The human right to water and sanitation: Using natural language processing to uncover patterns in academic publishing

Abstract

After years of advocacy and international negotiation, the General Assembly of the United Nations voted to officially recognize a stand-alone human right to water and sanitation on 28 July 2010. Since, academic scholarship has continued to grow in an effort to understand the implications of the codification of this human right. Yet, with this growth, it has become impractical if not impossible for scholars to keep up with the advancement of academic knowledge or to make sense of it in a systematic way. In short, to date, we know very little about the trends in the literature as they have unfolded over the past thirty years and the topics to which scholars have devoted significant attention within the broader field, particularly over time. This is an important area of inquiry, as developing a comprehensive understanding of where prior literature has focused and where it appears to be going offers scholars an opportunity to identify areas in need of refinement and/or increased attention. Given the practicalities of reading thousands of research papers each year, this project utilizes natural language processing (NLP) to identify topics and trends in academic literature on the human right to water and sanitation (HRtWS). NLP provides the opportunity to digest large quantities of text data through machine learning, culminating with descriptive information on trends and topics in the field since 1990. The results of this exercise show that the research related to the human right to water and sanitation has grown exponentially, particularly over the last decade, illustrates the multidisciplinary nature of the literature, and demonstrates the diversity of topics in the field.