Library for statistics extraction from texts in Russian.
Run the following command:
$ pip install rutsDependencies:
- python 3.8-3.10
- nltk
- pymorphy2
- razdel
- scipy
- spaCy
- numpy
- pandas
- matplotlib
- graphviz
The main functions are based on the textacy statistics adapted to Russian language. The library allows working both with raw texts and Doc-objects of the spaCy library.
API to explore the available functions.
The library allows creating your own tools for sentence and word extraction from a text, which can be further employed for counting statistics.
Example:
import re
from nltk.corpus import stopwords
from ruts import SentsExtractor, WordsExtractor
text = "Не имей 100 рублей, а имей 100 друзей"
se = SentsExtractor(tokenizer=re.compile(r', '))
se.extract(text)
('Не имей 100 рублей', 'а имей 100 друзей')
we = WordsExtractor(use_lexemes=True, stopwords=stopwords.words('russian'), filter_nums=True, ngram_range=(1, 2))
we.extract(text)
('иметь', 'рубль', 'иметь', 'друг', 'иметь_рубль', 'рубль_иметь', 'иметь_друг')
we.get_most_common(3)
[('иметь', 2), ('рубль', 1), ('друг', 1)]The library allows extracting the following statistics from a text:
- the number of sentences
- the number of words
- the number of unique words
- the number of long words
- the number of complex words
- the number of simple words
- the number of monosyllabic words
- the number of polysyllabic words
- the number of symbols
- the number of letters
- the number of spaces
- the number of syllables
- the number of punctuation marks
- word distribution by the number of letters
- word distribution by the number of syllables
Example:
from ruts import BasicStats
text = "Существуют три вида лжи: ложь, наглая ложь и статистика"
bs = BasicStats(text)
bs.get_stats()
{'c_letters': {1: 1, 3: 2, 4: 3, 6: 1, 10: 2},
'c_syllables': {1: 5, 2: 1, 3: 1, 4: 2},
'n_chars': 55,
'n_complex_words': 2,
'n_letters': 45,
'n_long_words': 3,
'n_monosyllable_words': 5,
'n_polysyllable_words': 4,
'n_punctuations': 2,
'n_sents': 1,
'n_simple_words': 7,
'n_spaces': 8,
'n_syllables': 18,
'n_unique_words': 8,
'n_words': 9}
bs.print_stats()
Статистика | Значение
------------------------------
Предложения | 1
Слова | 9
Уникальные слова | 8
Длинные слова | 3
Сложные слова | 2
Простые слова | 7
Односложные слова | 5
Многосложные слова | 4
Символы | 55
Буквы | 45
Пробелы | 8
Слоги | 18
Знаки препинания | 2The library allows counting the following readability metrics:
- Flesch Reading Ease
- Flesch-Kincaid Grade Level
- Coleman-Liau Index
- SMOG Index
- Automated Readability Index
- LIX readability measure
Coefficients for Russian language were borrowed from the Plain Russian Language project dedicated to counting readability coefficients based on a special corpus of texts with age labels.
Example:
from ruts import ReadabilityStats
text = "Ног нет, а хожу, рта нет, а скажу: когда спать, когда вставать, когда работу начинать"
rs = ReadabilityStats(text)
rs.get_stats()
{'automated_readability_index': 0.2941666666666656,
'coleman_liau_index': 0.2941666666666656,
'flesch_kincaid_grade': 3.4133333333333304,
'flesch_reading_easy': 83.16166666666666,
'lix': 48.333333333333336,
'smog_index': 0.05}
rs.print_stats()
Метрика | Значение
--------------------------------------------------
Тест Флеша-Кинкайда | 3.41
Индекс удобочитаемости Флеша | 83.16
Индекс Колман-Лиау | 0.29
Индекс SMOG | 0.05
Автоматический индекс удобочитаемости | 0.29
Индекс удобочитаемости LIX | 48.33 The library allows counting the following lexical diversity metrics for a text:
- Type-Token Ratio (TTR)
- Root Type-Token Ratio (RTTR)
- Corrected Type-Token Ratio (CTTR)
- Herdan Type-Token Ratio (HTTR)
- Summer Type-Token Ratio (STTR)
- Mass Type-Token Ratio (MTTR)
- Dugast Type-Token Ratio (DTTR)
- Moving Average Type-Token Ratio (MATTR)
- Mean Segmental Type-Token Ratio (MSTTR)
- Measure of Textual Lexical Diversity (MTLD)
- Moving Average Measure of Textual Lexical Diversity (MAMTLD)
- Hypergeometric Distribution D (HD-D)
- Simpson's Diversity Index
- Hapax Legomena Index
Some of the implementations were borrowed from the lexical_diversity project.
