Eng-Janvier/Kinyarwanda-keywords_extractor

This a package you can use for extracting kinyarwanda keywords from any given kinyarwanda text for a free of charge. This package was tested for extracting kinyarwanda text from more 73,000 kinyarwanda sentences and it accuracy was good. It is allowed to customize it depends on your project scope First install the package with: "!pip install nijas_keywords_extractor" by using anaconda distribution or jupyter notebook This package contains 13 functions to be executed 1. gen_keywords(df,col,stop_words): Thiss function accept the dataframe as argument and the column name to be transformed into keywords and a list of stop words 2. gen_data_fields(df,keywords,stop_words): This function accept 3 argument(dataframe, list of keywords generated and the list of words to ignore) 3. update_DF(fields,list_fields):This function accept 3 argument(list of dataframe columns or fields names and the data field values) 4. rename_keywords(df,col,dfmapa,col_map): This function adds a new fields that mapp each keyword into a new meaningful phrase. The function accepts 4 arguments (dataframe, column name that contains keywords,external dataframe that contains a column called col_map to be used for mapping ) 5. clean_keywords(df,dfmapa):accept 2 arguments(dataframe generated by the second function and the external dataset data contains the mapping data ) This function transform the keywords into clean data using external dataset. 6. remove_special_character(df,columns): accept 2 arguments(dataframe and the list of columns to process). 7. remove_extra_spaces(df,columns): accept 2 arguments(dataframe and the list of columns to process). 8. map_keywords(df,dat,questions): accept 2 arguments(cleaned dataset above,external mappa dataset that might contain keywords and corresponded mapping for each question; and the list of 4 questions asked) 9. keywords_counts(df,col): counts the number of each value in a particular column(col) of a dataframe(df) 10. return_df_with_seled_kwds(df,dat …