Deep Level Analysis of Legitimacy in Bengali News Sentences

The tremendous increase in the growth of misinformation in news articles has the potential threat for the adverse effects on society. Hence, the detection of misinformation in news data has become an appealing research area. The task of annotating and detecting distorted news article sentences is the immediate need in this research direction. Therefore, an attempt has been made to formulate the legitimacy annotation guideline followed by annotation and detection of the legitimacy in Bengali e-papers. The sentence-level manual annotation of Bengali news has been carried out in two levels, namely “Level-1 Shallow Level Classification” and “Level-2 Deep Level Classification” based on semantic properties of Bengali sentences. The tagging of 1,300 anonymous Bengali e-paper sentences has been done using the formulated guideline-based tags for both levels. The validation of the annotation guideline has been done by applying benchmark supervised machine learning algorithms using the lexical feature, syntactic feature, domain-specific feature, and Level-2 specific feature in both levels. Performance evaluation of these classifiers is done in terms of Accuracy, Precision, Recall, and F-Measure. In both levels, Support Vector Machine outperforms other benchmark classifiers with an accuracy of 72% and 65% in Level-1 and Level-2, respectively.

Corruption in the "First Order": Predicting the influence of scandal on voting behaviour and media perceptions during a General Election

<p><b>This thesis investigates responses in voting behaviour and media perceptions to the presence of media scandals about politicians and associated political parties during the 2017 New Zealand general election. A repeated measures design was used wherein 351 participants were recruited before the start of the election campaign, primarily from an Introductory Psychology course at Victoria University of Wellington. Follow-up surveys were conducted at three time points throughout the two month campaign. Participants were randomly allocated into one of two conditions for each follow-up survey. Half the participants were given a real news article to read about a media scandal, the other half read an article about a policy platform by the same political party. At the end of the election campaign, participants were asked about their voting behaviours. A second study was conducted after Labour Party leader, Jacinda Ardern, was announced Prime Minister with participants recruited via social media sites ‘Twitter’ and ‘Facebook’. In this study, 153 participants recalled information about scandals that were present in the media during the election campaign.</b></p> <p>Results showed that political scandals in news media do have an influence on voter perceptions, but not in an easily predictable way. Prior perceptions of political parties were the best predictors of who participants intended to vote for. Participants responded most strongly to public policy articles rather than scandal information, particularly those more knowledgeable of New Zealand’s political system, and therefore likely more engaged with politics in general. Further, there was evidence that information presented in the media influenced how participants viewed political parties that were not involved in the scandal, which is an important under a proportional voting system like MMP which requires understanding of the relationships between parties.</p> <p>Evidence was also found for a backlash effect towards the media wherein participants who were exposed to scandal information would displayed a decrease in trust towards the general media, consistent with the idea that one reason why voters may not respond negatively to scandal information reflects the decision that the source of the information is not credible. Future research should consider more targeted analysis on the different sources of news media, especially new media like blogs, social media, and entertainment news.</p>

Corruption in the "First Order": Predicting the influence of scandal on voting behaviour and media perceptions during a General Election

<p><b>This thesis investigates responses in voting behaviour and media perceptions to the presence of media scandals about politicians and associated political parties during the 2017 New Zealand general election. A repeated measures design was used wherein 351 participants were recruited before the start of the election campaign, primarily from an Introductory Psychology course at Victoria University of Wellington. Follow-up surveys were conducted at three time points throughout the two month campaign. Participants were randomly allocated into one of two conditions for each follow-up survey. Half the participants were given a real news article to read about a media scandal, the other half read an article about a policy platform by the same political party. At the end of the election campaign, participants were asked about their voting behaviours. A second study was conducted after Labour Party leader, Jacinda Ardern, was announced Prime Minister with participants recruited via social media sites ‘Twitter’ and ‘Facebook’. In this study, 153 participants recalled information about scandals that were present in the media during the election campaign.</b></p> <p>Results showed that political scandals in news media do have an influence on voter perceptions, but not in an easily predictable way. Prior perceptions of political parties were the best predictors of who participants intended to vote for. Participants responded most strongly to public policy articles rather than scandal information, particularly those more knowledgeable of New Zealand’s political system, and therefore likely more engaged with politics in general. Further, there was evidence that information presented in the media influenced how participants viewed political parties that were not involved in the scandal, which is an important under a proportional voting system like MMP which requires understanding of the relationships between parties.</p> <p>Evidence was also found for a backlash effect towards the media wherein participants who were exposed to scandal information would displayed a decrease in trust towards the general media, consistent with the idea that one reason why voters may not respond negatively to scandal information reflects the decision that the source of the information is not credible. Future research should consider more targeted analysis on the different sources of news media, especially new media like blogs, social media, and entertainment news.</p>

