|
|
--- |
|
|
language: hy |
|
|
language_name: Armenian |
|
|
language_family: armenian |
|
|
tags: |
|
|
- wikilangs |
|
|
- nlp |
|
|
- tokenizer |
|
|
- embeddings |
|
|
- n-gram |
|
|
- markov |
|
|
- wikipedia |
|
|
- feature-extraction |
|
|
- sentence-similarity |
|
|
- tokenization |
|
|
- n-grams |
|
|
- markov-chain |
|
|
- text-mining |
|
|
- fasttext |
|
|
- babelvec |
|
|
- vocabulous |
|
|
- vocabulary |
|
|
- monolingual |
|
|
- family-armenian |
|
|
license: mit |
|
|
library_name: wikilangs |
|
|
pipeline_tag: text-generation |
|
|
datasets: |
|
|
- omarkamali/wikipedia-monthly |
|
|
dataset_info: |
|
|
name: wikipedia-monthly |
|
|
description: Monthly snapshots of Wikipedia articles across 300+ languages |
|
|
metrics: |
|
|
- name: best_compression_ratio |
|
|
type: compression |
|
|
value: 5.067 |
|
|
- name: best_isotropy |
|
|
type: isotropy |
|
|
value: 0.7690 |
|
|
- name: vocabulary_size |
|
|
type: vocab |
|
|
value: 0 |
|
|
generated: 2026-01-10 |
|
|
--- |
|
|
|
|
|
# Armenian - Wikilangs Models |
|
|
## Comprehensive Research Report & Full Ablation Study |
|
|
|
|
|
This repository contains NLP models trained and evaluated by Wikilangs, specifically on **Armenian** Wikipedia data. |
|
|
We analyze tokenizers, n-gram models, Markov chains, vocabulary statistics, and word embeddings. |
|
|
|
|
|
## ๐ Repository Contents |
|
|
|
|
|
### Models & Assets |
|
|
|
|
|
- Tokenizers (8k, 16k, 32k, 64k) |
|
|
- N-gram models (2, 3, 4, 5-gram) |
|
|
- Markov chains (context of 1, 2, 3, 4 and 5) |
|
|
- Subword N-gram and Markov chains |
|
|
- Embeddings in various sizes and dimensions (aligned and unaligned) |
|
|
- Language Vocabulary |
|
|
- Language Statistics |
|
|
|
|
|
 |
|
|
|
|
|
### Analysis and Evaluation |
|
|
|
|
|
- [1. Tokenizer Evaluation](#1-tokenizer-evaluation) |
|
|
- [2. N-gram Model Evaluation](#2-n-gram-model-evaluation) |
|
|
- [3. Markov Chain Evaluation](#3-markov-chain-evaluation) |
|
|
- [4. Vocabulary Analysis](#4-vocabulary-analysis) |
|
|
- [5. Word Embeddings Evaluation](#5-word-embeddings-evaluation) |
|
|
- [6. Morphological Analysis (Experimental)](#6--morphological-analysis-experimental) |
|
|
- [7. Summary & Recommendations](#7-summary--recommendations) |
|
|
- [Metrics Glossary](#appendix-metrics-glossary--interpretation-guide) |
|
|
- [Visualizations Index](#visualizations-index) |
|
|
|
|
|
--- |
|
|
## 1. Tokenizer Evaluation |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
### Results |
|
|
|
|
|
| Vocab Size | Compression | Avg Token Len | UNK Rate | Total Tokens | |
|
|
|------------|-------------|---------------|----------|--------------| |
|
|
| **8k** | 3.834x | 3.86 | 0.1591% | 2,659,955 | |
|
|
| **16k** | 4.305x | 4.33 | 0.1786% | 2,368,713 | |
|
|
| **32k** | 4.718x | 4.75 | 0.1958% | 2,161,366 | |
|
|
| **64k** | 5.067x ๐ | 5.10 | 0.2102% | 2,012,394 | |
|
|
|
|
|
### Tokenization Examples |
|
|
|
|
|
Below are sample sentences tokenized with each vocabulary size: |
|
|
|
|
|
**Sample 1:** `ิฑีฝีฟีฅึีธีซีคีถีฅึีซ ึีกีถีฏ ิฑีฝีฟีฅึีธีซีคีถีฅึีซ ึีกีถีฏ ิฑีฝีฟีฅึีธีซีคีถีฅึีซ ึีกีถีฏ | || || 12 ีฐีธีฏีฟีฅีดีขีฅึ || ิฟีซ...` |
|
|
|
|
|
| Vocab | Tokens | Count | |
|
|
|-------|--------|-------| |
|
|
| 8k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+12 more)` | 22 | |
|
|
| 16k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+9 more)` | 19 | |
|
|
| 32k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+9 more)` | 19 | |
|
|
| 64k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+9 more)` | 19 | |
|
|
|
|
|
**Sample 2:** `ิฑีฝีฟีฅึีธีซีคีถีฅึีซ ึีกีถีฏ ิฑีฝีฟีฅึีธีซีคีถีฅึีซ ึีกีถีฏ ิฑีฝีฟีฅึีธีซีคีถีฅึีซ ึีกีถีฏ | || || 13 ีกีบึีซีฌ || ิฟีกีฟีกีฌีซ...` |
|
|
|
|
|
| Vocab | Tokens | Count | |
|
|
|-------|--------|-------| |
|
|
| 8k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+10 more)` | 20 | |
|
|
| 16k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+10 more)` | 20 | |
|
|
| 32k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+8 more)` | 18 | |
|
|
| 64k | `โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โีกีฝีฟีฅึีธีซีคีถีฅึีซ โึีกีถีฏ โ| โ|| โ|| โ ... (+8 more)` | 18 | |
|
|
|
|
|
**Sample 3:** `ิดึีตีกีชีถีธ, ีขีถีกีฏีกีพีกีตึีฅึีซ ีกีถีธึีถีถีฅึี ิฒีฅีฌีกีผีธึีฝีซีก ิดึีตีกีชีถีธ - ีฃีตีธึีฒีกีฏ ีีซีฟีฅีขีฝีฏีซ ีทึีปีกีถีธึีด, ...` |
|
|
|
|
|
| Vocab | Tokens | Count | |
|
|
|-------|--------|-------| |
|
|
| 8k | `โีคึ ีต ีกีช ีถีธ , โีขีถีกีฏีกีพีกีตึีฅึีซ โีกีถีธึีถ ีถีฅึี โีขีฅีฌีกีผีธึีฝ ีซีก ... (+28 more)` | 38 | |
