File size: 2,804 Bytes
9db6ef7
afaeb05
9db6ef7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dccd18f
9db6ef7
a815a18
dccd18f
a815a18
9db6ef7
 
dccd18f
9db6ef7
 
 
 
 
 
 
 
 
 
 
dccd18f
9db6ef7
 
 
 
 
 
 
 
 
 
dccd18f
9db6ef7
 
 
 
 
 
 
 
dccd18f
9db6ef7
 
1497999
dccd18f
 
d5a5920
dccd18f
d5a5920
dccd18f
 
 
6367012
510dc6c
dccd18f
510dc6c
dccd18f
 
510dc6c
dccd18f
 
6367012
510dc6c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
---
license: apache-2.0
language:
- en
- ru
tags:
- translation
- dictionary
- nlp
- en-ru
- statistical-alignment
size_categories:
- 10K<n<100K
dataset_info:
  features:
    - name: english_word
      dtype: string
    - name: russian_word
      dtype: string
    - name: count
      dtype: int64
    - name: probability
      dtype: float64
  splits:
    - name: train
      num_examples: 56624
---

# en-ru-statistical-dict-20m-corpus

## Dataset Description

This dataset is a high-quality English-Russian lemmatized dictionary extracted from a parallel corpus of **[20 million sentences](https://huggingface.co/datasets/KvaytG/en-ru-parallel-20m)**. The dictionary was built using statistical alignment and rigorous filtering to ensure accuracy and relevance.

## Methodology

1. **Source:** Parallel corpus of 20,000,000 sentence pairs.
2. **Preprocessing:** - Tokenization and Lemmatization (SpaCy for English, PyMorphy3 for Russian).
   - Rare words in the corpus were replaced with `<unk>` tokens during training to optimize memory.
3. **Alignment:** Performed using `fast_align` (IBM Model 2) with forward and backward passes.
4. **Symmetrization:** `grow-diag-final-and` algorithm.
5. **Filtering:**
   - Numerical tokens and special characters were removed.
   - Minimum occurrence threshold: `count >= 50`.
   - Relative probability threshold: `probability >= 0.02` (2%).

## Dataset Structure

The dataset is provided in a single TSV (Tab-Separated Values) file:

| Column         | Description                                                  |
|:---------------|:-------------------------------------------------------------|
| `english_word` | Lemmatized English word                                      |
| `russian_word` | Lemmatized Russian translation                               |
| `count`        | Absolute number of times this pair was aligned in the corpus |
| `probability`  | Relative probability P(ru\|en) for this translation         |

## Example

For the word **"book"**:
- `книга`: 0.8156
- `бронирование`: 0.051
- `забронировать`: 0.0297

... and so on.

## Usage

```python
from datasets import load_dataset
dataset = load_dataset("KvaytG/en-ru-statistical-dict-20m-corpus", split="train")
```

## License

This dataset is released under the **Apache License 2.0**.

## Citation

```bibtex
@misc{kvaytg_en_ru_statistical_dict_20m_corpus,
  author       = {KvaytG},
  title        = {English-Russian Statistical Dictionary from 20M parallel corpus},
  year         = {2026},
  publisher    = {Hugging Face},
  journal      = {Hugging Face Datasets},
  url          = {https://huggingface.co/datasets/KvaytG/en-ru-statistical-dict-20m-corpus},
  note         = {Built from en-ru-parallel-20m using fast_align + grow-diag-final-and}
}
```