How azchaimae/nlp-applications-with-word-embeddings-models-extrinsic-evaluation works
This repository is a research evaluation suite for comparing Arabic word embeddings on NLP tasks, rather than a hosted application. The workflow is to run a task-specific or combined evaluation script against labeled benchmarks and embedding files, then record accuracy and related metrics. The task scripts cover document categorization, named-entity recognition, part-of-speech tagging, and sentiment analysis; the combined FastText script also includes embedding-oriented evaluations such as analogy and clustering tasks. The excerpts show these as standalone scripts. Because they are partial, some internal details and connections are not represented.
Main entry points: Accuracy by NLP tasks_DC.py Document categorization · Accuracy by NLP tasks_NER.py Named entities · Accuracy by NLP tasks_POS.py Part of speech · Accuracy by NLP tasks_SA.py Sentiment analysis
Components 9
Task evaluations
- Document categorization
Accuracy by NLP tasks_DC.py - Named entities
Accuracy by NLP tasks_NER.py - Part of speech
Accuracy by NLP tasks_POS.py - Sentiment analysis
Accuracy by NLP tasks_SA.py
Combined evaluation
- Embedding task suite
FastText All accuracies.py
Other components
- Researcher
- Labeled benchmarks
- Embedding files
- Metric files
Connections 20
- Researcher → Document categorization: runs
README.md - Researcher → Named entities: runs
README.md - Researcher → Part of speech: runs
README.md - Researcher → Sentiment analysis: runs
README.md - Researcher → Embedding task suite: runs
README.md - Labeled benchmarks → Document categorization: provides labels
Accuracy by NLP tasks_DC.py - Labeled benchmarks → Named entities: provides labels
Accuracy by NLP tasks_NER.py - Labeled benchmarks → Part of speech: provides labels
Accuracy by NLP tasks_POS.py - Labeled benchmarks → Sentiment analysis: provides labels
Accuracy by NLP tasks_SA.py - Labeled benchmarks → Embedding task suite: provides labels
FastText All accuracies.py - Embedding files → Document categorization: supplies vectors
Accuracy by NLP tasks_DC.py - Embedding files → Named entities: supplies vectors
Accuracy by NLP tasks_NER.py - Embedding files → Part of speech: supplies vectors
Accuracy by NLP tasks_POS.py - Embedding files → Sentiment analysis: supplies vectors
Accuracy by NLP tasks_SA.py - Embedding files → Embedding task suite: supplies vectors
FastText All accuracies.py - Document categorization → Metric files: writes metrics
Accuracy by NLP tasks_DC.py - Named entities → Metric files: writes metrics
Accuracy by NLP tasks_NER.py - Part of speech → Metric files: writes metrics
Accuracy by NLP tasks_POS.py - Sentiment analysis → Metric files: writes metrics
Accuracy by NLP tasks_SA.py - Embedding task suite → Metric files: writes metrics
FastText All accuracies.py