Skip to content

Latest commit

 

History

History
81 lines (53 loc) · 2.18 KB

README.md

File metadata and controls

81 lines (53 loc) · 2.18 KB
Raw

SeEn

Sequential Enrichment of Recommendation Datasets

All the data used on this work is available at: Barros, Marcia (2022): seen_datasets. figshare. Dataset. https://doi.org/10.6084/m9.figshare.18857549.v1

Astro datasets

Create the sequential dataset ARM

git: https://github.com/lasigeBioTM/cARM

input files: Dias_updated_catalogue: /seen_datasets/astro_data/CATALOG-FINAL.csv

  1. Follow the Configuration steps on README

output is a sqlite database. The one created and used in this work is available at: /seen_datasets/astro_data/arm2021.db

  1. Create the sequential dataset:

cd authorClusterMatrix/ python createAuthorClusterMatrix_seq.py

  • aRMSeq file available at: /seen_datasets/astro_data/astro20_user_item_rating_item_name_year_20_ordered.csv

SeEn: https://github.com/lasigeBioTM/SeEn

create the enriched dataset ARM

  1. Create the similarity dataset for the open clusters of stars:

python ARM/astro_clusters_similarity/create_features_file.py

Used in this work:

  • input: membership tables: /seen_datasets/astro_data/memberships-tables/
  • output: features file: /seen_datasets/astro_data/clusters_features_mean.csv

ARM/astro_clusters_similarity/create_sim_file.py

Used in this work:

  • input: features file: /seen_datasets/astro_data/clusters_features_mean.csv
  • output: similarity file: /seen_datasets/astro_data/clusters_similatiry.csv
  1. create the enriched datasets: python ARM/astro_sampler/astro_sampler.py

Used in this work:

  • input:similarity : /seen_datasets/astro_data/clusters_similatiry.csv
  • output: enriched datasets: /seen_datasets/astro_data/seen_cos/

The datasets are ready to be tested.

#################################

Chemical datasets

Create the sequential dataset chERM

https://github.com/lasigeBioTM/CheRM

  • used in this work (original sequence): /seen_datasets/chem_data/cherm_20_order_year.csv

Create the database with the semantic similarity

https://github.com/lasigeBioTM/SemanticSimDBcreator

  • Used in this work: /seen_datasets/chem_data/cherm_sim_20_dump.sql

create the enriched dataset

python CHERM/semantic_sampler/src/main.py

  • Used in this work (enriched sequences): /seen_datasets/chem_data/sim_lin/