|
|
| 16k | `โีคึ ีต ีกีช ีถีธ , โีขีถีกีฏีกีพีกีตึีฅึีซ โีกีถีธึีถ ีถีฅึี โีขีฅีฌีกีผีธึีฝ ีซีก ... (+28 more)` | 38 | |
|
|
| 32k | `โีคึ ีตีกีช ีถีธ , โีขีถีกีฏีกีพีกีตึีฅึีซ โีกีถีธึีถ ีถีฅึี โีขีฅีฌีกีผีธึีฝ ีซีก โีคึ ... (+25 more)` | 35 | |
|
|
| 64k | `โีคึ ีตีกีช ีถีธ , โีขีถีกีฏีกีพีกีตึีฅึีซ โีกีถีธึีถ ีถีฅึี โีขีฅีฌีกีผีธึีฝ ีซีก โีคึ ... (+19 more)` | 29 | |
|
|
|
|
|
|
|
|
### Key Findings |
|
|
|
|
|
- **Best Compression:** 64k achieves 5.067x compression |
|
|
- **Lowest UNK Rate:** 8k with 0.1591% unknown tokens |
|
|
- **Trade-off:** Larger vocabularies improve compression but increase model size |
|
|
- **Recommendation:** 32k vocabulary provides optimal balance for production use |
|
|
|
|
|
--- |
|
|
## 2. N-gram Model Evaluation |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
### Results |
|
|
|
|
|
| N-gram | Variant | Perplexity | Entropy | Unique N-grams | Top-100 Coverage | Top-1000 Coverage | |
|
|
|--------|---------|------------|---------|----------------|------------------|-------------------| |
|
|
| **2-gram** | Word | 274,089 | 18.06 | 2,048,640 | 5.7% | 17.3% | |
|
|
| **2-gram** | Subword | 435 ๐ | 8.77 | 31,099 | 58.6% | 95.3% | |
|
|
| **3-gram** | Word | 587,646 | 19.16 | 3,025,054 | 4.6% | 14.3% | |
|
|
| **3-gram** | Subword | 3,630 | 11.83 | 285,926 | 26.0% | 63.1% | |
|
|
| **4-gram** | Word | 869,845 | 19.73 | 4,440,732 | 5.7% | 15.5% | |
|
|
| **4-gram** | Subword | 20,147 | 14.30 | 1,705,625 | 14.0% | 36.8% | |
|
|
| **5-gram** | Word | 457,726 | 18.80 | 2,884,886 | 8.6% | 20.6% | |
|
|
| **5-gram** | Subword | 78,248 | 16.26 | 5,627,732 | 8.9% | 23.9% | |
|
|
|
|
|
### Top 5 N-grams by Size |
|
|
|
|
|
**2-grams (Word):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ` | 162,524 | |
|
|
| 2 | `ีง ีธึ` | 122,895 | |
|
|
| 3 | `ีฎีกีถีธีฉีกีฃึีธึีฉีตีธึีถีถีฅึ ีกึีฟีกึีซีถ` | 120,383 | |
|
|
| 4 | `ีฅีฒีฅีฌ ีง` | 83,997 | |
|
|
| 5 | `ีง ีฉีพีกีฏีกีถีซ` | 83,509 | |
|
|
|
|
|
**3-grams (Word):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีฎีกีถีธีฉีกีฃึีธึีฉีตีธึีถีถีฅึ ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ` | 119,820 | |
|
|
| 2 | `ีทึีปีกีถีซ ีขีถีกีฏีกีพีกีตึีฅึ ีฃีตีธึีฒีฅึ` | 35,523 | |
|
|
| 3 | `ีด ีฉ ีก` | 25,916 | |
|
|
| 4 | `ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ ีผีธึีฝีกีฝีฟีกีถีซ` | 19,895 | |
|
|
| 5 | `ีง ีก ีด` | 19,670 | |
|
|
|
|
|
**4-grams (Word):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีฃีธีฟีซีถีฅึีจ worldtimezone com ีฏีกีตึีธึีด` | 18,028 | |
|
|
| 2 | `ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone com` | 18,028 | |
|
|
| 3 | `ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone` | 18,028 | |
|
|
| 4 | `ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ` | 18,006 | |
|
|
| 5 | `ีฐีฒีธึีดีถีฅึ ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ` | 18,006 | |
|
|
|
|
|
**5-grams (Word):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone com ีฏีกีตึีธึีด` | 18,028 | |
|
|
| 2 | `ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone com` | 18,028 | |
|
|
| 3 | `ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ` | 18,006 | |
|
|
| 4 | `ีฐีฒีธึีดีถีฅึ ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone` | 18,006 | |
|
|
| 5 | `ีฉีพีกีฏีกีถีซ ีธึีฏึีกีซีถีกีตีซ ีฐีกีดีกีบีฅีฟีกีฏีกีถ ีดีกึีคีกีฐีกีดีกึีซ ีกึีคีตีธึีถึีถีฅึีจ` | 17,821 | |
|
|
|
|
|
**2-grams (Subword):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีธ ึ` | 25,191,475 | |
|
|
| 2 | `ีก ีถ` | 21,219,860 | |
|
|
| 3 | `ีถ _` | 15,448,243 | |
|
|
| 4 | `ีฅ ึ` | 14,542,618 | |
|
|
| 5 | `ีซ _` | 12,944,110 | |
|
|
|
|
|
**3-grams (Subword):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีธ ึ ีด` | 7,197,254 | |
|
|
| 2 | `ีถ ีฅ ึ` | 6,928,522 | |
|
|
| 3 | `ีก ีถ _` | 6,546,164 | |
|
|
| 4 | `ีก ีฏ ีก` | 6,299,916 | |
|
|
| 5 | `ีฏ ีก ีถ` | 5,484,002 | |
|
|
|
|
|
**4-grams (Subword):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีก ีฏ ีก ีถ` | 4,881,415 | |
|
|
| 2 | `ีธ ึ ีฉ ีต` | 4,780,700 | |
|
|
| 3 | `ีธ ึ ีด _` | 4,302,874 | |
|
|
| 4 | `ีต ีธ ึ ีถ` | 3,153,212 | |
|
|
| 5 | `ีฏ ีก ีถ _` | 2,879,683 | |
|
|
|
|
|
**5-grams (Subword):** |
|
|
|
|
|
| Rank | N-gram | Count | |
|
|
|------|--------|-------| |
|
|
| 1 | `ีฉ ีต ีธ ึ ีถ` | 2,857,609 | |
|
|
| 2 | `ีธ ึ ีฉ ีต ีธ` | 2,854,879 | |
|
|
| 3 | `ึ ีฉ ีต ีธ ึ` | 2,854,454 | |
|
|
| 4 | `ีก ีฏ ีก ีถ _` | 2,763,929 | |
|
|
| 5 | `ีธ ึ ีฉ ีต ีก` | 1,924,954 | |
|
|
|
|
|
|
|
|
### Key Findings |
|