Predicting the impact of online news articles – is information necessary?

We exploit the Twitter platform to create a dataset of news articles derived from tweets concerning COVID-19, and use the associated tweets to define a number of popularity measures. The focus on (potentially) biomedical news articles allows the quantity of biomedically valid information (as extracted by biomedical relation extraction) to be included in the list of explored features. Aside from forming part of a systematic correlation exploration, the features – ranging from the semantic relations through readability measures to the article’s digital content – are used within a number of machine learning classifier and regression algorithms. Unsurprisingly, the results support that for more complex articles (as determined by a readability measure) more sophisticated syntactic structure may be expected. A weak correlation is found with information within an article suggesting that other factors, such as numbers of videos, have a notable impact on the popularity of a news article. The best popularity prediction performance is obtained using a random forest machine learning algorithm, and the feature describing the quantity of biomedical information is in the top 3 most important features in almost a third of the experiments performed. Additionally, this feature is found to be more valuable than the widely used named entity recognition.

Mixed findings in directly replicated experimental studies on fake news

Fake news mimics the look of legitimate news articles even if it does not mimic the standards of journalistic reporting. An increase in fake news has developed along with heightened concern about the veracity of news information, which has been highly politicized as fake news. These problems suggest whether standards of journalistic reporting can overcome the mimicry of real news, and whether the public can correctly identify real news. Here we ask two research questions. Does source information about the news article or its presentation influence the perception that a news article is fake news? What factors influence the perception of fake news? We conducted directly replicated experimental studies that presented four news articles to four subject pools. We show that source information and presentation have limited influence on participants’ judgments of a real news article as fake. Among those who evaluated the articles as fake news, our results show that the less participants thought the article presented a fair, balanced, evidence-based view, the more likely they were to judge it as fake news. These findings warrant discussion about the purpose of news organizations and news reporting as well as about how evidence and fairness work in news information.

A Pragmatic Analysis on Vague Language Used in Disaster News Articles on Thejakartapost.com

This paper presents the findings analysis of categories and functions on vague language used in disaster news articles on Thejakartapost.com based on the theory of Channell (1994). In the journalism context, especially in disaster news article, the information often contains vague language that has imprecise statement since it is harmful. Therefore, to avoid wrong statements, the reporters often use vague language in presenting information accurately. The study employed a qualitative descriptive method. All data were 24 news articles. There were 12 news articles of natural disasters and 12 news articles of human-caused disasters. The period was from April 2018 until March 2019. The findings of this study showed that there were three categories of vague language, namely vague additives to numbers that were realized by approximators and adjectives. The others were vagueness of choice of vague words that were realized by nouns, and vagueness by scalar implicatures that were realized by quantifiers, numbers, and exaggerations. In addition, they also have its functions of vague language. Firstly, giving the right amount of information, it is used since the reporters just shared the right number of information although the exact number was not available. Secondly, filling in lexical gaps of uncertainty, it is used since the reporters wanted to cover the imprecise information with another word, and generalized word that was difficult to identify. Last but not least, self-protection. It is used since the reporters wanted to protect and hedge their statements from imprecise information.Keywords: Vague Language, Categories, News Articles, Disasters, implicature