sentifish 1.11.4

A simple package for sentiment analysis

sentifish is a Python library for Sentiment analysis of textual data(only English).By using sentifish it is very easy to perform tasks such as part-of-speech tagging, noun phrase extraction, sentiment analysis, classification etc. A good inforamtion for users is that this new version of sentifish has greater speed of sentiment analysis as compare to older versions.

Installation

sentifish-py can be installed using pip similar to other Python packages. Do not use sudo with pip.

To install sentifish-py, simply:

$ pip install sentifish

Getting Started

To ensure you have installed sentifish successfully you can run the following command in the Python IDLE.

>>> import sentifish

Sentifish have some methods classes which are following described.

sentTokenizer(paragraph)

sentTokenizer( ) is a method. It takes a paragraph as input and then returns a list of sentences of input paragraph.

>>> from sentifish import sentTokenizer
>>> para="This is the first sentence. This is the second sentence.
this is the third sentence."
>>> para_lines=sentTokenizer(para)
>>> para_lines
['This is the first sentence.', 'This is the second sentence.', 'this is the third sentence.']

wordTokenizer(sentence)

wordTokenizer( ) is a method. It takes a paragraph or sentence as input and then returns a list of words, symbols, and numbers of input paragraph or sentence.

>>> from sentifish import wordTokenizer
>>> sent="This is an example sentence."
>>> word_list=wordTokenizer(sent)
>>> word_list
['This', 'is', 'an', 'example', 'sentence', '.']

Class Sentiment( )

Sentiment( ) is a class. By using this class we can find the sentiment of a texual data(it may be a word, sentence or a paragraph). This class has a constructor init(self,text) which takes the text data at the time of instantiation of Sentiment( )

>>> from sentifish import Sentiment
>>> text="Ram is a good boy and he always remains happy"
>>> obj=Sentiment(text)

analyze( )

Class Sentiment( ) has a method analyze( ) it returns a float number in between -1 to +1. +1 for strongly positive sentiment, 0 for neutral and -1 for strongly negative sentiment.

>>> from sentifish import Sentiment
>>> text="Ram is a good boy and he always remains happy"
>>> obj=Sentiment(text)
>>> polarity = obj.analyze( )
>>> polarity
0.75

isPositive( )

Class Sentiment( ) has a method isPositive( ) it return True if the sentiment of the input text is positive. Otherwise it returns False.

>>> from sentifish import Sentiment
>>> text="Ram is a good boy and he always remains happy"
>>> obj=Sentiment(text)
>>> polarity = obj.analyze( )
>>> polarity
0.75
>>> obj.isPositive( )
True

isNegative( )

Class Sentiment( ) has a method isNegative( ) it return True if the sentiment of the input text is negative. Otherwise it returns False.

>>> from sentifish import Sentiment
>>> text="Ram is a good boy and he always remains happy"
>>> obj=Sentiment(text)
>>> polarity = obj.analyze( )
>>> polarity
0.75
>>> obj.isPositive( )
True
>>> obj.isNegative( )
False

isNeutral( )

Class Sentiment( ) has a method isNeutral( ) it return True if the sentiment of the input text is neutral. Otherwise it returns False.

>>> from sentifish import Sentiment
>>> text="Ram is a good boy and he always remains happy"
>>> obj=Sentiment(text)
>>> polarity = obj.analyze( )
>>> polarity
0.75
>>> obj.isPositive( )
True
>>> obj.isNegative( )
False
>>> obj.isNeutral( )
False

NOTE:- It can analys text of only english language.

Class PosTag( )

PosTag( ) is a class used for tagging word with part of speech tags. Class PosTag( ) has a constructor which requires list of words at the time of instantiation. The tagged words will store in a list “tagged_words” which can be access by using object of PosTag( ) class.

>>> from sentifish import wordTokenizer
>>> sent="This is an example sentence."
>>> word_list=wordTokenizer(sent)
>>> from sentifish import PosTag
>>> obj=PosTag(word_list)
>>> obj.tagged_words
[('This', 'This', ['NN']), ('is', 'is', ['HV']), ('an', 'an',['IA']),
('example', 'example', ['NN']), ('sentence', 'sentence', ['VB']),('.', '.', ['SYM'])]

Class Characters( )

Characters( ) is a class which has collection of special characters, small alphabets, capital alphabets and detailed information of “pos tags”. To find tags use tags( ) method.

>>> from sentifish import Characters
>>> obj = Characters( )
>>> obj.tags( )
{'HV': 'Helping verb', 'WP': 'Wh-Pronoun', 'CD': 'Cardinal number','PR': 'Pronoun',
'IN': 'Preposition','INV': 'Negative word','INC':'Word enhancing sense of another word',
'CC': 'Conjunction', 'SYM': 'Symbol','VB': 'Verb base form', 'VBD': 'Verb past form',
'VBN': 'Verb past participle form',
'VBZ': 'Verb s/es/ies/ form','VBG': 'Verb ing form', 'JJ': 'Adjective', 'RB': 'Adverb',
'Nn': 'Noun', 'V': 'Verb', 'NN': 'Noun', 'IA': 'Indefinite articles'}

To find special chars use specialChars( ) method.

>>> obj.specialChars( )
['`', '~', '@', '#', '$', '%', '^', '&', '*', '-', '_', ';', ':',
'\\', '|', '/', ',', '<', '.', '>', '?', "'", '"', '!', '+', ' ']

Use capitalAlpha( ) and smallAlpha( ) method to get list of capital alphabets and small alphabets respectively.

