RealSinaEslami/NimFaseleh

📚 Nim Faseleh Corpus (Persian Zero-Width Non-Joiner Dataset)

🎯 Objective

The Nim Faseleh Corpus is designed to comprehensively cover Persian Zero-Width Non-Joiner (ZWNJ) usage patterns and compound word formations.

ZWNJ is one of the most critical yet under-addressed challenges in Persian Natural Language Processing. This dataset provides more than 14,000 standardized samples to support research and development in this area.

---

🔍 Why this dataset?

In Persian text, correct ZWNJ usage directly impacts both grammatical correctness and semantic meaning.

For example:

• «نمی‌روم» (correct) vs «نمی روم» (incorrect)
• «خودکارنویس» (correct) vs «خودکار نویس» (different meaning)

This dataset can serve as a gold standard corpus for training and evaluating text normalization and correction models.

---

📁 Dataset Structure

The data is provided in a CSV file named Nim_Faseleh_Corpus.csv.

Column	Type	Description	Example
Word	string	Word or compound expression	آب‌انبار
Category	string	Lexical category (noun/adjective/etc.)	Noun
Starting_Letter	string	First character for grouping	آ
Length	int	Number of characters	8
Is_Compound	int	Indicates ZWNJ/compound structure (0/1)	1
Component_Count	int	Number of word components	2

---

📊 Statistics

• Total samples: 14,000+
• Full coverage of Persian alphabet (32 letters)
• Includes nouns, adjectives, verbs, adverbs, and idiomatic compounds
• Balanced structural diversity across compound formations

---

💡 Applications

1. Text Normalization Models
   Training systems that convert informal or incorrect Persian text into standardized form.

2. ZWNJ Evaluation Benchmark
   A benchmark dataset for evaluating the performance of Persian NLP models.

3. ASR / TTS Post-processing
   Improving transcription quality in speech systems by enforcing orthographic consistency.

---

📜 License

This dataset is released under the Creative Commons Attribution-ShareAlike 4.0 (CC BY-SA 4.0) license.

---

✍️ Citation

If you use this dataset in your research, please cite it as follows:

@misc{NimFaseleh,
  author = {Sina Eslami},
  title = {Nim Faseleh Corpus: Persian Zero-Width Non-Joiner Dataset},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/AmirScorpion/NimFaseleh}
}

🤝 Contribution

Feedback, issues, and contributions are highly welcome.
Please use the Issues section to report problems or suggest improvements.

⭐ If you find this dataset useful, consider giving it a star to support the project.