semantic-systems / nfdi-search-engine

A lightweight, KG-driven search engine over different endpoints and APIs
https://nfdi-search.nliwod.org/
MIT License
5 stars 3 forks source link

Integration of GESIS and CORDIS into NFDI search engine #80

Closed Najeeb-Shams closed 1 year ago

Najeeb-Shams commented 1 year ago

Description:

This pull request aims to integrate GESIS and CORDIS into the NFDI search engine, enhancing and enriching its functionality.

Changes Made:

The relevant parts of the project have been synchronized and updated to support the integration.

huntila commented 1 year ago

@Najeeb-Shams, please check the returned values. For example, for the search terms 'Life in Digital Europe', the result comes from GESIS, i.e., id is invalid, authors should be a string not a list [], and shorten the description it is too long.

Example search result returned from search_gesis.py for 'Life in Digital Europe'

" https://search.gesis.org/research_data/datasearch-httpwww-da-ra-deoaip--oaioai-da-ra-de9087 Life in Digital Europe - 2003 Adults in the countries plus their partners if present ['Anderson, B.', 'Brynin, M.', 'Raban, Y.', 'Vicario, L.', 'Ling, R.', 'Nokolov, R.', 'Haisken-deNew, J. P.', 'Commission of the European Union']

https://search.gesis.org/research_data/datasearch-httpseasy-dans-knaw-nloai--oaieasy-dans-knaw-nleasy-dataset62234 Development of an Automatic Pollen Classification System - 2015 This research applied image processing and machine learning techniques to develop an automatic process for the identification of 5 pollen species using images from a digital microscope. ['Chudyk, C. (i3mainz, University of Applied Sciences Mainz)'] "

huntila commented 1 year ago

Also in search_gesis.py

Najeeb-Shams commented 1 year ago

I have tested the GESIS API and thoroughly reviewed the Genesis-API metadata. As a result, the API integration is functioning smoothly, and it successfully retrieves the required data in line with GESIS standards. If a particular dataset or article is not found, you will see the message "Sorry, no entry.....". This applies both to the API search and if you manually search for the same title/article on the GESIS platform.

Any empty lines, authors' type converted to string, and reductions in the description have been appropriately addressed and updated. All of these changes have been incorporated into a new pull request.