Build neural network models to identify and correct news headlines exaggerating obesity-related scientific findings

Purpose:Media exaggerations of health research may confuse readers’understanding,erode public trust in science and medicine,and cause disease mismanagement.This study built artificial intelligence(AI)models to automatically identify and correct news headlines exaggerating obesity-related research findings.Design/methodology/approach:We searched popular digital media outlets to collect 523 headlines exaggerating obesity-related research findings.The reasons for exaggerations include:inferring causality from observational studies,inferring human outcomes from animal research,inferring distant/end outcomes(e.g.,obesity)from immediate/intermediate outcomes(e.g.,calorie intake),and generalizing findings to the population from a subgroup or convenience sample.Each headline was paired with the title and abstract of the peer-reviewed journal publication covered by the news article.We drafted an exaggeration-free counterpart for each original headline and fined-tuned a BERT model to differentiate between them.We further fine-tuned three generative language models-BART,PEGASUS,and T5 to autogenerate exaggeration-free headlines based on a journal publication’s title and abstract.Model performance was evaluated using the ROUGE metrics by comparing model-generated headlines with journal publication titles.Findings:The fine-tuned BERT model achieved 92.5%accuracy in differentiating between exaggeration-free and original headlines.Baseline ROUGE scores averaged 0.311 for ROUGE-1,0.113 for ROUGE-2,0.253 for ROUGE-L,and 0.253 ROUGE-Lsum.PEGASUS,T5,and BART all outperformed the baseline.The best-performing BART model attained 0.447 for ROUGE-1,0.221 for ROUGE-2,0.402 for ROUGE-L,and 0.402 for ROUGE-Lsum.Originality/value:This study demonstrated the feasibility of leveraging AI to automatically identify and correct news headlines exaggerating obesity-related research findings.

Features of English News Headlines

Nowadays Chinese people are attaching great importance to English news that not only offers worldwide information but gives a hand to English learners with their improvement in English.As is called the eyes of English news,headlines are partic ularly important.Generally speaking,English news headlines are precise,brief and as attractive as possible.They have their own grammar features and vocabulary features while different devices are employed to make them expressive and striking.

复合图书馆原型——HeadLine

复合图书馆是近几年出现的一个新概念。HeadLine是一个复合图书馆原型,本文对其进行了简单介绍。

Study on Translation of English Financial News Headlines from the Perspective of Pragmatics

English financial news headlines are distinguished from other types of titles for its distinctive characteristics.Besidesthat,when it comes to the problem of different thinking modes,the core of the translation is to keep the pragmatic equivalence forthe sake of achieving effective communications in cross-cultural background.Based on analysis of the features of English financialnews headlines,this paper aims at searching for better translation strategies for these from the perspective of pragmatics.

An Analysis of Lexical Features of English News Headlines and Their Translation Skills

A headline is an inseparable component of news, which serves the role of eyes. So wring a brief, accurate and eye-catching headline is very important. With plenty of practical examples given, the paper discusses lexical features of English news headlines through different aspects, including the use of nouns and midget words, vogue words, derivatives, nonce words, loans and so on. It also introduces some translating strategies according to their distinctive lexical characteristics.

Analysis of Ambiguity in Newspaper Headlines from a Cognitive Perspective

Ambiguity in newspaper headlines have been studied by many people, which mostly focuses upon the traditional linguistics perspective: lexical, grammatical and syntactical. In this paper, the author intends to study the ambiguity in newspaper headlines from the cognitive perspective in order to find how people perceive the ambiguous headlines. To support the analysis, many examples of ambiguous headlines are collected to investigate into way people perceive them. Finally, the author concluded that in the cognitive perspective, people are prone to accept one of the two or more interpretations about the ambiguous headlines according to the following factors: their back ground knowledge, the common sense, the social-cultural environment as well as their personal altitudes, which are drawn upon from the image schema theory.

Analysis on Features of News Headlines Translation

This paper is holding the purpose of summing up the theories and methods based on many examples which demonstrate principles,features and technique skills in news headline translation.We know that the translation of news headlines which is considered as the eyes and window of news appears very essential as many people do not have enough time to read the whole text from beginning to the end.They tend to browse headlines at first and then pick up the news in which the headline attracts them.From the paper,we can draw a conclusion that English news headline translation is not an easy task,and we need more study and practice to make it better and do our utmost to help Chinese readers when they are reading English newspapers or magazines.

