Untitled

Abstract

Language is gendered if the context surrounding a mention is suggestive of a particular binary gender for that mention. Detecting the different ways in which language is gendered is an important task since gendered language can bias NLP models (such as for coreference resolution). This task is challenging since genderedness is often expressed in subtle ways. Existing approaches need considerable annotation efforts for each language, domain, and author, and often require handcrafted lexicons and features. Additionally, these approaches do not provide a quantifiable measure of how gendered the text is, nor are they applicable at the fine-grained mention level. In this paper, we use existing NLP pipelines to automatically annotate gender of mentions in the text. On corpora labeled using this method, we train a supervised classifier to predict the gender of any mention from its context and evaluate it on unseen text. The model confidence for a mention's gender can be used as a proxy to indicate the level of genderedness of the context. We test this gendered language detector on movie summaries, movie reviews, news articles, and fiction novels, achieving an AUC-ROC of up to 0.71, and observe that the model predictions agree with human judgments collected for this task. We also provide examples of detected gendered sentences from aforementioned domains.

Analyzing Gender Bias within Narrative Tropes

“…While similar to methods used in prior work (García et al, 2014), our genderedness score is limited by its lexicon and susceptible to gender generalization and explicit marking (Hitti et al, 2019). We leave exploration of more nuanced methods of capturing trope genderedness (Ananya et al, 2019) to future work.…”

“…Our work builds on computational research analyzing gender bias. Methods to measure gender bias include using contextual cues to develop probabilistic estimates (Ananya et al, 2019), and using gender directions in word embedding spaces (Bolukbasi et al, 2016). Other work engages directly with tvtropes.org: Kiesel and Grimnes (2010) build a wrapper for the website, but perform no analysis of its content.…”

“…Our work builds on computational research analyzing gender bias. Methods to measure gender bias include using contextual cues to develop probabilistic estimates (Ananya et al, 2019), and using gender directions in word embedding spaces (Bolukbasi et al, 2016) Analyzing bias through tropes is a popular area of research within social science. Hansen (2018) focus in on the titular princess character in the video game The Legend of Zelda as an example of the Damsel in Distress trope.…”

Continuous User Authentication on Multiple Smart Devices

“…Smart devices are typically equipped with built-in sensors [8], touch screens [9], keyboard interfaces [10], and other accessories [11] that can be used to capture behavioral data. Continuous authentication systems essentially employ behavioral data originating from the user interaction with the mobile device to extract invariant features of user behavior.…”

Study of Karyotypes in Patients of Breast Carcinoma in Gujarat State

“…Earlier, however, it was a matter of taboo and embarrassment that meant detection and diagnosis was rare. In the 1990"s the symbol of breast cancer -the pink ribbonbrought out a revolution against this cancer [5].…”