|
|
|
|
- **Best Perplexity:** 2-gram (subword) with 435 |
|
|
- **Entropy Trend:** Decreases with larger n-grams (more predictable) |
|
|
- **Coverage:** Top-1000 patterns cover ~24% of corpus |
|
|
- **Recommendation:** 4-gram or 5-gram for best predictive performance |
|
|
|
|
|
--- |
|
|
## 3. Markov Chain Evaluation |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
### Results |
|
|
|
|
|
| Context | Variant | Avg Entropy | Perplexity | Branching Factor | Unique Contexts | Predictability | |
|
|
|---------|---------|-------------|------------|------------------|-----------------|----------------| |
|
|
| **1** | Word | 0.8583 | 1.813 | 11.08 | 3,072,156 | 14.2% | |
|
|
| **1** | Subword | 1.3537 | 2.556 | 10.63 | 11,441 | 0.0% | |
|
|
| **2** | Word | 0.3131 | 1.242 | 2.02 | 34,021,781 | 68.7% | |
|
|
| **2** | Subword | 0.7180 | 1.645 | 5.38 | 121,571 | 28.2% | |
|
|
| **3** | Word | 0.1098 | 1.079 | 1.24 | 68,796,040 | 89.0% | |
|
|
| **3** | Subword | 0.7758 | 1.712 | 4.65 | 653,913 | 22.4% | |
|
|
| **4** | Word | 0.0398 ๐ | 1.028 | 1.07 | 84,964,993 | 96.0% | |
|
|
| **4** | Subword | 0.6964 | 1.620 | 3.55 | 3,038,021 | 30.4% | |
|
|
|
|
|
### Generated Text Samples (Word-based) |
|
|
|
|
|
Below are text samples generated from each word-based Markov chain model: |
|
|
|
|
|
**Context Size 1:** |
|
|
|
|
|
1. `ีง ีฐีซีฝีฟีฅึีซีฏีก ีฝีกึึีธึีด ึ ีซ ีกีผีกีปีซีถ ีบีกึีขีฅึีธึีฉีตีธึีถีจ 10 ีจ ีนีฅีถ ีกีถีพีกีถีฅีฌ ีฅีถ ีกีดีฅีถีกีฏีฅึีถีฅึ ีงีซีถ ีธึีบีฅีฝ ีขีธึีฝีกีขีกีถีกีฏีกีถ` |
|
|
2. `ึ ีถีซีฏีธีฌีกีต ีฝีดีธึีนีฏีธีพ ีกีพีกีฃ ีคีธึีฝีฟึ ีพีกึีฏีกีฎ ีจีฝีฟ ีผีค ีกีผีกีปีซีถ ีฐีกีฏีกึ
ีคีกีตีซีถ ีบีกีทีฟีบีกีถีธึีฉีตีกีถ ีฟีกีฏ ีง ึีซีฌีดีธึีด ีฉีพีกีฏีกีถีซีถ ...` |
|
|
3. `ีฅีถ ีทีกึีซีกีฉีซ ึ
ึีฅีถึีถีฅึีซีถ ีกีผีกีปีกึีฏีธึีฉีตีธึีถีถีฅึ ีกีถีฅีฌ ีซึ ีฐีกึีกีฆีกีฟ ีผีธีฝีฟีฝีฅีฌีดีกีท ีกีฏีธึีดีข ึีกีถีซ ีธึ ึ ีถีก ีฝีฟีฅีฒีฎีฅึ red w...` |
|
|
|
|
|
**Context Size 2:** |
|
|
|
|
|
1. `ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ ayuntamiento de torrejรณn de velasco la cocina espaรฑola antigua la cocina gitana las...` |
|
|
2. `ีง ีธึ ีกีตีค ีฉีฅีดีกีฟีซีฏีกีตีธีพ ีดีฅีฎ ีฐีกีตีฟีถีซีธึีฉีตีธึีถ ีฑีฅีผึ ีขีฅึีฅึ ีฌีตีธึีฏ ีขีฅีฝีธีถีซ ีฌีตีธึีฝีซ ึีซีฌีดีธึีด ีดีซีกีชีกีดีกีถีกีฏ ึีธึึีกีคึีพีธึีด...` |
|
|
3. `ีฎีกีถีธีฉีกีฃึีธึีฉีตีธึีถีถีฅึ ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ ะปะธัะฐ ัะฐะผะฐััะบะพะน ะณัะฑะตัะฝะธะธ ะฝะฐ ะณะพะด ัะฟะฑ ัะธะฟะพะณัะฐัะธั ะผ ะพ ะณ ัะธะบะป ะผะธะฝะธะฐัั...` |
|
|
|
|
|
**Context Size 3:** |
|
|
|
|
|
1. `ีฎีกีถีธีฉีกีฃึีธึีฉีตีธึีถีถีฅึ ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ ีฏีซีถีธีทีซีฟีก ีกีฝีฟีฅึีธีซีคีจ ึีธึึ ีดีธีฌีธึีกีฏีถีฅึีซ ีฏีฅีถีฟึีธีถีซ ีฏีกีตึีธึีด ีธึีฒีฅีฎึีซ ีฟีพีต...` |
|
|
2. `ีด ีฉ ีก ii i ีฐีกีฆีกึีกีดีตีกีฏีถีฅึ ีถีกีพีธีฌีธีฏ ีฃีตีธึีฒีซ ีทึีปีกีฏีกีตึีซ ีกีพีกีฆีกีขีฌีธึึีถีฅึีซึ ีฐีกีพีกึีพีกีฎ ีถีตีธึีฉีฅึีซ ีทีกึึีธึีด ีฅีฒีฅีฌ ีฅีถ ...` |
|
|
3. `ีกึีฟีกึีซีถ ีฐีฒีธึีดีถีฅึ ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone com ีฏีกีตึีธึีด ีขีกีทีฏีธึีฟีธีฝีฟีกีถ ีฐีกีถึีกีบีฅีฟีธึีฉีตีกีถ ีฆ...` |
|
|
|
|
|
**Context Size 4:** |
|
|
|
|
|
1. `ีผีธึีฝีกีฝีฟีกีถีซ ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone com ีฏีกีตึีธึีด ีฝีดีธีฌีฅีถีฝีฏีซ ีดีกึีฆีซ ีฝีดีธีฌีฅีถีฝีฏีซ ีทึีปีกีถีซ ีฏีกีฆีดีธึีด ีขีถีกีฏีน...` |
|
|
2. `ีฃีธีฟีซีถีฅึีจ worldtimezone com ีฏีกีตึีธึีด ีฌีฅีถีซีถีฃึีกีคีซ ีดีกึีฆีซ ีพีฝึีธีฌีธีชีฝีฏีซ ีทึีปีกีถีซ ีฏีกีฆีดีธึีด ีขีถีกีฏีนีธึีฉีตีธึีถีจ ีฉีพีกีฏีกีถีซีถ...` |
|
|
3. `ีชีกีดีกีตีซีถ ีฃีธีฟีซีถีฅึีจ worldtimezone com ีฏีกีตึีธึีด ีขีกีทีฏีธึีฟีธีฝีฟีกีถ ีฐีกีถึีกีบีฅีฟีธึีฉีตีกีถ ีธึึีซีดีฝีฏีซ ีทึีปีกีถีซ ีฏีกีฆีดีธึีด ีขีถีกีฏีน...` |
|
|
|
|
|
|
|
|
### Generated Text Samples (Subword-based) |
|
|
|
|
|
Below are text samples generated from each subword-based Markov chain model: |
|
|
|
|
|
**Context Size 1:** |
|
|
|
|
|
1. `_ีง_moronesestour` |
|
|
2. `ีกีฎีฅึีถีคีฅีถีธึีถีธีพีกีพีธ` |
|
|
3. `ีถึีกีตีฅึีฏีฝีฏีธีฌีธีถีนีฅีฝ` |
|
|
|
|
|
**Context Size 2:** |
|
|
|
|
|
1. `ีธึีฉีตีธึีด_ีดีฅึีซึ_ีฅีถีค` |
|
|
2. `ีกีถ_ีฐีฝีฏีพีฅึ_ีฃีธึีด,_ีธ` |
|
|
3. `ีถ_ีณีกึึีธึีฉีตีธึีถึ_ีฏีก` |
|
|
|
|
|
**Context Size 3:** |
|
|
|
|
|
1. `ีธึีด_ีดีฅีฎ_ีดีฅีฌีธีพ_ีกึีฐีซ` |
|
|
2. `ีถีฅึ_ีฏีกีด_ีคีฅ_ีดีฅีถีฟึีธึ` |
|
|
3. `ีกีถ_ีฟีพีตีกีฌีถีฅึีกีฆีดีซีถีนีบ` |
|
|
|
|
|
**Context Size 4:** |
|
|
|
|
|
1. `ีกีฏีกีถ_ีคีธีฌีซีพีธึีฉีตีธึีถีถีฅ` |
|
|
2. `ีธึีฉีตีกีถีจ_ีนีกึีซ_ีผีฅีฏีฟีธีถ` |
|
|
3. `ีธึีด_ีถึีกีถึ_ีฟีฒีกีดีกึีคีฏีก` |
|
|
|
|
|
|
|
|
### Key Findings |
|