A Study of Rhetorical Features of English News Headlines

As is known to all,headlines serve as a summary and advertisement for the main idea of the news so as to make the reader have a better understanding.Therefore,to some degree,we can say that the headline is the soul of a news report.Rhetoric,in all kinds of writing in all kinds of languages,can be applied and make an efficient effect.As for an English news headline,it is usually formed with certain rhetorical device to embody its soul and to appeal to more readers.This paper is to introduce and discuss the rhetorical features of the news headlines with examples and analysis.

Roman Urdu News Headline Classification Empowered with Machine Learning

Roman Urdu has been used for text messaging over the Internet for years especially in Indo-Pak Subcontinent.Persons from the subcontinent may speak the same Urdu language but they might be using different scripts for writing.The communication using the Roman characters,which are used in the script of Urdu language on social media,is now considered the most typical standard of communication in an Indian landmass that makes it an expensive information supply.English Text classification is a solved problem but there have been only a few efforts to examine the rich information supply of Roman Urdu in the past.This is due to the numerous complexities involved in the processing of Roman Urdu data.The complexities associated with Roman Urdu include the non-availability of the tagged corpus,lack of a set of rules,and lack of standardized spellings.A large amount of Roman Urdu news data is available on mainstream news websites and social media websites like Facebook,Twitter but meaningful information can only be extracted if data is in a structured format.We have developed a Roman Urdu news headline classifier,which will help to classify news into relevant categories on which further analysis and modeling can be done.The author of this research aims to develop the Roman Urdu news classifier,which will classify the news into five categories(health,business,technology,sports,international).First,we will develop the news dataset using scraping tools and then after preprocessing,we will compare the results of different machine learning algorithms like Logistic Regression(LR),Multinomial Naïve Bayes(MNB),Long short term memory(LSTM),and Convolutional Neural Network(CNN).After this,we will use a phonetic algorithm to control lexical variation and test news from different websites.The preliminary results suggest that a more accurate classification can be accomplished by monitoring noise inside data and by classifying the news.After applying above mentioned different machine learning algorithms,results have shown that Multinomial Naïve Bayes classifier is giving the best accuracy of 90.17%which is due to the noise lexical variation.

A Study of Stylistic Features of English News Headlines

From different aspects we can draw out different characteristics of news headlines. According to various study and research from different authors, we can conclude their characteristics are displayed on syntactical level, punctuation level, lexical level, rhetorical level and cultural level.

An Analysis of English News Headlines

A headline is an indispensable component of a news report and is the eye of the pages on the newspapers,through which the readers can gain as much information as possible in the newspaper within the shortest time and search for clues to stories that interest them,as the headline is the condensation of the whole story.Despite their own characteristic ways of editing headlines,English newspapers share something in common,namely brevity,accuracy,clarity and eye-catchingness.However,these features unique to English newspaper headlines constitute a big barrier to readers' understanding of the headlines and they may become quite at loss when reading headlines at the very beginning.This thesis aims to make an analysis of English news headlines in terms of their features at structural,lexical and rhetorical levels and hopes to help readers understand English news headline better.

A Study on the Ambiguity in English News Headlines from the Perspective of Cooperative Principle

In recent years,some news headlines do not focus on the contents of news,but exaggerate the fact.The main reason for this phenomenon is that the news editors disobey Cooperative Principle.As a result,it causes two different effects:one is negative;the other is positive.So the paper will find out the reasons of the ambiguity by using examples appearing in different newspapers and give a detailed analysis from the perspective of Cooperative Principle.

A Cognitive Study on Conceptual Metaphors in English News Headlines

The headline plays an important role in news report, which is a guide for readers to have a quick understanding the information of news. The headline is the"eye"of the news and it often has a lot of metaphors. The traditional theory of metaphor holds that as a rhetorical means, metaphor is mainly employed in literary works, while the theory of cognitive metaphor holds that metaphor is pervasive in our thought, action and language. The paper will discuss and analyze the characteristics of metaphors in headlines by analyzing a large number of English news examples. From the perspective of cognitive linguistics, this paper attempts to help readers to better understand the use of metaphors in headlines.

