SafeCity: Understanding Diverse Forms of Sexual Harassment Personal Stories

Karlekar, Sweta; Bansal, Mohit

doi:10.18653/v1/d18-1303

Cited by 32 publications

(38 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our models can automatically extract the key elements from the sexual harassment stories and at the same time categorize the stories in different dimensions. The proposed models outperformed the single task models, and achieved higher than previously reported accuracy in classifications of harassment forms (Karlekar and Bansal, 2018).…”

Section: Introductionmentioning

confidence: 62%

“…There are a very limited number of studies on sexual harassment stories shared online. Karlekar and Bansal (2018) were the first group to our knowledge that applied NLP to analyze large amount ( ∼10,000) of sexual harassment stories. Although their CNN-RNN classification models demonstrated high performance on classifying the forms of harassment, only the top 3 majority forms were studied.…”

Section: Related Workmentioning

confidence: 99%

“…Therefore, the information about perpetrators needs to be extracted as well as the location and time of events. Karlekar and Bansal (2018) applied several visualization techniques in order to capture such information, but it was not obtained explicitly. Our preliminary research demonstrated automatic extraction of key element and story classification in separate steps (Liu et al, 2019).…”

Section: Related Workmentioning

confidence: 99%

“…commenting, ogling and groping. A dataset of these stories was published by Karlekar and Bansal (2018). In addition to the forms of harassment, we manually annotated each story with the key elements (i.e.…”

Section: Data Collection and Annotationmentioning

confidence: 99%

“…Data Splits: We used the same splits of train, develop, and test sets used by Karlekar and Bansal (Karlekar and Bansal, 2018), with 7201, 990 and 1701 stories, respectively. In this study, we only considered single label classifications.…”

Section: Experimental Settingsmentioning

confidence: 99%

See 4 more Smart Citations

Uncover Sexual Harassment Patterns from Personal Stories by Joint Key Element Extraction and Categorization

Liu

Cifor

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

The number of personal stories about sexual harassment shared online has increased exponentially in recent years. This is in part inspired by the #MeToo and #TimesUp movements. Safecity is an online forum for people who experienced or witnessed sexual harassment to share their personal experiences. It has collected >10,000 stories so far. Sexual harassment occurred in a variety of situations, and categorization of the stories and extraction of their key elements will provide great help for the related parties to understand and address sexual harassment. In this study, we manually annotated those stories with labels in the dimensions of location, time, and harassers' characteristics, and marked the key elements related to these dimensions. Furthermore, we applied natural language processing technologies with joint learning schemes to automatically categorize these stories in those dimensions and extract key elements at the same time. We also uncovered significant patterns from the categorized sexual harassment stories. We believe our annotated data set, proposed algorithms, and analysis will help people who have been harassed, authorities, researchers and other related parties in various ways, such as automatically filling reports, enlightening the public in order to prevent future harassment, and enabling more effective, faster action to be taken.

show abstract

Section: Introductionmentioning

confidence: 62%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Data Collection and Annotationmentioning

confidence: 99%

Section: Experimental Settingsmentioning

confidence: 99%

See 3 more Smart Citations

Uncover Sexual Harassment Patterns from Personal Stories by Joint Key Element Extraction and Categorization

Liu

Cifor

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

show abstract

Space identification of sexual harassment reports with text mining

Karami

Swan

Moraes

2020

Proc Assoc Inf Sci Technol

View full text Add to dashboard Cite

Sexual harassment is an invisible problem that has been difficult to combat because victims are often reluctant to report. However, within the past years, the sheer volume of women who have spoken up about sexual harassment has brought the issue to the forefront. This change has been largely driven, in part, by Internet and social media technologies. Given the large size of data posted on these online technologies, it is impossible to manually analyze and organize it; therefore, there is a need to utilize data and text mining methods. In order to help the fight against sexual harassment, this study proposes a predictive framework to collect more than 14,000 sexual harassment reports on the everyday sexism project (ESP) website and identify the space (location) in the reports. Our framework achieves 85.33% accuracy for seven space classes including workplace, public space, home, public transport, school, university, and media. This paper also enriches experiments by merging similar classes (e.g., school and university) and applies a feature selection method to reduce the number of features for efficiency and effectiveness purposes. This enrichment process offers promising results for different sets of classes and features, ranging from 86% – 93% accuracy.

show abstract

#MeTooMaastricht: Building a Chatbot to Assist Survivors of Sexual Harassment

Devrim

Glazunov

Jaramillo

et al. 2020

Machine Learning and Knowledge Discovery in Databases

View full text Add to dashboard Cite

Inspired by the recent social movement of #MeToo, we are building a chatbot to assist survivors of sexual harassment cases (designed for the city of Maastricht but can easily be extended). The motivation behind this work is twofold: properly assist survivors of such events by directing them to appropriate institutions that can offer them help and increase the incident documentation so as to gather more data about harassment cases which are currently under reported. We break down the problem into three data science/machine learning components: harassment type identification (treated as a classification problem), spatiotemporal information extraction (treated as Named Entity Recognition problem) and dialogue with the users (treated as a slot-filling based chatbot). We are able to achieve a success rate of more than 98% for the identification of a harassment-or-not case and around 80% for the specific type harassement identification. Locations and dates are identified with more than 90% accuracy and time occurences prove more challenging with almost 80%. Finally, initial validation of the chatbot shows great potential for the further development and deployment of such a beneficial for the whole society tool.

show abstract

SafeCity: Understanding Diverse Forms of Sexual Harassment Personal Stories

Cited by 32 publications

References 22 publications

Uncover Sexual Harassment Patterns from Personal Stories by Joint Key Element Extraction and Categorization

Uncover Sexual Harassment Patterns from Personal Stories by Joint Key Element Extraction and Categorization

Space identification of sexual harassment reports with text mining

#MeTooMaastricht: Building a Chatbot to Assist Survivors of Sexual Harassment

Contact Info

Product

Resources

About