ted-merge.py

# coding: UTF-8
import nltk
import csv 

#　変数
word_list = open('word-verified.csv','r')
csv = csv.reader(word_list)
file = open('word_merged.csv','w')

word = []
count = []
definition =[]
tag = []
row = 0

# csv読み込み
for c in csv:
    word.append(c[0])
    count.append(c[1])
    definition.append(c[2])

# 検索したい単語
for w in word:
    print(row)
    
    if definition[row] != "*not valid*":
        comp = definition[row]
        
        for i in range(row,len(word)):
            #　　単語定義が同じ場合はカウントを足し合わせ、not vaildを挿入            
            if row != i and comp == definition[i]:
                #　　定義(現在形)を代入
                word[row] = definition[i]
                count[row] = int(count[row]) + int(count[i])
                definition[i] = "*not valid*"

    row = row + 1

row = 0

for w in word:
    #　　not validの場合は飛ばす
    if definition[row] != "*not valid*":
        file.write(str(word[row])+","+str(count[row])+"\n")
    row = row + 1