27 datasets found

The MQ2009 query set is a large-scale query dataset, containing 200,000 queries.

The Wikimarks dataset, which consists of 30 million deduplicated paragraphs from all Wikipedia articles.

The dataset used for the Retrieve-Cluster-Summarize system, consisting of 117 article-level queries and 126 test queries.

The dataset used in this paper for keyphrase extraction utilizing document logical structure.

The dataset used in this paper for keyphrase extraction from academic articles.

The BBC News dataset was used for sentiment analysis of news articles.

The NIPS full paper dataset is a collection of text documents.

The ClueWeb09B collection is a large-scale web search dataset, containing 31 million web pages, 31 million queries, and 1.5 billion documents.

The dataset used in this paper for novel token prediction in e-commerce search.

The dataset contains three passage-ranking datasets: WikipassageQA, InsuranceQA v2, and MS-MARCO.

This paper proposes a method for learning-to-rank with contextual neural language modeling.

This paper proposes a method for learning to rank, which is a key task in information retrieval.

The dataset used in the paper is the TREC Dynamic Domain 2015 ad-hoc retrieval task, which includes search result diversification. The dataset consists of 23 official runs and...

The dataset used in the paper is the TREC Web Track 2014 ad-hoc retrieval task, which includes search result diversification. The dataset consists of 50 test topics and 10,000...

Web pages are a valuable source of information for many natural language processing and information retrieval tasks. Extracting the main content from those documents is...

The dataset used in the paper is the Robust04 dataset, a news corpus containing 0.5M documents and 249 queries.

The BEIR dataset is a large-scale zero-shot evaluation dataset for information retrieval models, consisting of 13,000 documents and 1,000 questions.

The dataset used in the paper is a collection of search engine result pages (SERPs) with their corresponding relevance scores.

The dataset used in the paper is a subset of the Wikipedia corpus, consisting of 7500 English Wikipedia articles belonging to one of the following categories: People, Cities,...

The dataset used for training and evaluating IR systems, containing a large collection of documents and queries.