viberary_training_data.md

Data source:

https://sites.google.com/eng.ucsd.edu/ucsdbookgraph/home

@inproceedings{DBLP:conf/recsys/WanM18,
  author       = {Mengting Wan and
                  Julian J. McAuley},
  editor       = {Sole Pera and
                  Michael D. Ekstrand and
                  Xavier Amatriain and
                  John O'Donovan},
  title        = {Item recommendation on monotonic behavior chains},
  booktitle    = {Proceedings of the 12th {ACM} Conference on Recommender Systems, RecSys
                  2018, Vancouver, BC, Canada, October 2-7, 2018},
  pages        = {86--94},
  publisher    = {{ACM}},
  year         = {2018},
  url          = {https://doi.org/10.1145/3240323.3240369},
  doi          = {10.1145/3240323.3240369},
  timestamp    = {Mon, 22 Jul 2019 19:11:02 +0200},
  biburl       = {https://dblp.org/rec/conf/recsys/WanM18.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

Run in DuckDB:

CREATE TABLE goodreads_reviews AS SELECT * FROM read_json_auto('/Users/vicki/viberary/data/goodreads_reviews_dedup.json',ignore_errors='true',lines='true') ;

CREATE TABLE goodreads AS SELECT * FROM read_json_auto('/Users/vicki/viberary/data/goodreads_books.json',lines='true');

CREATE TABLE goodreads_authors AS SELECT * FROM read_json_auto('/Users/vicki/viberary/data/goodreads_book_authors.json',lines='true');

CREATE table authorid as select REGEXP_EXTRACT(authors, '[0-9]+') as author_id, title, description, average_rating, book_id FROM goodreads;

COPY (SELECT review_text,title,description,authorid.average_rating, goodreads_authors.name as author FROM authorid  JOIN goodreads_reviews  ON authorid.book_id = goodreads_reviews.book_id JOIN goodreads_authors ON authorid.author_id = goodreads_authors.author_id where authorid.author_id NOT ILIKE '' ) TO '20230630_training.parquet' (FORMAT PARQUET);