我有一個帶有預定義文本的 Xlsx 文件，其中只有一列。用戶將輸入一個或多個單詞，輸出將是包含一個或多個單詞的文本。import numpy as npimport pandas as pdimport timeimport refrom sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer, TfidfTransformerfrom sklearn.metrics.pairwise import linear_kernel, cosine_similarityfrom sklearn.metrics.pairwise import pairwise_distancesimport pickledef load_df(path):    df = pd.read_excel(path)    print(df.shape)    return dfdef splitDataFrameList(df, target_column, separator):    def splitListToRows(row, row_accumulator, target_column, separator):        split_row = row[target_column].split(separator)        for s in split_row:            new_row = row.to_dict()            new_row[target_column] = s            row_accumulator.append(new_row)    new_rows = []    df.apply(splitListToRows, axis=1, args=(new_rows, target_column, separator))    new_df = pd.DataFrame(new_rows)    return new_dfclass Autocompleter:    def __init__(self):        pass    def import_json(self, json_filename):        print("load Excel file...")        df = load_df(json_filename)        return df    def process_data(self, new_df):        # print("select representative threads...")        # new_df = new_df[new_df.IsFromCustomer == False]        print("split sentenses on punctuation...")        for sep in ['. ', ', ', '? ', '! ', '; ']:            new_df = splitDataFrameList(new_df, 'UserSays', sep)        print("UserSays Cleaning using simple regex...")在輸入中，如果我什么都不輸入，它會為我提供這個輸出['How to access outlook on open network?', 'Email access outside ril network', 'Log in outlook away from office']這是不希望的，如果只有一個文本匹配它會給出以下輸出input - sccm['What is sccm', 'How to access outlook on open network?', 'Email access outside ril network']我希望以這樣的方式輸出，如果輸入的單詞或單詞不存在于 xlsx 文件中，那么輸出不應該返回任何東西。和