|
|
|
|
- **Best Predictability:** Context-4 (word) with 96.0% predictability |
|
|
- **Branching Factor:** Decreases with context size (more deterministic) |
|
|
- **Memory Trade-off:** Larger contexts require more storage (3,038,021 contexts) |
|
|
- **Recommendation:** Context-3 or Context-4 for text generation |
|
|
|
|
|
--- |
|
|
## 4. Vocabulary Analysis |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
### Statistics |
|
|
|
|
|
| Metric | Value | |
|
|
|--------|-------| |
|
|
| Vocabulary Size | 1,233,415 | |
|
|
| Total Tokens | 101,630,095 | |
|
|
| Mean Frequency | 82.40 | |
|
|
| Median Frequency | 4 | |
|
|
| Frequency Std Dev | 4969.84 | |
|
|
|
|
|
### Most Common Words |
|
|
|
|
|
| Rank | Word | Frequency | |
|
|
|------|------|-----------| |
|
|
| 1 | ีง | 4,241,724 | |
|
|
| 2 | ึ | 2,378,989 | |
|
|
| 3 | ีฅีถ | 1,229,938 | |
|
|
| 4 | ีงึ | 646,182 | |
|
|
| 5 | ีฉีพีกีฏีกีถีซีถ | 577,819 | |
|
|
| 6 | ีฉีพีกีฏีกีถีซ | 566,535 | |
|
|
| 7 | ีธึ | 463,952 | |
|
|
| 8 | ีฐีกีดีกึ | 422,395 | |
|
|
| 9 | ีซ | 378,745 | |
|
|
| 10 | ีซึ | 371,982 | |
|
|
|
|
|
### Least Common Words (from vocabulary) |
|
|
|
|
|
| Rank | Word | Frequency | |
|
|
|------|------|-----------| |
|
|
| 1 | ีฝีกีถีฃีฏีธึีถีธึึ | 2 | |
|
|
| 2 | ีกึีนีฅีฐีธึีด | 2 | |
|
|
| 3 | ีผีฅีฃีฅีถีฝีซีธึีด | 2 | |
|
|
| 4 | ีคีฅีดีธีฃีธึีฃีกีถีซีถ | 2 | |
|
|
| 5 | ีทีถีกีบีถ | 2 | |
|
|
| 6 | ีผีกีฐีดีกีถีธึีฌีฌีก | 2 | |
|
|
| 7 | ีฌีกีฏีกีถีพีกีฌ | 2 | |
|
|
| 8 | ีฌีกีฏีกีถีพีกีฌีถ | 2 | |
|
|
| 9 | ีกีฏีฌีฅีฐีซ | 2 | |
|
|
| 10 | jsrn | 2 | |
|
|
|
|
|
### Zipf's Law Analysis |
|
|
|
|
|
| Metric | Value | |
|
|
|--------|-------| |
|
|
| Zipf Coefficient | 0.9590 | |
|
|
| Rยฒ (Goodness of Fit) | 0.995010 | |
|
|
| Adherence Quality | **excellent** | |
|
|
|
|
|
### Coverage Analysis |
|
|
|
|
|
| Top N Words | Coverage | |
|
|
|-------------|----------| |
|
|
| Top 100 | 25.0% | |
|
|
| Top 1,000 | 47.2% | |
|
|
| Top 5,000 | 65.8% | |
|
|
| Top 10,000 | 73.5% | |
|
|
|
|
|
### Key Findings |
|
|
|
|
|
- **Zipf Compliance:** Rยฒ=0.9950 indicates excellent adherence to Zipf's law |
|
|
- **High Frequency Dominance:** Top 100 words cover 25.0% of corpus |
|
|
- **Long Tail:** 1,223,415 words needed for remaining 26.5% coverage |
|
|
|
|
|
--- |
|
|
## 5. Word Embeddings Evaluation |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
|
|
|
### 5.1 Cross-Lingual Alignment |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
|
|
|
### 5.2 Model Comparison |
|
|
|
|
|
| Model | Dimension | Isotropy | Semantic Density | Alignment R@1 | Alignment R@10 | |
|
|
|-------|-----------|----------|------------------|---------------|----------------| |
|
|
| **mono_32d** | 32 | 0.7690 | 0.3405 | N/A | N/A | |
|
|
| **mono_64d** | 64 | 0.7404 | 0.3145 | N/A | N/A | |
|
|
| **mono_128d** | 128 | 0.6370 | 0.2680 | N/A | N/A | |
|
|
| **aligned_32d** | 32 | 0.7690 ๐ | 0.3747 | 0.1620 | 0.5260 | |
|
|
| **aligned_64d** | 64 | 0.7404 | 0.3024 | 0.2940 | 0.7100 | |
|
|
| **aligned_128d** | 128 | 0.6370 | 0.2604 | 0.4280 | 0.8220 | |
|
|
|
|
|
### Key Findings |
|
|
|
|
|
- **Best Isotropy:** aligned_32d with 0.7690 (more uniform distribution) |
|
|
- **Semantic Density:** Average pairwise similarity of 0.3101. Lower values indicate better semantic separation. |
|
|
- **Alignment Quality:** Aligned models achieve up to 42.8% R@1 in cross-lingual retrieval. |
|
|
- **Recommendation:** 128d aligned for best cross-lingual performance |
|
|
|
|
|
--- |
|
|
## 6. Morphological Analysis (Experimental) |
|
|
|
|
|
This section presents an automated morphological analysis derived from the statistical divergence between word-level and subword-level models. By analyzing where subword predictability spikes and where word-level coverage fails, we can infer linguistic structures without supervised data. |
|
|
|
|
|
### 6.1 Productivity & Complexity |
|
|
|
|
|
| Metric | Value | Interpretation | Recommendation | |
|
|
|--------|-------|----------------|----------------| |
|
|
| Productivity Index | **5.000** | High morphological productivity | Reliable analysis | |
|
|
| Idiomaticity Gap | **-0.140** | Low formulaic content | - | |
|
|
|
|
|
### 6.2 Affix Inventory (Productive Units) |
|
|
|
|
|
These are the most productive prefixes and suffixes identified by sampling the vocabulary for global substitutability patterns. A unit is considered an affix if stripping it leaves a valid stem that appears in other contexts. |
|
|
|
|
|