Example:
from ruts import DiversityStats
text = "Ног нет, а хожу, рта нет, а скажу: когда спать, когда вставать, когда работу начинать"
ds = DiversityStats(text)
ds.get_stats()
{'ttr': 0.7333333333333333,
'rttr': 2.840187787218772,
'cttr': 2.008316044185609,
'httr': 0.8854692840710253,
'sttr': 0.2500605793160845,
'mttr': 0.0973825075623254,
'dttr': 10.268784661968104,
'mattr': 0.7333333333333333,
'msttr': 0.7333333333333333,
'mtld': 15.0,
'mamtld': 11.875,
'hdd': -1,
'simpson_index': 21.0,
'hapax_index': 431.2334616537499}
ds.print_stats()
Метрика | Значение
----------------------------------------------------------------------
Type-Token Ratio (TTR) | 0.92
Root Type-Token Ratio (RTTR) | 7.17
Corrected Type-Token Ratio (CTTR) | 5.07
Herdan Type-Token Ratio (HTTR) | 0.98
Summer Type-Token Ratio (STTR) | 0.96
Mass Type-Token Ratio (MTTR) | 0.01
Dugast Type-Token Ratio (DTTR) | 85.82
Moving Average Type-Token Ratio (MATTR) | 0.91
Mean Segmental Type-Token Ratio (MSTTR) | 0.94
Measure of Textual Lexical Diversity (MTLD) | 208.38
Moving Average Measure of Textual Lexical Diversity (MTLD) | 1.00
Hypergeometric Distribution D (HD-D) | 0.94
Индекс Симпсона | 305.00
Гапакс-индекс | 2499.46 The library allows extracting the following morphological features:
- part of speech
- animacy
- aspect
- case
- gender
- involvement
- mood
- number
- person
- tense
- transitivity
- voice
Morphological analysis is made using pymorphy2. Descriptions of morphological features were borrowed from OpenCorpora.
Example:
from ruts import MorphStats
text = "Постарайтесь получить то, что любите, иначе придется полюбить то, что получили"
ms = MorphStats(text)
ms.pos
('VERB', 'INFN', 'CONJ', 'CONJ', 'VERB', 'ADVB', 'VERB', 'INFN', 'CONJ', 'CONJ', 'VERB')
ms.get_stats()
{'animacy': {None: 11},
'aspect': {None: 5, 'impf': 1, 'perf': 5},
'case': {None: 11},
'gender': {None: 11},
'involvement': {None: 10, 'excl': 1},
'mood': {None: 7, 'impr': 1, 'indc': 3},
'number': {None: 7, 'plur': 3, 'sing': 1},
'person': {None: 9, '2per': 1, '3per': 1},
'pos': {'ADVB': 1, 'CONJ': 4, 'INFN': 2, 'VERB': 4},
'tense': {None: 8, 'futr': 1, 'past': 1, 'pres': 1},
'transitivity': {None: 5, 'intr': 2, 'tran': 4},
'voice': {None: 11}}
ms.explain_text(filter_none=True)
(('Постарайтесь',
{'aspect': 'perf',
'involvement': 'excl',
'mood': 'impr',
'number': 'plur',
'pos': 'VERB',
'transitivity': 'intr'}),
('получить', {'aspect': 'perf', 'pos': 'INFN', 'transitivity': 'tran'}),
('то', {'pos': 'CONJ'}),
('что', {'pos': 'CONJ'}),
('любите',
{'aspect': 'impf',
'mood': 'indc',
'number': 'plur',
'person': '2per',
'pos': 'VERB',
'tense': 'pres',
'transitivity': 'tran'}),
('иначе', {'pos': 'ADVB'}),
('придется',
{'aspect': 'perf',
'mood': 'indc',
'number': 'sing',
'person': '3per',
'pos': 'VERB',
'tense': 'futr',
'transitivity': 'intr'}),
('полюбить', {'aspect': 'perf', 'pos': 'INFN', 'transitivity': 'tran'}),
('то', {'pos': 'CONJ'}),
('что', {'pos': 'CONJ'}),
('получили',
{'aspect': 'perf',
'mood': 'indc',
'number': 'plur',
'pos': 'VERB',
'tense': 'past',
'transitivity': 'tran'}))
ms.print_stats('pos', 'tense')
---------------Часть речи---------------
Глагол (личная форма) | 4
Союз | 4
Глагол (инфинитив) | 2
Наречие | 1
-----------------Время------------------
Неизвестно | 8
Настоящее | 1
Будущее | 1
Прошедшее | 1 Library allows working with a number of preprocessed datasets:
- sov_chrest_lit - soviet reading-books for literature classes
- stalin_works - the collected works of Stalin
One can work solely with texts (without title info) or texts with metadata. There is also an opportunity to filter texts on different criteria.
