Jul-27-2018, 06:11 AM
Hello,
I am trying to create a simple UI using tkinter that performs sentiment analysis on an input text. I am quite new to it and implementing NLP for the first time. Following is the mechanism that I want to follow-
1. Take text as input from the user in the textbox.
2. Create a button, when clicked on, performs the sentiment analysis on it.
3. Display the output, i.e. positive and negative polarities in another text box.
Given below is the program that I written-
from tkinter import *
import nltk.classify.util
from nltk.classify import NaiveBayesClassifier
from nltk.corpus import names
#create windows object
window=Tk()
window.geometry("1024x800") #set the window size using .geometry command
window.title("Sentiment Analysis") #this is the title of the GUI
def saui(words):
return dict([(word, True) for word in words])
positive_vocab = ['awesome', 'outstanding', 'fantastic', 'terrific', 'good',
'nice', 'great', ':)']
negative_vocab = ['bad', 'terrible', 'useless', 'hate', ':(']
neutral_vocab = ['movie', 'the', 'sound', 'was', 'is', 'actors', 'did',
'know', 'words', 'not']
pos_feat = [(saui(pos), 'pos') for pos in positive_vocab]
neg_feat = [(saui(neg), 'neg') for neg in negative_vocab]
neu_feat = [(saui(neu), 'neu') for neu in neutral_vocab]
train_set = neg_feat + pos_feat + neu_feat
classifier = NaiveBayesClassifier.train(train_set)
#set the heading of the GUI
heading=Label(window,text="This is how it works",font=
("arial",12,"bold"),fg="#26482B").pack()
#set the label, like 'enter text'
label1 = Label(window,text="Enter Some Text", font=
("arial",10,"bold"),fg="black")
label1.place(x=10,y=50)
#take some text as input in the textbox and put that input in a variable
ip = Text(window,width=60,height=30,bg="white")
ip.place(x=10,y=90)
sentence=ip.get('1.0',END)
sentence = sentence.lower()
words = sentence.split(' ')
#set another label that indicates the results
label2 = Label(window,text="Results", font=("arial",10,"bold"),fg="black")
label2.place(x=680,y=50)
#insert another textbox where you want to show the results
button1 = Button(window,text="Analyze",width=10,height=1,bg="#9B9D9C",fg="black",font=
("arial",8,"bold"),command=saui(words))
button1.place(x=200,y=586)
op = Text(window,width=20,height=15,bg="white")
op.place(x=680,y=90)
#Predict
neg = 0
pos = 0
for word in words:
classResult = classifier.classify(saui(word))
if classResult == 'neg':
neg = neg + 1
if classResult == 'pos':
pos = pos + 1
output1 = 'Positive: ' + str(float(pos) / len(words))
output2 = 'Negative: ' + str(float(neg) / len(words))
op.insert(0.0, output1)
op.insert(1.0,output2)
window.mainloop()
Earlier I encountered a number of errors like missing arguments, variable not defined, but I have resolved them. As of now, no such error is encountered.
However, when I click on the button, it is not displaying the output in the "op" textbox and is not even evaluating the polarities.
How I do resolve this issue?
Kindly help me with it.
I am trying to create a simple UI using tkinter that performs sentiment analysis on an input text. I am quite new to it and implementing NLP for the first time. Following is the mechanism that I want to follow-
1. Take text as input from the user in the textbox.
2. Create a button, when clicked on, performs the sentiment analysis on it.
3. Display the output, i.e. positive and negative polarities in another text box.
Given below is the program that I written-
from tkinter import *
import nltk.classify.util
from nltk.classify import NaiveBayesClassifier
from nltk.corpus import names
#create windows object
window=Tk()
window.geometry("1024x800") #set the window size using .geometry command
window.title("Sentiment Analysis") #this is the title of the GUI
def saui(words):
return dict([(word, True) for word in words])
positive_vocab = ['awesome', 'outstanding', 'fantastic', 'terrific', 'good',
'nice', 'great', ':)']
negative_vocab = ['bad', 'terrible', 'useless', 'hate', ':(']
neutral_vocab = ['movie', 'the', 'sound', 'was', 'is', 'actors', 'did',
'know', 'words', 'not']
pos_feat = [(saui(pos), 'pos') for pos in positive_vocab]
neg_feat = [(saui(neg), 'neg') for neg in negative_vocab]
neu_feat = [(saui(neu), 'neu') for neu in neutral_vocab]
train_set = neg_feat + pos_feat + neu_feat
classifier = NaiveBayesClassifier.train(train_set)
#set the heading of the GUI
heading=Label(window,text="This is how it works",font=
("arial",12,"bold"),fg="#26482B").pack()
#set the label, like 'enter text'
label1 = Label(window,text="Enter Some Text", font=
("arial",10,"bold"),fg="black")
label1.place(x=10,y=50)
#take some text as input in the textbox and put that input in a variable
ip = Text(window,width=60,height=30,bg="white")
ip.place(x=10,y=90)
sentence=ip.get('1.0',END)
sentence = sentence.lower()
words = sentence.split(' ')
#set another label that indicates the results
label2 = Label(window,text="Results", font=("arial",10,"bold"),fg="black")
label2.place(x=680,y=50)
#insert another textbox where you want to show the results
button1 = Button(window,text="Analyze",width=10,height=1,bg="#9B9D9C",fg="black",font=
("arial",8,"bold"),command=saui(words))
button1.place(x=200,y=586)
op = Text(window,width=20,height=15,bg="white")
op.place(x=680,y=90)
#Predict
neg = 0
pos = 0
for word in words:
classResult = classifier.classify(saui(word))
if classResult == 'neg':
neg = neg + 1
if classResult == 'pos':
pos = pos + 1
output1 = 'Positive: ' + str(float(pos) / len(words))
output2 = 'Negative: ' + str(float(neg) / len(words))
op.insert(0.0, output1)
op.insert(1.0,output2)
window.mainloop()
Earlier I encountered a number of errors like missing arguments, variable not defined, but I have resolved them. As of now, no such error is encountered.
However, when I click on the button, it is not displaying the output in the "op" textbox and is not even evaluating the polarities.
How I do resolve this issue?
Kindly help me with it.