Theme-rheme Analysis of English News Headlines

As a special type of writing,news report has attracted more and more attention in different countries.Many people have developed an interest in reading English news,which provides them with timely information about important happenings around the world.Under the frame work of Systemic-Functional Grammar,based on English news headlines,the theme-rheme structure of English headlines is systemically analyzed in this thesis.Some detailed researches are made to investigate the influence of the Theme-rheme Structure in English news headlines.Then the news headline as well as its contents could be understood and enjoyed while reading newspaper.

Translation of News Headlines Under Contextual Theory

With the coming of all kinds of information,the world is becoming one indivisible planet.People not only pay attention to major events in their own countries,but also gradually pay attention to international news and learn about the world situation.As one of the world’s common languages,English news accounts for a large proportion of the news market.Usually when people choose to read current events,they will look at the headlines first,and then at the content of the article.Therefore,whether the translated news headlines can accurately convey information and catch readers’attention in a short time plays a very important role in the dissemination of news.Based on the context theory,this paper studies and analyzes the translation of some news headlines,expecting to provide some suggestions for news translation.

Studyon the E-C Translation of Financial NewsHeadlines inthe Context of the Skopos Theory

As the heart of news,financial news headline determines whetherpeople are interested in the content.The purpose of translating financial news headlines is to accurately convey the financial information of foreign media to domestic readers and attract domestic readers’attention by means of rhetoric,typesetting,and so on.This paper mainly discusses how the Skopos Theory is applied in translating financial news headlines through case studies.In the context of Skopos Theory,translation principles for different financial news headlines will also be explored and summarized.

Sub-Saharan Migrants in Moroccan Print Media:Discourses of Violence,Security,and Illegality before and after the Migration Reform

Morocco currently occupies a curious position as both a country of transit and a country of installation for migrants originating from Sub-Saharan countries.Even though the new migration policy reform has provided migrants with some new rights,public opinion is still unready to shape and reshape their perceptions of’black’African migrants.The situation of those migrants has grabbed media attention,especially in newspapers;hundreds of articles and reports have been written in newspapers,criticising or denouncing the phenomenon.Although the currently available body research has tended to neglect the linguistic ways in which such media convey meaning along the side of their use of images,little attention has been paid to language-based media.As such,this paper scrutinises(i)the depiction of Sub-Saharan migrants in Moroccan print newspapers headlines before and after the migration reform,and(ii)it considers how reports on‘black’African migrants in Morocco may lead to propagating stereotypical acts and xenophobia towards this category.The study examines the topics covered and reveals the patterns used in the coverage of related topics,such as illegality and violence,before and after the new migration policy in 2013.With these objectives in mind,this study is based on qualitative data of news items and reports from the two daily printed newspapers,Assabah and Almassae.Relying on Critical Discourse Analysis and Content Analysis has demonstrated an overgeneralisation of Sub-Saharan migrants in news headlines in general.The results also show a dramatic change in the thematic coverage during the selected period.

Recent Advances on Neural Headline Generation

Recently, neural models have been proposed for headline generation by learning to map documents to headlines with recurrent neural network. In this work, we give a detailed introduction and comparison of existing work and recent improvements in neural headline generation, with particular attention on how encoders, decoders and neural model training strategies alter the overall performance of the headline generation system. Furthermore, we perform quantitative analysis of most existing neural headline generation systems and summarize several key factors that impact the performance of headline generation systems. Meanwhile, we carry on detailed error analysis to typical neural headline generation systems in order to gain more comprehension. Our results and conclusions are hoped to benefit future research studies.

Translation Strategies of Sports Headlines from English to Chinese

Headlines are indispensable components in news reports. The translation of headlines is a vital to the whole. However, so far it failed to attract enough attention of domestic translators. Based on English sports news examples, this paper probes into the profiles of English sports headlines including the definition, characteristics and function. In particular, the paper proposes some approaches to proper translation of English sports headlines into Chinese. Translators should employ different strategies under specific circumstances.

Shawn Mendes

Shawn Mendes is a Canadian singer and songwriter.He gained a following in 2013,posting song covers on the video-sharing application Vine.The following year,he caught the attention of artist manager Andrew Gertler and Island Records A&R Ziggy Chareton,which led to him signing a deal with the record label.He has since released three studio albums,headlined three world tours,and received several nominations and awards.