Example:
from ruts.datasets import SovChLit
sc = SovChLit()
sc.info
{'description': 'Корпус советских хрестоматий по литературе',
'url': 'https://dataverse.harvard.edu/file.xhtml?fileId=3670902&version=DRAFT',
'Наименование': 'sov_chrest_lit'}
for i in sc.get_records(max_len=100, category='Весна', limit=1):
pprint(i)
{'author': 'Е. Трутнева',
'book': 'Родная речь. Книга для чтения в I классе начальной школы',
'category': 'Весна',
'file': PosixPath('../ruTS/ruts_data/texts/sov_chrest_lit/grade_1/155'),
'grade': 1,
'subject': 'Дождик',
'text': 'Дождик, дождик, поливай, будет хлеба каравай!\n'
'Дождик, дождик, припусти, дай гороху подрасти!',
'type': 'Стихотворение',
'year': 1963}
for i in sc.get_texts(text_type='Басня', limit=1):
pprint(i)
('— Соседка, слышала ль ты добрую молву? — вбежавши, крысе мышь сказала:\n'
'— Ведь кошка, говорят, попалась в когти льву. Вот отдохнуть и нам пора '
'настала!\n'
'— Не радуйся, мой свет,— ей крыса говорит в ответ,— и не надейся '
'по-пустому.\n'
'Коль до когтей у них дойдёт, то, верно, льву не быть живому: сильнее кошки '
'зверя нет.')Library allows visualizing text with the help of the following graphs:
- Zipf's law
- Literature Fingerprinting
- Word Tree
Example:
from collections import Counter
from nltk.corpus import stopwords
from ruts import WordsExtractor
from ruts.datasets import SovChLit
from ruts.visualizers import zipf
sc = SovChLit()
text = '\n'.join([text for text in sc.get_texts(limit=100)])
we = WordsExtractor(use_lexemes=True, stopwords=stopwords.words('russian'), filter_nums=True)
tokens_with_count = Counter(we.extract(text))
zipf(tokens_with_count, num_words=100, num_labels=10, log=False, show_theory=True, alpha=1.1)Library allows creating the following classes of spaCy components:
- BasicStats
- DiversityStats
- MorphStats
- ReadabilityStats
Russian-language spaCy model can be downloaded by running the command:
$ python -m spacy download ru_core_news_smExample:
import ruts
import spacy
nlp = spacy.load('ru_core_news_sm')
nlp.add_pipe('basic', last=True)
doc = nlp("Существуют три вида лжи: ложь, наглая ложь и статистика")
doc._.basic.c_letters
{1: 1, 3: 2, 4: 3, 6: 1, 10: 2}
doc._.basic.get_stats()
{'c_letters': {1: 1, 3: 2, 4: 3, 6: 1, 10: 2},
'c_syllables': {1: 5, 2: 1, 3: 1, 4: 2},
'n_chars': 55,
'n_complex_words': 2,
'n_letters': 45,
'n_long_words': 3,
'n_monosyllable_words': 5,
'n_polysyllable_words': 4,
'n_punctuations': 2,
'n_sents': 1,
'n_simple_words': 7,
'n_spaces': 8,
'n_syllables': 18,
'n_unique_words': 8,
'n_words': 9}- docs - project documentation
- ruts:
- basic_stats.py - basic text statistics
- components.py - spaCy components
- constants.py - main constants
- diversity_stats.py - lexical diversity metrics
- extractors.py - tools for object extraction from a text
- morph_stats.py - morphological statistics
- readability_stats.py - readability metrics
- utils.py - subsidiary tools
- datasets:
- dataset.py - basic class for working with datasets
- sov_chrest_lit.py - soviet reading-books for literature classes
- stalin_works.py - the collected works of Stalin
- visualizers - tools for text visualization:
- fingerprinting.py - Literature Fingerprinting
- word_tree.py - Word Tree
- zipf.py - Zipf's law
- tests:
- test_basic_stats.py - tests for basic text statistics
- test_components.py - tests for spaCy components
- test_diversity_stats.py - tests for lexical diversity metrics
- test_extractors.py - tests for object extraction tools
- test_morph_stats - tests for morphological statistics
- test_readability_stats.py - tests for readability metrics
- datasets - tests for datasets:
- test_dataset.py - tests for basic class for working with datasets
- test_sov_chrest_lit.py - tests for dataset soviet reading-books for literature classes
- test_stalin_works.py - tests for dataset the collected works of Stalin
- visualizers - tests for tools for text visualization:
- test_fingerprinting.py - tests for visualization Literature Fingerprinting
- test_word_tree.py - tests for visualization Word Tree
- test_zipf.py - tests for visualization Zipf's law
- Sergey Shkarin ([email protected])
- Ekaterina Smirnova ([email protected])
Please use the following BibTeX entry for citing ruTS if you use it in your research or software. Citations are helpful for the continued development and maintenance of this library.
@software{ruTS,
author = {Sergey Shkarin},
title = {{ruTS, a library for statistics extraction from texts in Russian}},
year = 2023,
publisher = {Moscow},
url = {https://github.com/SergeyShk/ruTS}
}