list index out of range error for large data set

Question

I'm trying to implement sentiment analysis model for analysing data. The code below is working for dataset below 50, but there are 10000+ of data in the CSV file.

I have googled a lot but still couldn't find any solution.

Here's the code:

import re
import csv
fh=open("dataset.csv","r")
dataset={}
no_of_items={}
for row in reader:
     no_of_items.setdefault(row[1],0)
     no_of_items[row[1]]+=1
     dataset.setdefault(row[1],{})
     split_data=re.split('[^a-zA-Z\']',row[0])

for i in split_data:

    if len(i) > 2:          
        dataset[row[1]].setdefault(i.lower(),0)         
        dataset[row[1]][i.lower()]+=1           
        feature_set.setdefault(i.lower(),{})            
        feature_set[i.lower()].setdefault(row[1],0)         
        feature_set[i.lower()][row[1]]+=1

Output:

no_of_items.setdefault(row[1],0) IndexError: list index out of range