irds/tripclick_train_torso

`tripclick/train/torso`数据集由ir-datasets包提供，主要用于文本检索任务。该数据集包含105,964个查询（即主题）和966,898个相关性评估（qrels）。文档数据需要使用`irds/tripclick`数据集。

The `tripclick/train/torso` dataset is provided by the ir-datasets package and is primarily intended for text retrieval tasks. It contains 105,964 queries (i.e., topics) and 966,898 relevance judgments (qrels). The document data for this dataset must be accessed using the `irds/tripclick` dataset.

原始信息汇总

数据集概述

数据集名称

tripclick/train/torso

数据来源

• 源数据集：irds/tripclick

数据内容

• queries (查询/主题)：总数105,964
• qrels (相关性评估)：总数966,898
• docs 数据集：irds/tripclick

数据使用示例

python from datasets import load_dataset

queries = load_dataset(irds/tripclick_train_torso, queries) for record in queries: record # {query_id: ..., text: ...}

qrels = load_dataset(irds/tripclick_train_torso, qrels) for record in qrels: record # {query_id: ..., doc_id: ..., relevance: ..., iteration: ...}

引用信息

@inproceedings{Rekabsaz2021TripClick, title={TripClick: The Log Files of a Large Health Web Search Engine}, author={Navid Rekabsaz and Oleg Lesota and Markus Schedl and Jon Brassey and Carsten Eickhoff}, year={2021}, booktitle={SIGIR} }