#### Productive Prefixes |
|
|
| Prefix | Examples | |
|
|
|--------|----------| |
|
|
| `-ีก` | ีกีขีกีปีซึ, ีกีดีถีฅีฆีซีก, ีกีตีฎีตีกีดีถีฅึ | |
|
|
| `-ีดีก` | ีดีกีคีฏีกีถึ, ีดีกีถีฃีกีฒีถีฅึ, ีดีกึีคีกีฐีกีทีพีซีถ | |
|
|
| `-ีฏีก` | ีฏีกีดีบีฅีกีคีธึ, ีฏีกีพีณีกีบีกีฟ, ีฏีกีขีกึีคีซีถีฏีก | |
|
|
| `-ีฏ` | ีฏีธีถีธึีกีตีซ, ีฏีกีดีบีฅีกีคีธึ, ีฏีฅีบีซีฟีกีฌ | |
|
|
| `-ีฏีธ` | ีฏีธีถีธึีกีตีซ, ีฏีธีถีฟึีกีบีธึีถีฏีฟีฅึีซ, ีฏีธีดีบีกีถีตีธีถีซ | |
|
|
| `-ีฐีก` | ีฐีกีดึ
ีฃีฟีกีฃีธึีฎีธึีดีจ, ีฐีกีตีงีถีฅึีฃีธีซ, ีฐีกึีฏีกีพีธ | |
|
|
| `-ีฝ` | ีฝีฅีญีดีฅึ, ีฝีฅีฏีฏีธ, ีฝีกีฒีกีฉีฅีฌีตีกีถีถ | |
|
|
| `-ีด` | ีดีกีคีฏีกีถึ, ีดีกีถีฃีกีฒีถีฅึ, ีดีฅึีฏีกึีธึึีซีน | |
|
|
|
|
|
#### Productive Suffixes |
|
|
| Suffix | Examples | |
|
|
|--------|----------| |
|
|
| `-ีถ` | ีพีฅึีกีขีถีกีฏีซีนีถีฅึีซีถ, ีญีธึึีกีตีซีถ, ีทึีปีกีถีกีฃีซีฎีถ | |
|
|
| `-ีซ` | ีฏีธีถีธึีกีตีซ, ีฏีธีถีฟึีกีบีธึีถีฏีฟีฅึีซ, ีพีซีพีซีซ | |
|
|
| `-ีจ` | ีฆีกีถีฃีกีฏีกีฟีถีฅึีจ, ีทีฅึีฆีซีถีฃีฅึีจ, ีฌีฅีตีทีดีกีถีซีธีฆีถีฅึีจ | |
|
|
| `-ีซีถ` | ีพีฅึีกีขีถีกีฏีซีนีถีฅึีซีถ, ีญีธึึีกีตีซีถ, ีขีฅีถีฆีซีถีซีถ | |
|
|
| `-ีกีถ` | ีนีฅีขีธีฟีกึึีฝีฏีกีตีกีถ, ีฏีกึีซีขีซีกีถ, ีถีฅีถีฃีกึีธีญีธึีฉีตีกีถ | |
|
|
| `-ึีซ` | ีฏีธีถีฟึีกีบีธึีถีฏีฟีฅึีซ, ีคีกีฟีกีพีธึีถีฅึีซ, ีกีถีฆีฃีกีตีกึีธึึีซีนีถีฅึีซ | |
|
|
| `-ึีด` | ีผีกีฆีดีกีขีฅีดีธึีด, ีฐีฅีพึีธีถีธึีด, ีบีกึีขีฅึีกีฃึีธึีด | |
|
|
| `-ีด` | ีผีกีฆีดีกีขีฅีดีธึีด, ีฐีฅีพึีธีถีธึีด, ีฝีซีฟีฏีธีด | |
|
|
|
|
|
### 6.3 Bound Stems (Lexical Roots) |
|
|
|
|
|
Bound stems are high-frequency subword units that are semantically cohesive but rarely appear as standalone words. These often correspond to the 'core' of a word that requires inflection or derivation to be valid. |
|
|
|
|
|
| Stem | Cohesion | Substitutability | Examples | |
|
|
|------|----------|------------------|----------| |
|
|
| `ีธึีฉีต` | 1.85x | 608 contexts | ีธึีฉีตีกีฏ, ีธึีฉีตีกีถ, ีธึีฉีตีธีก | |
|
|
| `ีถีถีฅึ` | 1.61x | 522 contexts | ีกีถีถีฅึ, ีฆีถีถีฅึ, ีธีถีถีฅึ | |
|
|
| `ึีถีฅึ` | 1.51x | 463 contexts | ีกึีถีฅึ, ีฉึีถีฅึ, ีฌีฅึีถีฅึ | |
|
|
| `ีดีขีฅึ` | 1.58x | 352 contexts | ีซีดีขีฅึ, ีฉีดีขีฅึ, ีญีดีขีฅึ | |
|
|
| `ีกีฟีธึ` | 1.40x | 704 contexts | ีขีกีฟีธึ, ีฏีกีฟีธึ, ีฐีกีฟีธึ | |
|
|
| `ึีพีกีฎ` | 1.64x | 255 contexts | ีปึีพีกีฎ, ีฐึีพีกีฎ, ีฝึีพีกีฎ | |
|
|
| `ีฏีถีฅึ` | 1.49x | 432 contexts | ีดีฏีถีฅึ, ีฏีถีฅึีฅ, ีฝีฏีถีฅึ | |
|
|
| `ีกีฆีดีก` | 1.60x | 200 contexts | ีฏีกีฆีดีก, ีกีฆีดีกีฉ, ีกีฆีดีกีถ | |
|
|
| `ึีดีถีฅ` | 1.69x | 134 contexts | ีธึีดีถีฅึีซ, ีฐีฒึีดีถีฅึ, ีฝีธึีดีถีฅึ | |
|
|
| `ึีฉีตีธ` | 1.72x | 115 contexts | ีธึีฉีตีธีก, ีธึีฉีตีธึีถ, ีฌีผีธึีฉีตีธ | |
|
|
| `ีทีญีกีฟ` | 1.82x | 71 contexts | ีกีทีญีกีฟ, ีกีทีญีกีฟีฅ, ีกีทีญีกีฟีซ | |
|
|
| `ีกีฒีกึ` | 1.73x | 82 contexts | ีดีกีฒีกึ, ีกีฒีกึีซ, ึีกีฒีกึ | |
|
|
|
|
|
### 6.4 Affix Compatibility (Co-occurrence) |
|
|
|
|
|
This table shows which prefixes and suffixes most frequently co-occur on the same stems, revealing the 'stacking' rules of the language's morphology. |
|
|
|
|
|
| Prefix | Suffix | Frequency | Examples | |
|
|
|--------|--------|-----------|----------| |
|
|
| `-ีก` | `-ีถ` | 126 words | ีกีถีฟีซึีธีถ, ีกีบีกีฏีฅีฃีธึีฎีธึีฉีตีกีถ | |
|
|
| `-ีก` | `-ีซ` | 101 words | ีกีฃีถีซีกีตีซ, ีกีพีฟีธีคึีธีดีซ | |
|
|
| `-ีฝ` | `-ีซ` | 84 words | ีฝีกีซีคีซ, ีฝีกีพีกีชีซ | |
|
|
| `-ีฝ` | `-ีถ` | 80 words | ีฝีธึีกีบีกีฐีธีพีดีกีถ, ีฝีกีถีคีธีพีซีถ | |
|
|
| `-ีฏ` | `-ีถ` | 78 words | ีฏีกีดีฅีถึีซีถ, ีฏีกีฆีดีกีฏีฅึีบีธึีฉีตีธึีถีถีฅึีซีถ | |
|
|
| `-ีฏ` | `-ีซ` | 74 words | ีฏึีซีธีฌีซีฉีธีฝึีฅึีกีตีซ, ีฏีฅีถีฝีกีดีซีปีธึีถีฅึีซ | |
|
|
| `-ีก` | `-ีจ` | 73 words | ีกึีซีฝีจ, ีกีฃีกึีคีจ | |
|
|
| `-ีด` | `-ีซ` | 72 words | ีดีกึีฝีซีดีกีฌีซีฆีดีซ, ีดีกีธึีฝีธีฌีธีฝีซ | |
|
|
| `-ีด` | `-ีถ` | 68 words | ีดีธึีฉีซึีถ, ีดีกีฐีธึีซีถ | |
|
|
| `-ีดีก` | `-ีถ` | 55 words | ีดีกีฐีธึีซีถ, ีดีกีธึีคีกีถ | |
|
|
|
|
|
### 6.5 Recursive Morpheme Segmentation |
|
|
|
|
|
Using **Recursive Hierarchical Substitutability**, we decompose complex words into their constituent morphemes. This approach handles nested affixes (e.g., `prefix-prefix-root-suffix`). |
|
|
|
|
|
| Word | Suggested Split | Confidence | Stem | |
|
|
|------|-----------------|------------|------| |
|
|
| ีฏึีกีฝีถีธีบีธีฌีซีฅ | **`ีฏึีกีฝีถีธีบีธีฌ-ีซ-ีฅ`** | 7.5 | `ีซ` | |
|
|
| ีคีซีคีฅึีดีถีฅึ | **`ีคีซีคีฅึีด-ีถ-ีฅึ`** | 7.5 | `ีถ` | |
|
|
| ีฑีถีกีฃีถีคีซีฏีถีฅึ | **`ีฑีถีกีฃีถีคีซีฏ-ีถ-ีฅึ`** | 7.5 | `ีถ` | |
|
|
| ีพีฅึีกีฐีฝีฏีพีกีฎ | **`ีพีฅึีกีฐีฝีฏ-ีพ-ีกีฎ`** | 7.5 | `ีพ` | |
