README.md

---
language:
- ar
- ru
- en
license: apache-2.0
tags:
- translation
- arabic
- russian
- english
- dictionary
- multilingual
- lexical-corpus
task_categories:
- translation
---

# Arabic-Russian-English Dictionary Collection

## Description
A comprehensive multilingual dictionary collection extracted from lexical files. Contains parallel translation pairs for Arabic-Russian, Russian-Arabic, and English-Russian.

## 📊 Dataset Statistics

### Overview
| Metric | Value |
|--------|-------|
| Total Records | 80,093 |
| Arabic-Russian Pairs | 17,479 |
| Russian-Arabic Pairs | 11,432 |
| English-Russian Pairs | 51,182 |

### Vocabulary Size
| Language | Unique Words |
|----------|--------------|
| Arabic | 9,643 |
| Russian | 11,431 |
| English | 51,182 |

### Average Word Length
| Language | Average Length |
|----------|----------------|
| Arabic | 4.1 characters |
| Russian | 11.1 characters |
| English | 9.6 characters |

### Translation Directions
| Direction                 | Count   | Percentage   |
|:--------------------------|:--------|:-------------|
| ar_ru (Arabic → Russian)  | 17,479  | 21.8%        |
| ru_ar (Russian → Arabic)  | 11,432  | 14.3%        |
| en_ru (English → Russian) | 51,182  | 63.9%        |

## 📁 Data Format

Each record contains:
- `direction`: translation direction (ar_ru, ru_ar, en_ru)
- `lemma`: source word/phrase
- `translation`: target word/phrase
- `transliteration`: pronunciation guide (where available)
- `notes`: grammatical annotations
- `sense_id`: for words with multiple meanings

### Example Records

**Arabic → Russian:**
```json
{
    "direction": "ar_ru",
    "lemma": "كتاب",
    "transliteration": "kitab",
    "translation": "книга",
    "notes": null
}
```

**Russian → Arabic:**
```json
{
    "direction": "ru_ar",
    "lemma": "книга",
    "transliteration": "ааа",
    "translation": "كتاب",
    "notes": null
}
```

**English → Russian:**
```json
{
    "direction": "en_ru",
    "lemma": "book",
    "transliteration": null,
    "translation": "книга",
    "notes": null
}
```

## 🚀 Usage

```python
from datasets import load_dataset

# Load Arabic-Russian dictionary
dataset = load_dataset("TajikNLPWorld/Arabic-Russian-Dictionary")
print(dataset["train"][0])

# Load Russian-Arabic dictionary
dataset = load_dataset("TajikNLPWorld/Russian-Arabic-Dictionary")
print(dataset["train"][0])

# Load English-Russian dictionary
dataset = load_dataset("TajikNLPWorld/English-Russian-Dictionary")
print(dataset["train"][0])

# Load combined dataset
dataset = load_dataset("TajikNLPWorld/Combined-Dictionary")
print(dataset["train"][0])
```

## 📜 License
Apache 2.0

## 🤝 Citation
```bibtex
@dataset{tajiknlpworld_dictionary_2024,
    title = {Arabic-Russian-English Dictionary Collection},
    author = {TajikNLPWorld},
    year = {2024},
    publisher = {Hugging Face},
    url = {https://huggingface.co/TajikNLPWorld}
}
```

---
*Dataset automatically generated from lexical files*