Using Natural Language Processing and Qualitative Analysis to Intervene in Gang Violence: A Collaboration Between Social Work Researchers and Data

📝 Abstract

The U.S. has the highest rate of firearm-related deaths when compared to other industrialized countries. Violence particularly affects low-income, urban neighborhoods in cities like Chicago, which saw a 40% increase in firearm violence from 2014 to 2015 to more than 3,000 shooting victims. While recent studies have found that urban, gang-involved individuals curate a unique and complex communication style within and between social media platforms, organizations focused on reducing gang violence are struggling to keep up with the growing complexity of social media platforms and the sheer volume of data they present. In this paper, describe the Digital Urban Violence Analysis Approach (DUVVA), a collaborative qualitative analysis method used in a collaboration between data scientists and social work researchers to develop a suite of systems for decoding the high- stress language of urban, gang-involved youth. Our approach leverages principles of grounded theory when analyzing approximately 800 tweets posted by Chicago gang members and participation of youth from Chicago neighborhoods to create a language resource for natural language processing (NLP) methods. In uncovering the unique language and communication style, we developed automated tools with the potential to detect aggressive language on social media and aid individuals and groups in performing violence prevention and interruption.

💡 Analysis

The U.S. has the highest rate of firearm-related deaths when compared to other industrialized countries. Violence particularly affects low-income, urban neighborhoods in cities like Chicago, which saw a 40% increase in firearm violence from 2014 to 2015 to more than 3,000 shooting victims. While recent studies have found that urban, gang-involved individuals curate a unique and complex communication style within and between social media platforms, organizations focused on reducing gang violence are struggling to keep up with the growing complexity of social media platforms and the sheer volume of data they present. In this paper, describe the Digital Urban Violence Analysis Approach (DUVVA), a collaborative qualitative analysis method used in a collaboration between data scientists and social work researchers to develop a suite of systems for decoding the high- stress language of urban, gang-involved youth. Our approach leverages principles of grounded theory when analyzing approximately 800 tweets posted by Chicago gang members and participation of youth from Chicago neighborhoods to create a language resource for natural language processing (NLP) methods. In uncovering the unique language and communication style, we developed automated tools with the potential to detect aggressive language on social media and aid individuals and groups in performing violence prevention and interruption.

📄 Content

1 Using Natural Language Processing and Qualitative Analysis to Intervene in Gang Violence: A Collaboration Between a Social Work Researchers and Data Scientists

Desmond Upton Patton Kathleen McKeown Owen Rambow Columbia University Columbia University Columbia University New York, NY, USA New York, NY, USA New York, NY, USA dp2787@columbia.edu Kathy@cs.columbia.edu Ocr2101@columbia.edu

Jamie Macbeth Fairfield University Fairfield, CT, USA Jamie.Macbeth@gmail.com

ABSTRACT The U.S. has the highest rate of firearm-related deaths when compared to other industrialized countries. Violence particularly affects low- income, urban neighborhoods in cities like Chicago, which saw a 40% increase in firearm violence from 2014 to 2015 to more than 3,000 shooting victims. While recent studies have found that urban, gang-involved individuals curate a unique and complex communication style within and between social media platforms, organizations focused on reducing gang violence are struggling to keep up with the growing complexity of social media platforms and the sheer volume of data they present. In this paper, describe the Digital Urban Violence Analysis Approach (DUVVA), a collaborative qualitative analysis method used in a collaboration between data scientists and social work researchers to develop a suite of systems for decoding the high- stress
language of urban, gang-involved youth.
Our approach leverages principles of grounded theory when analyzing approximately 800 tweets posted by Chicago gang members and participation of youth from Chicago neighborhoods to create a language resource for natural language processing (NLP) methods. In uncovering the unique language and communication style, we developed automated tools with the potential to detect aggressive language on social media and aid individuals and groups in performing violence prevention and interruption. Keywords: Social Media, Gang Violence, Qualitative Methods, Natural Language Processing

1. INTRODUCTION In this paper, we describe the Digital Urban Violence Analysis approach (DUVVA), a collaborative qualitative analysis method used to inform natural language processing techniques that predict clusters of aggressive language on Twitter that may escalate into fatal and non-fatal firearm violence. Recently, several studies (Patton et al 2015; Patton et al. 2016a; Patton et al. 2016b) examined the relationship between social media and gang violence. These study found that young people living in violent, urban neighborhood taunt each other, make threats and brag about violence on social media platforms in ways that may lead to firearm violence, a behavior known as Internet banging or cyberbanging (Patton, Eschmann, Butler, 2013). These studies used in-depth, qualitative methods that include coding each post by hand to code text from known gang-involved youth in Bloomberg Data for Good Exchange Conference. 25-Sep-2016, New York City, NY, USA.

2 Chicago. However, this method alone is inefficient when analyzing massive amounts of social media data that requires an in-depth understanding of community context and language. We aim to develop a more efficient, automatic coding process using natural language processing (NLP) tools informed by in-depth qualitative insights of offline conditions and mechanisms that shape urban gang-related violence. The development of the NLP tool could be particularly helpful for violence prevention organizations that use social media as a part of their violence intervention methods. One study of Chicago gang communication on Twitter found that urban, gang-involved individuals curate a unique, complex communication style within and between social media platforms that warrant careful interpretation. Many studies of Twitter are “big data” studies that employ quantitative computer- based analysis of tens of thousands or even millions of data points. However, there are many obstacles to collecting quantitative data of gang- related behavior on Twitter. Identifying hashtags or keywords is difficult given the diverse kinds of users and communication patterns and styles that vary by city, neighborhood and between gangs, crews, and cliques. Furthermore, the linguistic style and extensive use of emoji’s and other images among gang-affiliated youth renders most quantitative tools such as scripts or emotion detection software inadequate. There is a need to understand more accurately how social media reflects the lived reality of marginalized young people who live in low-income, violent urban neighborhoods. To achieve this goal, we must first understand how language is used to communicate offline identities, networks and exposure to violence and trauma

This content is AI-processed based on ArXiv data.