|
|
| ีดีฅีฟีกีฏีธีฃีถีซีฟีซีพ | **`ีดีฅีฟีกีฏีธีฃีถีซีฟ-ีซ-ีพ`** | 7.5 | `ีซ` | |
|
|
| ีพีฅึีกีฐีกีถีฑีถีพีฅีฌ | **`ีพีฅึีกีฐีกีถีฑีถ-ีพ-ีฅีฌ`** | 7.5 | `ีพ` | |
|
|
| ีซึีกีฏีกีถีธึีฉีซึีถ | **`ีซึีกีฏีกีถีธึีฉ-ีซ-ึีถ`** | 7.5 | `ีซ` | |
|
|
| ีฏีธึีฌีซีขีกีฌีซีซ | **`ีฏีธึีฌีซีขีกีฌ-ีซ-ีซ`** | 7.5 | `ีซ` | |
|
|
| ีฐีกีดีกีฝีกึีกีถีซ | **`ีฐีกีดีกีฝีกึ-ีกีถ-ีซ`** | 7.5 | `ีกีถ` | |
|
|
| ีถีกีญีกีดีกึีคีซีฏ | **`ีถีกีญีกีดีกึีค-ีซ-ีฏ`** | 7.5 | `ีซ` | |
|
|
| ีดีกีตึีกีกีฝีบีกีถีจ | **`ีดีกีตึีกีกีฝีบ-ีกีถ-ีจ`** | 7.5 | `ีกีถ` | |
|
|
| ีดีฟีกีตีถีธึีฉีตีธึีถีซึ | **`ีดีฟีกีตีถีธึีฉีตีธึ-ีถ-ีซึ`** | 7.5 | `ีถ` | |
|
|
| ีดีซีฏึีธีขีซีธีฟีซีกีถ | **`ีดีซีฏึีธีขีซีธีฟ-ีซ-ีกีถ`** | 7.5 | `ีซ` | |
|
|
| ีฝีฏีฆีขีถีกีฏีกีถีซึ | **`ีฝีฏีฆีขีถีกีฏ-ีกีถ-ีซึ`** | 7.5 | `ีกีถ` | |
|
|
| ีฟึีซีบีฟีกีดีซีถีถีฅึ | **`ีฟึีซีบีฟีกีดีซีถ-ีถ-ีฅึ`** | 7.5 | `ีถ` | |
|
|
|
|
|
### 6.6 Linguistic Interpretation |
|
|
|
|
|
> **Automated Insight:** |
|
|
The language Armenian shows high morphological productivity. The subword models are significantly more efficient than word models, suggesting a rich system of affixation or compounding. |
|
|
|
|
|
--- |
|
|
## 7. Summary & Recommendations |
|
|
|
|
|
 |
|
|
|
|
|
### Production Recommendations |
|
|
|
|
|
| Component | Recommended | Rationale | |
|
|
|-----------|-------------|-----------| |
|
|
| Tokenizer | **64k BPE** | Best compression (5.07x) | |
|
|
| N-gram | **2-gram** | Lowest perplexity (435) | |
|
|
| Markov | **Context-4** | Highest predictability (96.0%) | |
|
|
| Embeddings | **100d** | Balanced semantic capture and isotropy | |
|
|
|
|
|
|
|
|
--- |
|
|
## Appendix: Metrics Glossary & Interpretation Guide |
|
|
|
|
|
This section provides definitions, intuitions, and guidance for interpreting the metrics used throughout this report. |
|
|
|
|
|
### Tokenizer Metrics |
|
|
|
|
|
**Compression Ratio** |
|
|
> *Definition:* The ratio of characters to tokens (chars/token). Measures how efficiently the tokenizer represents text. |
|
|
> |
|
|
> *Intuition:* Higher compression means fewer tokens needed to represent the same text, reducing sequence lengths for downstream models. A 3x compression means ~3 characters per token on average. |
|
|
> |
|
|
> *What to seek:* Higher is generally better for efficiency, but extremely high compression may indicate overly aggressive merging that loses morphological information. |
|
|
|
|
|
**Average Token Length (Fertility)** |
|
|
> *Definition:* Mean number of characters per token produced by the tokenizer. |
|
|
> |
|
|
> *Intuition:* Reflects the granularity of tokenization. Longer tokens capture more context but may struggle with rare words; shorter tokens are more flexible but increase sequence length. |
|
|
> |
|
|
> *What to seek:* Balance between 2-5 characters for most languages. Arabic/morphologically-rich languages may benefit from slightly longer tokens. |
|
|
|
|
|
**Unknown Token Rate (OOV Rate)** |
|
|
> *Definition:* Percentage of tokens that map to the unknown/UNK token, indicating words the tokenizer cannot represent. |
|
|
> |
|
|
> *Intuition:* Lower OOV means better vocabulary coverage. High OOV indicates the tokenizer encounters many unseen character sequences. |
|
|
> |
|
|
> *What to seek:* Below 1% is excellent; below 5% is acceptable. BPE tokenizers typically achieve very low OOV due to subword fallback. |
|
|
|
|
|
### N-gram Model Metrics |
|
|
|
|
|
**Perplexity** |
|
|
> *Definition:* Measures how "surprised" the model is by test data. Mathematically: 2^(cross-entropy). Lower values indicate better prediction. |
|
|
> |
|
|
> *Intuition:* If perplexity is 100, the model is as uncertain as if choosing uniformly among 100 options at each step. A perplexity of 10 means effectively choosing among 10 equally likely options. |
|
|
> |
|
|
> *What to seek:* Lower is better. Perplexity decreases with larger n-grams (more context). Values vary widely by language and corpus size. |
|
|
|
|
|
**Entropy** |
|
|
> *Definition:* Average information content (in bits) needed to encode the next token given the context. Related to perplexity: perplexity = 2^entropy. |
|
|
> |
|
|
> *Intuition:* High entropy means high uncertainty/randomness; low entropy means predictable patterns. Natural language typically has entropy between 1-4 bits per character. |
|
|
> |
|
|
> *What to seek:* Lower entropy indicates more predictable text patterns. Entropy should decrease as n-gram size increases. |
|
|
|
|
|
