Print Next Line In Pandas Dataframe

I have a pandas dataframe consisting of chat bot data and I want to print the User's input in response to a specific chat bot message eg: (Row 1) Bot: Hi, What are your hobbies? (R

Solution 1:

You should first get the index of the row that matches the question by index function of DataFrame. To get partial match to your question, use str.contains.

df = pd.DataFrame({'data':
               ["(Row 1) Bot: Hi, What are your hobbies?",
                "(Row 2) User1: Cricket, FootBall",                               
                "(Row 3) Bot: Hi, What is your name?",
                "(Row 4) User2: Alexa",
                "(Row 5) Bot: Hi, What are your hobbies?",
                "(Row 6) User3: Tennis, Baseball"]
               })

idx = df[df['data'].str.contains("Hi, What are your hobbies?")].index.tolist()
for i in idx:
  if i < len(df) - 1:
    print(df.iloc[i + 1].values[0])

Output:

(Row2) User1: Cricket, FootBall
(Row6) User3: Tennis, Baseball

So, in the above code, idx holds a list of indexes that match your query. In the last line, you print the values of the next row corresponding those indices.

Solution 2:

Using the Data Frame declaration from @CentAu

import pandas as pd

Chat_Column = pd.DataFrame({'data':
               ["(Row 1) Bot: Hi, What are your hobbies?",
                "(Row 2) User1: Cricket, FootBall",
                "(Row 3) Bot: Hi, What is your name?",
                "(Row 4) User2: Alexa",
                "(Row 5) Bot: Hi, What are your hobbies?",
                "(Row 6) User3: Tennis, Baseball"]
               })

for ndx, row in Chat_Column.iterrows():
    if"Bot: Hi, What are your hobbies?"in row["data"]:
        print(Chat_Column.iloc[ndx+1]["data"])

Solution 3:

d = {'a':['Question1','Answer1','Question2','Answer2','Question3','Answer3']}
df = pd.DataFrame(d)
printdf['a'].shift(-1)[df['a'] == 'Question2'].values[0]

But I would recommend putting the dataframe into two columns (Quesiton and Answer).

Point of note - using pandas native indexing will be considerably faster than iterating through a list of strings and testing if statements on every iteration. That said, it is probably irrelevant if it is only a couple hundred / thousand lines, but could be obvious if the DataFrame increases in size.