This repository contains resources developed within the following paper:

S. Zhang and K. Balog. Ad Hoc Table Retrieval using Semantic Similarity. In: Proceedings of the Web Conference 2018 (WWW '18), April 2018.

The table corpus is WikiTables, which comprises 1.6M tables extracted from Wikipedia. We proproceeed it and make it public downloadable here.

The data/queries.txt file contains the search queries. Queries #1-#30 queries constitute Query subset 1 (QS-1), queries #31-#60 constitute Query subset 2 (QS-2).

The data/qrels.txt file contains the relevance assessments (in TREC qrels format).

We utilize word2vec trained on Google news, and you can find it here. You can find the graph embeddings here.

The rankings/ folder contains the table rankings generated by the various methods (in TREC runfile format).

The evaluation scores are reported using trec_eval.

@inproceedings{Zhang:2018:AHT,
    author = {Zhang, Shuo and Balog, Krisztian},
    title = {Ad Hoc Table Retrieval using Semantic Similarity},
    booktitle = {Proceedings of the Web Conference 2018},
    year = {2018},
    pages = {1553--1562},
}

If you have any questions, please contact Shuo Zhang at [email protected].