import pandas as pd

reviews = pd.read_csv('https://raw.githubusercontent.com/umsi-data-science/data/main/amazon_food_reviews.zip')

reviews_sample = reviews.head(10)

reviews_sample

reviews_sample.ProfileName.str.lower()

reviews_sample.ProfileName.str.upper()

reviews_sample.Summary.str.len()

reviews_sample.columns

reviews_sample

reviews_sample.columns.str.lower()

reviews_sample.columns.str.strip().str.lower().str.replace(' ','_')

reviews_sample.columns = reviews_sample.columns.str.strip().str.lower().str.replace(' ','_')

reviews_sample

reviews_sample.productid.head()

reviews_sample.productid.str.split('00')

reviews_sample.productid.str.split('00').str.get(1)

reviews_sample.productid.str.split('00').str[1]

reviews_sample.summary

reviews_sample.summary.str.lower().str.replace('dog','health')

reviews_sample.summary.str.lower().str.replace('Chris|Sam|Sidra|Ye Chan','_SOMETHING_')

reviews_sample.summary.head(10)

reviews_sample.summary.str.extract(r'(Dog)')

reviews_sample.summary

reviews_sample.summary.str.extract(r'(Dog|Taffy)')

reviews_sample.summary.str.extract(r'(Dog|[Tt]affy)')

# returns a Series reviews_sample.summary.str.extract(r'(Dog|[Tt]affy)', expand = False)

reviews_sample.summary.str.extractall(r'(Dog|[Tt]affy)')

reviews_sample.summary

reviews_sample.summary.str.extractall(r'(a)')

reviews_sample.text.iloc[0]

pattern = r'[Gg]ood' # The same as (pattern = r'Good/good')

reviews_sample.text.str.contains(pattern)

reviews_sample.text.str.match(pattern)

pattern = r'.*[Gg]ood.*'

reviews_sample.text.str.match(pattern)

reviews.Text.str.contains(r'\b[Vv]egan\b').sum()

reviews.Text.str.contains('<.*>').sum()

reviews['text_no_html'] = reviews.Text.str.replace('<[^<]+?>', '', regex=True) reviews