User Modeling in MDE - Data Sheet and Scripts

This replication package contains both the filled out excel template and the used scripts to filter the results at an abstract/title level for ScienceDirect and SpringerLink. For ScienceDirect: 

Perform the search on the ScienceDirect page and download the results.

ScienceDirect proposes a .bib file download. Transform this .bib file into a .ris file using any available online tools for such a conversion.

Execute the script 'sciencedirect_filter_script.py' that will ingest a ScienceDirect.ris file and produce a file called 'output_filtered_included.ris' file.

For SpringerLink:

Perform the search on SpringerLink and download the resulting .csv file. and name it SearchResults.csv.

The downloaded file does not contain the abstracts, that is why the 'springerlink_abstract_scrape.py' script that will perform a scraping to fetch the abstracts. Execute 'python springerlink_abstract_scrape.py SearchResults.py output.csv'. This will start the scraping and save the results in output.csv.

Perform a manual check if some abstracts are missing. Simply text search the .csv file for 'ABSTRACT NOT FOUND ERROR' and manually enter the missing results.

Now use the second script by executing 'python springerlink_filter_script.py' and it will generate a file called 'outputfinalkeyword.ris'

The data extraction excel file contains all the relevant information concerning the data extraction. This includes the reviewed papers, a summary of the numbers, the RQs and metadata.