**Coverage (Top-K)** |
|
|
> *Definition:* Percentage of corpus occurrences explained by the top K most frequent n-grams. |
|
|
> |
|
|
> *Intuition:* High coverage with few patterns indicates repetitive/formulaic text; low coverage suggests diverse vocabulary usage. |
|
|
> |
|
|
> *What to seek:* Depends on use case. For language modeling, moderate coverage (40-60% with top-1000) is typical for natural text. |
|
|
|
|
|
### Markov Chain Metrics |
|
|
|
|
|
**Average Entropy** |
|
|
> *Definition:* Mean entropy across all contexts, measuring average uncertainty in next-word prediction. |
|
|
> |
|
|
> *Intuition:* Lower entropy means the model is more confident about what comes next. Context-1 has high entropy (many possible next words); Context-4 has low entropy (few likely continuations). |
|
|
> |
|
|
> *What to seek:* Decreasing entropy with larger context sizes. Very low entropy (<0.1) indicates highly deterministic transitions. |
|
|
|
|
|
**Branching Factor** |
|
|
> *Definition:* Average number of unique next tokens observed for each context. |
|
|
> |
|
|
> *Intuition:* High branching = many possible continuations (flexible but uncertain); low branching = few options (predictable but potentially repetitive). |
|
|
> |
|
|
> *What to seek:* Branching factor should decrease with context size. Values near 1.0 indicate nearly deterministic chains. |
|
|
|
|
|
**Predictability** |
|
|
> *Definition:* Derived metric: (1 - normalized_entropy) ร 100%. Indicates how deterministic the model's predictions are. |
|
|
> |
|
|
> *Intuition:* 100% predictability means the next word is always certain; 0% means completely random. Real text falls between these extremes. |
|
|
> |
|
|
> *What to seek:* Higher predictability for text generation quality, but too high (>98%) may produce repetitive output. |
|
|
|
|
|
### Vocabulary & Zipf's Law Metrics |
|
|
|
|
|
**Zipf's Coefficient** |
|
|
> *Definition:* The slope of the log-log plot of word frequency vs. rank. Zipf's law predicts this should be approximately -1. |
|
|
> |
|
|
> *Intuition:* A coefficient near -1 indicates the corpus follows natural language patterns where a few words are very common and most words are rare. |
|
|
> |
|
|
> *What to seek:* Values between -0.8 and -1.2 indicate healthy natural language distribution. Deviations may suggest domain-specific or artificial text. |
|
|
|
|
|
**Rยฒ (Coefficient of Determination)** |
|
|
> *Definition:* Measures how well the linear fit explains the frequency-rank relationship. Ranges from 0 to 1. |
|
|
> |
|
|
> *Intuition:* Rยฒ near 1.0 means the data closely follows Zipf's law; lower values indicate deviation from expected word frequency patterns. |
|
|
> |
|
|
> *What to seek:* Rยฒ > 0.95 is excellent; > 0.99 indicates near-perfect Zipf adherence typical of large natural corpora. |
|
|
|
|
|
**Vocabulary Coverage** |
|
|
> *Definition:* Cumulative percentage of corpus tokens accounted for by the top N words. |
|
|
> |
|
|
> *Intuition:* Shows how concentrated word usage is. If top-100 words cover 50% of text, the corpus relies heavily on common words. |
|
|
> |
|
|
> *What to seek:* Top-100 covering 30-50% is typical. Higher coverage indicates more repetitive text; lower suggests richer vocabulary. |
|
|
|
|
|
### Word Embedding Metrics |
|
|
|
|
|
**Isotropy** |
|
|
> *Definition:* Measures how uniformly distributed vectors are in the embedding space. Computed as the ratio of minimum to maximum singular values. |
|
|
> |
|
|
> *Intuition:* High isotropy (near 1.0) means vectors spread evenly in all directions; low isotropy means vectors cluster in certain directions, reducing expressiveness. |
|
|
> |
|
|
> *What to seek:* Higher isotropy generally indicates better-quality embeddings. Values > 0.1 are reasonable; > 0.3 is good. Lower-dimensional embeddings tend to have higher isotropy. |
|
|
|
|
|
**Average Norm** |
|
|
> *Definition:* Mean magnitude (L2 norm) of word vectors in the embedding space. |
|
|
> |
|
|
> *Intuition:* Indicates the typical "length" of vectors. Consistent norms suggest stable training; high variance may indicate some words are undertrained. |
|
|
> |
|
|
> *What to seek:* Relatively consistent norms across models. The absolute value matters less than consistency (low std deviation). |
|
|
|
|
|
**Cosine Similarity** |
|
|
> *Definition:* Measures angular similarity between vectors, ranging from -1 (opposite) to 1 (identical direction). |
|
|
> |
|
|
