amitjslearn / wordipy

A dummy package which can be used to get abbrevations, tupled things, amounts written in letter to numbers.

wordipy

wordipy is dummy package which can be used in NLP related applications for example to get abbrevations, tupled things, amounts written in letter to numbers.

Dependencies

The following needs to be installed if you dont have it.

• Python 3 (3.7 or higher)
• NLTK
  pip3 install --user -U nltk
         or
  conda install -c anaconda nltk
• word2number
  pip3 install word2number
         or
  conda install -c conda-forge word2number

---

Clone this repository

git clone https://github.com/amitjslearn/wordipy.git

Open the terminal, cd into this repo directory (or command prompt) (not sure for windows)

python3 setup.py install

Usage

Open python in terminal

>>> from wordipy import abbreviationize
>>> abbreviationize("C M")
'CM'
>>> abbreviationize("Triple A")
'AAA'
>>> abbreviationize('Double Bam', sep=",")
'Bam, Bam'

for more help type help(abbreviationize)

>>> help(abbreviationize)

---

>>> from wordipy import amountize
>>> amountize("three hundred dollars")
'$300'
>>> amountize("two yen")
'¥2'

for more help type help(amountize)

>>> help(amountize)

---

Future Work

1. abbreviationize(): input to abbreviationize should be a capital like "C M" to get the output as CM for ex "c m" will not give "CM". We can take care of it.

2. amountize(): the amount which is already a number can be retained as it is

3. Stop words can be removed if needed

4. Lemmatization and stemming can be done properly wherever required

5. Dollars can also be usd, likewise for rupees can be inr (or even Rs) etc. So we can add synonyms for currency names

Note

1. Only most commonly used tuples (like single, double, triple, quadruple) are used, other tuples can be added as per the needs.
2. Only some of the currencies are added (like dollar, yen) other currencies can be added if required.
3. The amount/numbers given should be positive numbers upto the range of 999,999,999,999 (i.e. billions).