47 datasets found

Individual text corpora (ICs) were generated from 214 participants with a mean number of 5 million word tokens.

The OpenWebText corpus is a collection of text data from the web.

The RECIPES dataset contains various annotated states (e.g., shape, composition, location, etc.) for ingredients in cooking recipes.

The PROPARA dataset comprises procedural text about scientific processes. The location states of participant entities at each time step (sentence) in these processes are labeled...

The dataset used in the paper is a collection of 1.1M tweets from Twitter, with approximately 3000 tweets per user from various domains such as politics, sports, entertainment,...

The STS benchmark (Cer et al., 2017) and SICK-Relatedness dataset (Marelli et al., 2014) respectively contain 8.6K and 9.8K labeled sentence pairs, the sizes of which are...

Radiology reports dataset is used to test the proposed Hierarchical Latent Word Clustering algorithm.

NIPS dataset is used to test the proposed Hierarchical Latent Word Clustering algorithm.

Hierarchical Latent Word Clustering dataset is used to test the proposed Hierarchical Latent Word Clustering algorithm.

The dataset used in this paper is a collection of short texts, including tweets, Pascal Flickr captions, and search snippets.

The Pile dataset contains 3.5 million samples of diverse text for language modeling.

Probase is a probabilistic knowledge base and it contains millions of entities and concepts. One of the advantages of Probase is that in comparison with the well-known knowledge...

The dataset used in the paper is a word2vec embedding trained on a corpus of Google News texts.

The Twitter Dataset is a collection of tweets annotated with Plutchik's emotions, consisting of tweets in three different languages: English, Dutch, and German.

The dataset used in this paper is the PubMed abstracts dataset, which contains approximately 11 million abstracts.

The Penn Tree Bank dataset is a corpus split into a training, validation and testing set of 929k words, a validation set of 73k words, and a test set of 82k words. The...

MS MARCO V1 corpus

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,...

Matrix factorization (MF) and Autoencoder (AE) are among the most successful approaches of unsupervised learning.

The dataset used in the paper is the Wikipedia dataset, which contains over six million English Wikipedia articles with a full-text field associated with 50 training queries...