get_user_network module

features.get_user_network.get_user_network(user_df, conversation_id_col, speaker_id_col)

Takes in data at the user level, and generates a “user_list” per user per conversation. This “user_list” contains the other participating in this conversation. This is a user level feature.

Parameters:

• user_df (pd.DataFrame) – The dataset for which we are generating a “user_list” per user per conversation.

• conversation_id_col (str) – The name of the column containing conversation identifiers.

• speaker_id_col (str) – The name of the column containing the speaker’s unique identifier.

Returns:

Updated user_df with a ‘user_list’ column

Return type:

pd.DataFrame

features.get_user_network.remove_active_user(df, speaker_id_col)

Removes the active user from their own ‘user_list’ in each row of the DataFrame.

Parameters:

• df (pd.DataFrame) – Contains ‘speaker_nickname’ and ‘user_list’ columns.

• speaker_id_col (str) – The name of the column containing the speaker’s unique identifier.

Returns:

Modified pd.DataFrame with ‘user_list’ updated to remove the active user.

Return type:

pd.DataFrame