Sentiment analysis on csv file Python 3.8

Question

I have a csv file where I wish to perform a sentiment analysis on this dataset containing survey data.

So far this is what I have tried (thanks to Rupin from a previous question!):

import csv
from collections import Counter

with open('myfile.csv', 'r') as f:
    reader = csv.reader(f, delimiter='\t')
    alist = []
    iterreader = iter(reader)
    next(iterreader, None)
    for row in iterreader:
        clean_rows = row[0].replace(",", " ").rsplit()

        alist.append(clean_rows)
        word_count = Counter(clean_rows)
        mostWcommon = word_count.most_common(3)
        print(mostWcommon)

The output is nearly okay, the only problem that I have is that Python is splitting in different rows of a list, hence I have something like this as my output:

1. ['experienced', 1]
2. ['experienced, 1]
3. ['experienced, 1]

I wish to split everything in one row so that I can have the real word frequency... Any suggestions?

Thanks!

Answer 1

You are creating a new Counter for each row and printing only that result. If you want a total count, you can create the counter outside the rows loop and update it with data from each row:

import csv
from collections import Counter

with open('myfile.csv', 'r') as f:
    reader = csv.reader(f, delimiter='\t')
    alist = []
    iterreader = iter(reader)
    next(iterreader, None)
    c = Conter()
    for row in iterreader:
        clean_rows = row[0].replace(",", " ").rsplit()
        alist.append(clean_rows)
        c.update(clean_rows)
    mostWcommon = word_count.most_common(3)
    print(mostWcommon)