遍历文本并除去字符串中重复的部分

ls =[]
for word in df['name'].values:
    result = jieba.tokenize(word)
    wd=[]
    start_index = []
    end_index = []
    for tk in result:
        wd.append(tk[0])
        start_index.append(tk[1])
        end_index.append(tk[2])
    data1 = {'wd':wd,
          'start_index':start_index,
          'end_index':end_index}
    data1 = pd.DataFrame(data1)
    keep_last_drop = data1.drop_duplicates(subset=['wd'],keep='last')#(按'word'列去重，保留后一项)注意！：这里的保留后一项keep=last很关键
    keep_False_drop = data1.drop_duplicates(subset=['wd'],keep=False)#未重复的数据
    dup_df=keep_last_drop.append(keep_False_drop).drop_duplicates(subset=['wd'],keep=False)#得出重复的数据
    dup_df = dup_df.loc[dup_df.wd !='('].loc[dup_df.wd!=')']#不考虑重复的括号
    dup_df=dup_df.reset_index()
#     print(dup_df)#输出表格中的重复值位置
    if len(dup_df)==0:
        x=word
    elif len(dup_df)==1:
        x = word[:dup_df.loc[0,'start_index']]+word[dup_df.loc[0,'end_index']:]
    elif len(dup_df) == 2:
        if dup_df.loc[0,'end_index'] == dup_df.loc[1,'start_index']:
            x = word[:dup_df.loc[0,'start_index']]+word[dup_df.loc[1,'end_index']:]
        else:
            x = word[:dup_df.loc[0,'start_index']]+word[dup_df.loc[0,'end_index']:dup_df.loc[1,'start_index']]+word[dup_df.loc[1,'end_index']:]
    elif len(dup_df) ==3:
        x = word[:dup_df.loc[0,'start_index']]+word[dup_df.loc[0,'end_index']:dup_df.loc[1,'start_index']]+word[dup_df.loc[1,'end_index']:dup_df.loc[2,'start_index']]+word[dup_df.loc[2,'end_index']:]
        print('-------------------------------------------------------------------------------重复值为三个：\n',dup_df)
    else:
        print('-------------------------------------------------------------------------------重复值为四个及以上：\n',dup_df)
    ls.append(x)
len(ls)