1
$\begingroup$

I have a pdf file that contains information . I would like to extract few key terms/phrase along with a value for example (current balance : CHF (swiss francs) 1,000)

I can convert pdf file to text using pdfminer . But how i can extract the above keyword using tensorflow text classification or other methods.

Can anyone suggest how i can start with it? I haven't come across a single example with tensorflow. There is a question here (Keyword/phrase extraction from Text using Deep Learning libraries ). but this doesn't help me to understand much as I am a beginner.

I used NLTK to tokenize words and read no of words by using

import re
text = re.sub(r'[_"\-;%()|+&=*.!?:#$@\[\]/]', ' ', f)
from nltk.tokenize import word_tokenize
from nltk.corpus import stopwords

tokens = word_tokenize(text)
stop_words = set(stopwords.words('english'))
words = [w for w in tokens if not w in stop_words]
freq = nltk.FreqDist(words) 
for key,val in freq.items():
    frq = str(key) + ':' + str(val)

So i have this text that i could like to extract the key phrase current balance : CHF (swiss francs) 1,000 along with the value using Neural Network. How can i do that? I came across many posts but couldnt get what i need.

PS: every time i upload a new file, i want my algorithm to find out the same keyword and use the value next to it.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.