Following is new feature that I wish to develop in DBpedia’s projects.
-
Currently, DBpedia handles huge amount of data and that is increasing monthly. Users that are fetching/using the data would also be searching for similar datasets (manually) each and every month as new datasets are added. So, basically introducing a Recommendation System on the platform will help in providing users, similar datasets on the basis on their past downloaded/searched datasets.
-
As new users won’t have any past activities, recommendation system would recommend datasets that are trending or most downloaded.
Basically categorizing the datasets and forming several clusters on the basis on similarities would eventually build recommendation system