cleantxty

Python package to clean strings and making them reasonable for NLP.

cleantxty is a an open-source python package cleaning text from raw text format. Source code for the library can be found here.

Features

cleantxt has two main methods,

• clean: to clean raw text and return the cleaned text
• clean_words: to clean raw text and return a list of clean words

other menthods that can be used simultaneoulsy are:

• remove_link: to remove link from the text
• remove_extra_white_space: to remove extra white space from the text
• lower_text: to make case of the text to lower case
• upper_text: to make case of the text to upper case
• remove_stopwords: to remove stopwords from the text
• remove_digits: to remove digits from the text
• remove_punctuations: to remove punctuations from the text
• custom_regex: to use custom regex and appy to text
• stem_text: to stem the provided text

Installation

cleantext requires Python 3 and NLTK to execute.

To install using pip, use

pip install cleantxty

Usage

• Import the library:

import cleantxty

• Choose a method:

To return the text in a string format,

cleantxty.clean("raw_text_here") 

To return a list of words from the text,

cleantxty.clean_words("raw_text_here") 

To choose a specific set of cleaning operations,

cleantxty.clean("raw_text_here",
default_case= "lower", # lower by default change to upper for upper case result
regex=None  # Provide custom regex to use
)

cleantxty.clean_words("raw_text_here",
default_case= "lower", # lower by default change to upper for upper case result
regex=None  # Provide custom regex to use
)

Examples

import cleantxty
cleantxty.clean('This is A s$ple ? tExt3% to   cleaN566556+wow8 ')

returns,

'this is a sample text to clean'

import cleantxty
cleantext.clean_words('This is A s$ample !!!! tExt3% to   cleaN566556+2+59*/133')

returns,

['sampl', 'text', 'clean']

from cleantxty import clean
text = "my id, name1@dom1.com and your, name2@dom2.in"
clean(text, regex=r"[a-z0-9\.\-+_]+@[a-z0-9\.\-+_]+\.[a-z]+")

returns,

"my id, email and your, email"

License

MIT

For any questions, issues, bugs, and suggestions please visit here