> *Intuition:* Words with similar meanings should have high cosine similarity. This is the standard metric for semantic relatedness in embeddings. |
|
|
> |
|
|
> *What to seek:* Semantically related words should score > 0.5; unrelated words should be near 0. Synonyms often score > 0.7. |
|
|
|
|
|
**t-SNE Visualization** |
|
|
> *Definition:* t-Distributed Stochastic Neighbor Embedding - a dimensionality reduction technique that preserves local structure for visualization. |
|
|
> |
|
|
> *Intuition:* Clusters in t-SNE plots indicate groups of semantically related words. Spread indicates vocabulary diversity; tight clusters suggest semantic coherence. |
|
|
> |
|
|
> *What to seek:* Meaningful clusters (e.g., numbers together, verbs together). Avoid over-interpreting distances - t-SNE preserves local, not global, structure. |
|
|
|
|
|
### General Interpretation Guidelines |
|
|
|
|
|
1. **Compare within model families:** Metrics are most meaningful when comparing models of the same type (e.g., 8k vs 64k tokenizer). |
|
|
2. **Consider trade-offs:** Better performance on one metric often comes at the cost of another (e.g., compression vs. OOV rate). |
|
|
3. **Context matters:** Optimal values depend on downstream tasks. Text generation may prioritize different metrics than classification. |
|
|
4. **Corpus influence:** All metrics are influenced by corpus characteristics. Wikipedia text differs from social media or literature. |
|
|
5. **Language-specific patterns:** Morphologically rich languages (like Arabic) may show different optimal ranges than analytic languages. |
|
|
|
|
|
|
|
|
### Visualizations Index |
|
|
|
|
|
| Visualization | Description | |
|
|
|---------------|-------------| |
|
|
| Tokenizer Compression | Compression ratios by vocabulary size | |
|
|
| Tokenizer Fertility | Average token length by vocabulary | |
|
|
| Tokenizer OOV | Unknown token rates | |
|
|
| Tokenizer Total Tokens | Total tokens by vocabulary | |
|
|
| N-gram Perplexity | Perplexity by n-gram size | |
|
|
| N-gram Entropy | Entropy by n-gram size | |
|
|
| N-gram Coverage | Top pattern coverage | |
|
|
| N-gram Unique | Unique n-gram counts | |
|
|
| Markov Entropy | Entropy by context size | |
|
|
| Markov Branching | Branching factor by context | |
|
|
| Markov Contexts | Unique context counts | |
|
|
| Zipf's Law | Frequency-rank distribution with fit | |
|
|
| Vocab Frequency | Word frequency distribution | |
|
|
| Top 20 Words | Most frequent words | |
|
|
| Vocab Coverage | Cumulative coverage curve | |
|
|
| Embedding Isotropy | Vector space uniformity | |
|
|
| Embedding Norms | Vector magnitude distribution | |
|
|
| Embedding Similarity | Word similarity heatmap | |
|
|
| Nearest Neighbors | Similar words for key terms | |
|
|
| t-SNE Words | 2D word embedding visualization | |
|
|
| t-SNE Sentences | 2D sentence embedding visualization | |
|
|
| Position Encoding | Encoding method comparison | |
|
|
| Model Sizes | Storage requirements | |
|
|
| Performance Dashboard | Comprehensive performance overview | |
|
|
|
|
|
--- |
|
|
## About This Project |
|
|
|
|
|
### Data Source |
|
|
|
|
|
Models trained on [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly) - a monthly snapshot of Wikipedia articles across 300+ languages. |
|
|
|
|
|
### Project |
|
|
|
|
|
A project by **[Wikilangs](https://wikilangs.org)** - Open-source NLP models for every Wikipedia language. |
|
|
|
|
|
### Maintainer |
|
|
|
|
|
[Omar Kamali](https://omarkamali.com) - [Omneity Labs](https://omneitylabs.com) |
|
|
|
|
|
### Citation |
|
|
|
|
|
If you use these models in your research, please cite: |
|
|
|
|
|
```bibtex |
|
|
@misc{wikilangs2025, |
|
|
author = {Kamali, Omar}, |
|
|
title = {Wikilangs: Open NLP Models for Wikipedia Languages}, |
|
|
year = {2025}, |
|
|
doi = {10.5281/zenodo.18073153}, |
|
|
publisher = {Zenodo}, |
|
|
url = {https://huggingface.co/wikilangs} |
|
|
institution = {Omneity Labs} |
|
|
} |
|
|
``` |
|
|
|
|
|
### License |
|
|
|
|
|
MIT License - Free for academic and commercial use. |
|
|
|
|
|
### Links |
|
|
|
|
|
- ๐ Website: [wikilangs.org](https://wikilangs.org) |
|
|
- ๐ค Models: [huggingface.co/wikilangs](https://huggingface.co/wikilangs) |
|
|
- ๐ Data: [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly) |
|
|
- ๐ค Author: [Omar Kamali](https://huggingface.co/omarkamali) |
|
|
- ๐ค Sponsor: [Featherless AI](https://featherless.ai) |
|
|
--- |
|
|
*Generated by Wikilangs Models Pipeline* |
|
|
|
|
|
*Report Date: 2026-01-10 18:05:40* |
|
|
|