>>> obj.capitalAlpha( )
['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M',
'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z']
>>> obj.smallAlpha( )
['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm',
'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z']

Class FreqDist( )

FreqDist( ) is a class and by using this class we can find the number of occurrence of word, symbol, and number in a sentence. FreqDist( ) class has a constructor which takes a sentences containing words, symbols or numbers and makes a dictionary in which it takes words as key and number of occurrence of word as value.

>>> from sentifish import FreqDist
>>> para="This is the first sentence. This is the second sentence. this is the third sentence."
>>> obj = FreqDist(para)

to obtain dictionary of words or tokens use class variable “words_dict”.

>>> from sentifish import FreqDist
>>> para="This is the first sentence. This is the second sentence.
this is the third sentence."
>>> obj = FreqDist(para)
>>> obj.words_dict
{'This': 2, 'is': 3, 'the': 3, 'first': 1, 'sentence': 3, '.': 3, 'second': 1, 'this': 1, 'third': 1}

To obtain number of distinct words or tokens in the sentence or input text use “dict_size” class variable.

>>> from sentifish import FreqDist
>>> para="This is the first sentence. This is the second sentence. this is the third sentence."
>>> obj = FreqDist(para)
>>> obj.words_dict
{'This': 2, 'is': 3, 'the': 3, 'first': 1, 'sentence': 3, '.': 3, 'second': 1, 'this': 1, 'third': 1}
>>> obj.dict_size
9

most_common(num)

most_common( ) is method which takes an integer number as input and returns a list of tuple of words or tokens which have high frequency in the sentence with their frequency. Number of words in the list will equal to the input integer.

>>> from sentifish import FreqDist
>>> para="This is the first sentence. This is the second sentence.
this is the third sentence."
>>> obj = FreqDist(para)
>>> obj.words_dict
{'This': 2, 'is': 3, 'the': 3, 'first': 1, 'sentence': 3, '.': 3,'second': 1, 'this': 1, 'third': 1}
>>> obj.dict_size
9
>>> obj.most_common(2)
[('is', 3), ('the', 3)]

least_common(num)

least_common( ) is method which takes an integer number as input and returns a list of tuple of words or tokens which are least common in the sentence. Number of words in the list will equal to the input integer.

>>> from sentifish import FreqDist
>>> para="This is the first sentence. This is the second sentence. this is the third sentence."
>>> obj = FreqDist(para)
>>> obj.words_dict
{'This': 2, 'is': 3, 'the': 3, 'first': 1, 'sentence': 3, '.': 3,'second': 1, 'this': 1, 'third': 1}
>>> obj.dict_size
9
>>> obj.most_common(2)
[('is', 3), ('the', 3)]
>>> obj.least_common(3)
[('second', 1), ('this', 1), ('third', 1)]

Class Lemmatizer( )

Lemmatizer( ) is a class. By using lemmatizer class user can find the base form of verb from any other form of verb.

lemmatize(word)

Lemmatizer( ) class have a method of name lemmatize. It takes a word of other form and returns the base form of word.

>>> from sentifish import Lemmatizer
>>> obj = Lemmatizer( )
>>> obj.lemmatize("went")
‘go’

Class Polarity( )

Polarity( ) is a class. This class is very useful to fix the polarity of words.

fix_polarity(tagged_words_list)

it is an method of the polarity class. It takes tagged words list as input and then fix the sentiment polarity of words and then returns a list.

>>> from sentifish import wordTokenizer
>>> text="Ram is a good boy and he always remains happy"
>>> word_list=wordTokenizer(text)
>>> word_list
['Ram', 'is', 'a', 'good', 'boy', 'and', 'he', 'always', 'remains', 'happy']
>>> from sentifish import PosTag
>>> obj1 = PosTag(word_list)
>>> obj1.tagged_words
[('Ram', 'Ram', ['NN']), ('is', 'is', ['HV']), ('a', 'a', ['IA']), ('good', 'good', ['JJ']),
('boy', 'boy', ['Nn']), ('and', 'and', ['CC']), ('he', 'he', ['PR']), ('always', 'always', ['RB']),
('remains', 'remain', ['VBZ']),('happy', 'happy', ['JJ'])]
>>> from sentifish import Polarity
>>> obj.fix_polarity(obj1.tagged_words)
[('Ram', 'Ram', ['NN', 0.0]), ('is', 'is', ['HV', 0.0]), ('a', 'a', ['IA', 0.0]), ('good', 'good', ['JJ', 0.7]),
('boy', 'boy', ['Nn', 0.0]), ('and', 'and', ['CC', 0.0]), ('he', 'he', ['PR', 0.0]),
('always', 'always', ['RB', 0.0]), ('remains', 'remain', ['VBZ',0.0]), ('happy', 'happy', ['JJ', 0.8])]

remove_stopwords(text)

remove_stopwords(text) is a method and it takes text as input and return a list of words after removing stop words from the input text. Stop words are words which have not any sentiment polarity.

>>> from sentifish import remove_stopwords
>>> text="Ram is a good boy and he always remains happy"
>>> remove_stopwords(text)
['Ram', 'good', 'boy', 'remains', 'happy']

A list of stop word can be found from the Words() class.

>>> from sentifish import Words
>>> obj = Words()
>>> obj.stop_words()
['i', 'me', 'my', 'myself', 'we', 'our', 'ours',…………… 'aren']

remove_bitmap(words_list)

remove_bitmap( ) is a method and it takes text as input and return a list of words after removing words of other languages than english.

>>> from sentifish import remove_bitmap