How ahmetaa/zemberek-nlp works
Zemberek-NLP is a Turkish NLP library and service suite. A typical analysis workflow accepts text, tokenizes it, and applies morphology to produce analyses; sentence analysis can also disambiguate. Other documented capabilities include normalization, named-entity recognition, classification, language identification, and language modeling. Console applications and a gRPC server expose selected functions. The graph preserves those documented branches; several subsystem internals are represented at their module level because their implementation was not sampled.
Main entry points: TurkishMorphology.java Morphology API · RuleBasedAnalyzer.java Morphological Analysis · TurkishMorphotactics.java Turkish Morphotactics · TurkishTokenizer.java Tokenization
Components 21
Interfaces
- Console Apps
apps/src/main/java/zemberek/apps/ConsoleApp.java - gRPC Server
grpc/src/main/java/zemberek/grpc/server/ZemberekGrpcServer.java - Morphology Service
grpc/src/main/java/zemberek/grpc/server/MorphologyServiceImpl.java - Preprocessing Service
grpc/src/main/java/zemberek/grpc/server/PreprocessingServiceImpl.java - Language ID Service
grpc/src/main/java/zemberek/grpc/server/LanguageIdServiceImpl.java
Text Processing
- Tokenization
tokenization/src/main/java/zemberek/tokenization/TurkishTokenizer.java - Sentence Detection
tokenization/src/main/java/zemberek/tokenization/TurkishSentenceExtractor.java - Normalization
normalization/src/main/java/zemberek/normalization/TurkishSentenceNormalizer.java
Language Analysis
- Morphology API
morphology/src/main/java/zemberek/morphology/TurkishMorphology.java - Morphological Analysis
morphology/src/main/java/zemberek/morphology/analysis/RuleBasedAnalyzer.java - Turkish Morphotactics
morphology/src/main/java/zemberek/morphology/morphotactics/TurkishMorphotactics.java - Root Lexicon
morphology/src/main/java/zemberek/morphology/lexicon/RootLexicon.java - Word Generation
morphology/src/main/java/zemberek/morphology/generator/WordGenerator.java
Models
- Named Entities
ner/src/main/java/zemberek/ner/PerceptronNer.java - Text Classification
classification/src/main/java/zemberek/classification/FastTextClassifier.java - Language Identification
lang-id/src/main/java/zemberek/langid/LanguageIdentifier.java - Language Modeling
lm/src/main/java/zemberek/lm/NgramLanguageModel.java - Word Embeddings
core/src/main/java/zemberek/core/embeddings/FastText.java
Core Foundation
- Core Utilities
core/src/main/java/zemberek/core/io/Files.java - Turkish Language Data
core/src/main/java/zemberek/core/turkish/TurkishAlphabet.java
Other components
- Text User
Connections 17
- Morphology Service → Morphology API: analyzes text
grpc/src/main/java/zemberek/grpc/server/MorphologyServiceImpl.java - Preprocessing Service → Tokenization: tokenizes text
grpc/src/main/java/zemberek/grpc/server/PreprocessingServiceImpl.java - Language ID Service → Language Identification: identifies language
grpc/src/main/java/zemberek/grpc/server/LanguageIdServiceImpl.java - Morphology API → Tokenization: tokenizes sentences
morphology/src/main/java/zemberek/morphology/TurkishMorphology.java - Morphology API → Morphological Analysis: delegates analysis
morphology/src/main/java/zemberek/morphology/TurkishMorphology.java - Morphology API → Word Generation: creates generator
morphology/src/main/java/zemberek/morphology/TurkishMorphology.java - Turkish Morphotactics → Root Lexicon: uses lexicon
morphology/src/main/java/zemberek/morphology/morphotactics/TurkishMorphotactics.java - Text User → Console Apps: runs applications
README.md - Text User → gRPC Server: accesses services
README.md - Morphology API → Turkish Morphotactics: configures analysis
morphology/src/main/java/zemberek/morphology/TurkishMorphology.java - Text User → Normalization: normalizes text
README.md - Text User → Named Entities: extracts entities
README.md - Text User → Text Classification: classifies text
README.md - Text User → Language Modeling: uses language models
README.md - Text User → Sentence Detection: detects sentences
README.md - Word Embeddings → Core Utilities: uses utilities
- Morphological Analysis → Turkish Language Data: uses language data