Overview

Istella-S LETOR is a smaller sample of the Istella Learning to Rank dataset, obtained by sampling irrelevant query-document pairs. It is suitable for experiments requiring a more compact benchmark.

• 33,018 queries; 220 features per query-document pair
• 3,408,630 pairs; approximately 103 examples per query
• Training/validation/test split: 60% / 20% / 20%

How to cite this dataset

If you use this dataset in your research, please acknowledge Istella SpA and cite the following publications:

Claudio Lucchese, Franco Maria Nardini, Salvatore Orlando, Raffaele Perego, Fabrizio Silvestri, Salvatore Trani. 2016. Post-Learning Optimization of Tree Ensembles for Efficient Ranking. Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval. 10.1145/2911451.2914763.

@inproceedings{istellaS2016,
  author = {Claudio Lucchese and Franco Maria Nardini and Salvatore Orlando and Raffaele Perego and Fabrizio Silvestri and Salvatore Trani},
  title = {Post-Learning Optimization of Tree Ensembles for Efficient Ranking},
  booktitle = {Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval},
  year = {2016},
  doi = {10.1145/2911451.2914763},
  url = {https://doi.org/10.1145/2911451.2914763},
  pages = {949--952},
}

Domenico Dato, Claudio Lucchese, Franco Maria Nardini, Salvatore Orlando, Raffaele Perego, Nicola Tonellotto, Rossano Venturini. 2016. Fast Ranking with Additive Ensembles of Oblivious and Non-Oblivious Regression Trees. ACM Transactions on Information Systems. 10.1145/2987380.

@article{istellaLetor2016,
  author = {Domenico Dato and Claudio Lucchese and Franco Maria Nardini and Salvatore Orlando and Raffaele Perego and Nicola Tonellotto and Rossano Venturini},
  title = {Fast Ranking with Additive Ensembles of Oblivious and Non-Oblivious Regression Trees},
  journal = {ACM Transactions on Information Systems},
  year = {2016},
  doi = {10.1145/2987380},
  url = {https://doi.org/10.1145/2987380},
  volume = {35},
  number = {2},
  articleno = {15},
  numpages = {31},
}

Download

The dataset is available solely for non-commercial use. Read and accept the license before downloading the files. Read the license.