Gender Bias in the News: A Scalable Topic Modelling and Visualization Framework

Abstract

We present a topic modelling and data visualization methodology to examine gender-based disparities in news articles by topic. Existing research in topic modelling is largely focused on the text mining of closed corpora, i.e., those that include a fixed collection of composite texts. We showcase a methodology to discover topics via Latent Dirichlet Allocation, which can reliably produce human-interpretable topics over an open news corpus that continually grows with time. Our system generates topics, or distributions of keywords, for news articles on a monthly basis, to consistently detect key events and trends aligned with events in the real world. Findings from 2 years worth of news articles in mainstream English-language Canadian media indicate that certain topics feature either women or men more prominently and exhibit different types of language. Perhaps unsurprisingly, topics such as lifestyle, entertainment, and healthcare tend to be prominent in articles that quote more women than men. Topics such as sports, politics, and business are characteristic of articles that quote more men than women. The data shows a self-reinforcing gendered division of duties and representation in society. Quoting female sources more frequently in a caregiving role and quoting male sources more frequently in political and business roles enshrines women’s status as caregivers and men’s status as leaders and breadwinners. Our results can help journalists and policy makers better understand the unequal gender representation of those quoted in the news and facilitate news organizations’ efforts to achieve gender parity in their sources. The proposed methodology is robust, reproducible, and scalable to very large corpora, and can be used for similar studies involving unsupervised topic modelling and language analyses.

Cobertura periodística de las candidatas a la gubernatura durante las campañas de México en 2021

“…En ese sentido, las mujeres son más citadas en noticias sobre salud, estilo de vida y temas feministas, mientras que los hombres tienen carta abierta para hablar sobre cualquier tipo de tema, generalmente sobre política y negocios (Kassova, 2020). De esta manera, las voces de las mujeres quedan relegadas a cuestiones derivadas del ámbito doméstico, mientras que los hombres tratan de asuntos del ámbito público (Rao y Taboada, 2021). Así, se ha documentado que, cuando los medios necesitan de la intervención de un político para que contextualice la noticia, prefieren contactar a un político hombre que a una mujer (Sjøvaag et al, 2019).…”

“…En cuanto al tercer comportamiento registrado, el cuerpo de las noticias analizadas presentó menos declaraciones de las mujeres en forma de citas directas que de los hombres, coincidiendo con los hallazgos de trabajos anteriores (Asr et al, 2021;Humprecht y Esser, 2017;Kassova, 2020;Rao y Taboada, 2021;Rodelo, 2020a;Sjøvaag et al, 2019;Van der Pas y Aaldering, 2020;Vidal, 2020b;Vogler y Schwaiger, 2021). Lo problemático de esta cuestión es que las posturas y propuestas de campaña de las candidatas no son presentadas a partir de las mismas palabras en que fueron expresadas por ellas (Humprecht y Esser, 2017).…”

Text Adversarial Attacks and Defenses: Issues, Taxonomy, and Perspectives

“…erefore, we can focus on more securityrelated scenarios [167], such as language style forgery, fake news recognition, malicious code, and other related areas of research. We can also consider studying multi-lingual adversarial instances in the context of the ethnicity of the researchers, as well as studying textual adversarial applications in more practical scenarios such as mental health [168,169], android applications [170][171][172][173], Ethereum smart contract [174], network traffic [175], and discrimination [176].…”

MMSS: A storytelling simulation software to mitigate misinformation on social media

“…These tools took advantage of the large amount of data generated on the social networks every second. A non-exclusive list of applications is health monitoring [3], organized crime [4], hate speech [5], and gender bias [6]. On the other hand, simulation framework designs were introduced for crisis management as well.…”