Extract Work Experience Years from CSV Using Python Regex: 4 Practical Methods

1. Introduction

In a recent Python community question, a user asked how to extract numeric work experience years from a CSV column for later regression analysis.

2. Solution Process

Four methods are presented, two contributed by the author and two by another contributor.

Method 1

Custom function using conditional checks and regular expressions to handle various formats.

def work_year(y):
    y = y.strip()
    if y == '无需经验':
        return 0
    elif y == '在校生/应届生':
        return 0
    elif '-' in y and '年经验' in y:
        low_experience = re.findall(re.compile('(
        d*\.?\d+)'), y)[0]
        high_experience = re.findall(re.compile('(
        d?\.?\d+)'), y)[1]
        s = round((float(low_experience) + float(high_experience)) / 2, 0)
        return s
    elif '年经验' in y or '年以上经验' in y:
        year = re.findall(re.compile('^(\d+)'), y)[0]
        return year
    else:
        return y

df['new'] = df['工作经验'].apply(work_year)
df.head()

Result shown in the following image:

Method 2

Similar logic with a slight variation in handling hyphenated ranges.

def work_year(y):
    if y == '无需经验':
        return 0
    elif y == '在校生/应届生':
        return 0
    elif '-' in y:
        low_experience = re.findall(re.compile('(\d*\.?\d+)'), y)[0]
        high_experience = re.findall(re.compile('(\d?\.?\d+)'), y)[1]
        s = round((float(low_experience) + float(high_experience)) / 2, 0)
        return s
    elif y[0].isnumeric():
        year = re.findall(re.compile('^(\d+)'), y)[0]
        return year
    else:
        return y

df['col1'] = df['工作经验'].str.strip().apply(work_year)
df

Method 3

Uses a single regular expression to capture one or two numbers and computes their average.

def work_year(y):
    search_year = re.search(r'(\d+)?-?(\d+)', y)

    def average(args):
        x = tuple(args)
        length = len(x)
        return round(sum(x) / length, 0)

    if search_year:
        return average([int(i) for i in search_year.groups() if i])
    else:
        return 0

df['new1'] = df['工作经验'].apply(work_year)

Method 4

Leverages pandas str.extract with the same regex and computes the mean directly.

df['new2'] = df['工作经验'].str.extract(r'(\d+)?-?(\d+)').astype(float).mean(axis=1).fillna(0).round(0)

3. Summary

The four approaches demonstrate practical ways to parse work experience strings in a CSV file, allowing the extracted numeric values to be used for further analysis such as multiple regression.