DarijaDZ

DarijaDZ is a large-scale corpus of user-generated text collected from Algerian YouTube and TikTok channels. The corpus contains approximately 22.5 million comment documents and 259.22 million word-level tokens, with content written primarily in Algerian Darija script alongside Latin/Arabizi writing and mixed-script content.

---

Dataset Description

Motivation

Algerian Darija is an under-resourced language variety with comparatively limited publicly available large-scale textual resources.

DarijaDZ was created to provide a large collection of naturally occurring Algerian online text that can be used for:

• language-model pretraining
• continued pretraining
• Arabic/Darija NLP research
• dialect identification
• sentiment analysis
• named entity recognition
• machine translation
• tokenizer research
• studying Arabic-script and Arabizi writing
• studying code-switching and informal online language

---

Corpus Statistics

Metric	Value
Documents	22,522,176
Word-level tokens	259,220,300
Mean tokens / document	11.51

---

Document-level distribution

Category	Percentage
Darija (Algerian dialect, Arabic script)	70.32%
MSA (Modern Standard Arabic)	9.22%
Arabize (Darija in Latin script / Arabizi)	9.63%
Code-switched (Arabic + Latin script mixed)	3.05%
French	6.11%
English	1.15%
Other (no linguistic content -- emoji, digits, symbols only)	0.52%

The imbalance between Arabic-script Darija and Arabizi (Darija in Latin script) motivated the development of an Arabic-script → Arabizi transliteration tool to help balance the data.

darija-arabizi-transliterator

Data Sources

The corpus is collected from two platforms: YouTube (via its official Data API v3) and TikTok (via a resumable scraper, no official/sanctioned API exists for this).

YouTube Collection

• Source: YouTube
• API: YouTube Data API v3
• Collected data: top-level comments and replies
• Scraper: resumable collection pipeline
• Channels: 49
• Raw video files processed: 105,196

TikTok Collection

• Source: TikTok
• Discovery: yt-dlp (channel video listing), no official API
• Collected data: top-level comments and replies
• Scraper: resumable collection pipeline
• Channels: 61
• Raw video files processed: 37,958

The source channels cover several content categories, including:

• News and current affairs
• Education / BAC preparation
• Cooking and lifestyle
• Entertainment and stories
• Commentary and vlogs

---

Cleaning and Preprocessing

The cleaning pipeline was empirically developed using a 1,000-document reservoir sample and subsequently applied to the corpus.

The corpus was processed using the following cleaning rules:

• URLs → replaced with [URL]
• @mentions → replaced with [MENTION]
• Zero-width and invisible characters → removed
• Directional-isolate control characters → removed
• Video timestamps → removed
• Consecutive emoji runs → collapsed
• Character elongation → collapsed while preserving expressive variation
• Excessive punctuation runs → collapsed
• Near-empty documents → removed
• Arabic diacritics (Tashkeel) → removed
• Unicode NFKC normalization → applied
• Near-duplicate documents → removed using MinHash/LSH

Intended Uses

Current release

The current dataset is most appropriate for:

• unsupervised language modeling
• language-model pretraining
• continued pretraining
• tokenizer development
• corpus analysis
• Arabic/Darija linguistic research
• studying informal online language
• studying Arabic/Latin script variation
• studying code-switching

Data Format

The dataset is distributed in a Hugging Face-compatible dataset format.

the dataset can be loaded using :

from datasets import load_dataset

dataset = load_dataset("nasrellahkharroubi/DarijaDz")

print(dataset)

---

Acknowledgments

DarijaDZ was created by Kharroubi Nasrellah , Nour Tliba and Rabah Arslene Djellouli, a fourth-year students at ENSIA

---

Citation

If you use DarijaDZ in your research, please cite the dataset using the following information:

Kharroubi Nasrellah , Nour Tliba , Rabah Arslene Djellouli .
DarijaDZ: Algerian Darija Corpus.
2026.

---