config)

Parameters:

• embeddings (bool): Enable embedding generation
• graph (bool): Enable knowledge graph construction
• normalize (bool): Enable data normalization
• conflict_resolution (str): Default conflict resolution strategy

---

Full Documentation

For complete module documentation, see:

• MODULES_DOCUMENTATION.md - Detailed module documentation
• GitHub Repository - Source code