如何有效检查Spark Dataframe中是否包含单词列表? [英] How to efficiently check if a list of words is contained in a Spark Dataframe?

问题描述

使用PySpark数据帧，我正在尝试尽可能高效地执行以下操作.我有一个数据框，其中的一列包含文本和我要过滤行的单词列表.所以:

Using PySpark dataframes I'm trying to do the following as efficiently as possible. I have a dataframe with a column which contains text and a list of words I want to filter rows by. So:

数据框看起来像这样

df:
col1    col2   col_with_text
a       b      foo is tasty
12      34     blah blahhh
yeh     0      bar of yums

列表将为list = [foo,bar] 因此结果将是:

The list will be list = [foo,bar] And thus result will be:

result:
col1    col2   col_with_text
a       b      foo
yeh     0      bar

然后，不仅将完成相同的字符串匹配，而且还将通过使用SequenceMatcher进行测试以进行相似性测试.这是我已经尝试过的:

Afterwards not only identical string matching will be done but also tested for similarity by using SequenceMatcher or so. This is what I already tried:

def check_keywords(x):
   words_list = ['foo','bar']

   for word in x
       if word == words_list[0] or word == words_list[1]:
           return x

result = df.map(lambda x: check_keywords(x)).collect()

不幸的是我没有成功，有人可以帮我吗? 预先感谢.

Unfortunately I was unsuccesfull, could someone help me out? Thanks in advance.

推荐答案

您应该考虑使用pyspark sql模块函数而不是编写UDF，因为它有几个基于regexp的函数:

You should consider using pyspark sql module functions instead of writing a UDF, there are several regexp based functions:

首先让我们从一个更完整的示例数据框架开始:

First let's start with a more complete sample data frame:

df = sc.parallelize([["a","b","foo is tasty"],["12","34","blah blahhh"],["yeh","0","bar of yums"], 
                     ['haha', '1', 'foobar none'], ['hehe', '2', 'something bar else']])\
    .toDF(["col1","col2","col_with_text"])

如果要根据行中是否包含words_list中的单词之一来过滤行，可以使用rlike:

If you want to filter lines based on whether they contain one of the words in words_list, you can use rlike:

import pyspark.sql.functions as psf
words_list = ['foo','bar']
df.filter(psf.col('col_with_text').rlike('(^|\s)(' + '|'.join(words_list) + ')(\s|$)')).show()

    +----+----+------------------+
    |col1|col2|     col_with_text|
    +----+----+------------------+
    |   a|   b|      foo is tasty|
    | yeh|   0|       bar of yums|
    |hehe|   2|something bar else|
    +----+----+------------------+

如果要提取与正则表达式匹配的字符串，则可以使用regexp_extract:

If you want to extract the strings matching the regular expression, you can use regexp_extract:

df.withColumn(
        'extracted_word', 
        psf.regexp_extract('col_with_text', '(?=^|\s)(' + '|'.join(words_list) + ')(?=\s|$)', 0))\
    .show()

    +----+----+------------------+--------------+
    |col1|col2|     col_with_text|extracted_word|
    +----+----+------------------+--------------+
    |   a|   b|      foo is tasty|           foo|
    |  12|  34|       blah blahhh|              |
    | yeh|   0|       bar of yums|           bar|
    |haha|   1|       foobar none|              |
    |hehe|   2|something bar else|              |
    +----+----+------------------+--------------+

这篇关于如何有效检查Spark Dataframe中是否包含单词列表?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！