The NewSHead dataset is a multi-doc headline dataset used in NHNet for training a headline summarization model.

Last update: Jul 15, 2022

Related tags

Overview

NewSHead

This repository contains the raw dataset used in NHNet [1] for the task of News Story Headline Generation. The code of data processing and training is available under Tensorflow Models - NHNet.

A news story is defined as a list of articles about the same event with a coherent topic. The released dataset contains 369,940 English stories with 932,571 unique URLs, among which we have 359,940 stories for training, 5,000 for validation, and 5,000 for testing, respectively. Each news story contains at least three (and up to five) articles.

The dataset is collected from news stories published between May 2018 and May 2019, where a proprietary clustering algorithm iteratively loads articles published in a time window and groups them based on content similarity¹. Up to five representative articles are picked from the cluster for generating the story headline². Curators from a crowd-sourcing platform are requested to provide a headline of up to 35 characters to describe the major information covered by the story.

Example Headlines:

International Space Station flyover
Drilling for oil in Pakistan
Review of 'Mr. Local'
MLB: Pirates vs Padres
Braves re-sign Jerry Blevins

Download Link

Tools to Process

Citation

If you use or discuss this dataset in your work, please cite our paper:

@InProceedings{headline2020,
  title = {{Generating Representative Headlines for News Stories}},
  author = {Gu, Xiaotao and Mao, Yuning and Han, Jiawei and Liu, Jialu and Yu, Hongkun and Wu, You and Yu, Cong
and Finnie, Daniel and Zhai, Jiaqi and Zukoski, Nicholas},
  booktitle = {Proc. of the the Web Conf. 2020},
  year = {2020}
}

Analysis

We did broad topic analysis for the 932,571 articles in our dataset. A histogram is attached as below.

Among all the 369,940 stories, each headline is required to be between 10 and 35 characters.

Such lengths of curated story headlines are much shorter than traditional summaries, and even shorter than article titles in our dataset depicted below

References

[1] Xiaotao Gu, Yuning Mao, Jiawei Han, Jialu Liu, Hongkun Yu, You Wu, Cong Yu, Daniel Finnie, Jiaqi Zhai and Nicholas Zukoski "Generating Representative Headlines for News Stories": https://arxiv.org/abs/2001.09386. World Wide Web Conf. (WWW’2020).

Footnote

1 Clustering algorithm could contain noise. It is possible if some articles in a story are not relevant to the rest.

2 These articles presented don't necessarily map to articles we would show on Google products such as Search and News App.

You might also like...

An Analysis Toolkit for Natural Language Generation (Translation, Captioning, Summarization, etc.)

VizSeq is a Python toolkit for visual analysis on text generation tasks like machine translation, summarization, image captioning, speech translation

310 Feb 1, 2021

Summarization, translation, sentiment-analysis, text-generation and more at blazing speed using a T5 version implemented in ONNX.

Summarization, translation, Q&A, text generation and more at blazing speed using a T5 version implemented in ONNX. This package is still in alpha stag

137 Feb 1, 2021

Package for controllable summarization

summarizers summarizers is package for controllable summarization based CTRLsum. currently, we only supports English. It doesn't work in other languag

72 Dec 7, 2022

The guide to tackle with the Text Summarization

1.2k Dec 30, 2022

FactSumm: Factual Consistency Scorer for Abstractive Summarization

FactSumm: Factual Consistency Scorer for Abstractive Summarization FactSumm is a toolkit that scores Factualy Consistency for Abstract Summarization W

83 Jan 9, 2023

code for modular summarization work published in ACL2021 by Krishna et al

This repository contains the code for running modular summarization pipelines as described in the publication Krishna K, Khosla K, Bigham J, Lipton ZC

Approximately Correct Machine Intelligence (ACMI) Lab

21 Nov 24, 2022

code for modular summarization work published in ACL2021 by Krishna et al

This repository contains the code for running modular summarization pipelines as described in the publication Krishna K, Khosla K, Bigham J, Lipton ZC

6 Jun 4, 2021

This repository contains the code, data, and models of the paper titled "XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Languages" published in Findings of the Association for Computational Linguistics: ACL 2021.

XL-Sum This repository contains the code, data, and models of the paper titled "XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Lang

189 Jan 2, 2023

Codes for processing meeting summarization datasets AMI and ICSI.

Meeting Summarization Dataset Meeting plays an essential part in our daily life, which allows us to share information and collaborate with others. Wit

39 Dec 14, 2022

Releases(v1.0-config)

v1.0-config(Jun 9, 2020)

Config file for crawling articles using news-please.
Source code(tar.gz)
Source code(zip)
news_please.zip(28.38 MB)
v1.0(Mar 27, 2020)

First version includes 369,940 news stories with 932,571 unique URLs, among which we have 359,940 stories for training, 5,000 for validation, and 5,000 for testing, respectively.
Source code(tar.gz)
Source code(zip)
dataset.zip(66.60 MB)

The NewSHead dataset is a multi-doc headline dataset used in NHNet for training a headline summarization model.

Related tags

Overview

NewSHead

Citation

Analysis

References

Footnote

You might also like...

An Analysis Toolkit for Natural Language Generation (Translation, Captioning, Summarization, etc.)

Summarization, translation, sentiment-analysis, text-generation and more at blazing speed using a T5 version implemented in ONNX.

Package for controllable summarization

The guide to tackle with the Text Summarization

FactSumm: Factual Consistency Scorer for Abstractive Summarization

code for modular summarization work published in ACL2021 by Krishna et al

code for modular summarization work published in ACL2021 by Krishna et al

This repository contains the code, data, and models of the paper titled "XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Languages" published in Findings of the Association for Computational Linguistics: ACL 2021.

Codes for processing meeting summarization datasets AMI and ICSI.

Releases(v1.0-config)

v1.0-config(Jun 9, 2020)

v1.0(Mar 27, 2020)

Owner

Google Research Datasets

Practical Natural Language Processing Tools for Humans is build on the top of Senna Natural Language Processing (NLP)

Graphical user interface for Argos Translate

Material for GW4SHM workshop, 16/03/2022.

SAVI2I: Continuous and Diverse Image-to-Image Translation via Signed Attribute Vectors

This repository serves as a place to document a toy attempt on how to create a generative text model in Catalan, based on GPT-2

An ActivityWatch watcher to pose questions to the user and record her answers.

A python package to fine-tune transformer-based models for named entity recognition (NER).

Natural Language Processing library built with AllenNLP 🌲🌱

An easier way to build neural search on the cloud

Code and dataset for the EMNLP 2021 Finding paper "Can NLI Models Verify QA Systems’ Predictions?"

GPT-3 command line interaction

Transformers4Rec is a flexible and efficient library for sequential and session-based recommendation, available for both PyTorch and Tensorflow.

Fake news detector filters - Smart filter project allow to classify the quality of information and web pages

BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese

Japanese Long-Unit-Word Tokenizer with RemBertTokenizerFast of Transformers

Data and evaluation code for the paper WikiNEuRal: Combined Neural and Knowledge-based Silver Data Creation for Multilingual NER (EMNLP 2021).

Research code for the paper "Fine-tuning wav2vec2 for speaker recognition"

NLTK Source

Longformer: The Long-Document Transformer

REST API for sentence tokenization and embedding using Multilingual Universal Sentence Encoder.