Page 299 - AI Ver 1.0 Class 10
P. 299
Installing NLTK
To install NLTK open command prompt in your computer and type:
pip install nltk
To install in anaconda environment, the command is:
conda install -c anaconda nltk
Downloading NLTK data
After the installation of NLTK, import nltk:
>>> import nltk
then next step is to install all the packages of nltk:
>>>nltk.download()
Some important commands of NLTK:
Tokenization is the process of converting large textual data into smaller parts called tokens. These tokens help in
nlp for finding patterns and are used for further processing through stemming and lemmatization.
Tokenizing a sentence will split big sentences into smaller sentences.
>>>data="Hello friends. Hope you are enjoying doing NLP. Wish you a wonderful
learning experience"
>>>from nltk.tokenize import sent_tokenize
>>>sent_token=sent_tokenize(data)
>>>print(sent_token)
Tokenizing a word will split a sentence into words.
>>> word_token=nltk.word_tokenize(data)
>>>print(word_token)
Stemming is the process of extracting base word from the given word:
>>>from nltk.stem import PorterStemmer
>>>ps = PorterStemmer()
>>>ps.stem('enjoying')
enjoy
>>>ps.stem('learning')
learn
Lemmatization is the process of extracting a base word called lemma. It is considered a better way than stemming
because stemming just removes the suffix without considering the actual meaning of the word.
Natural Language Processing 297

