Abstract

 The menstrual health of women serves as a critical marker of their overall health and quality of life. Considering the established link between irregular menstrual cycles, associated issues, and cardiovascular diseases, along with the potential implications for premature mortality, there is an urgent need to prioritize research efforts dedicated to assessing the impact of the COVID-19 pandemic on women's menstrual health. It's surprising to observe that the machine-learning community has largely overlooked this issue, which significantly impacts nearly half of the global population. The work presented in this paper utilizes change modeling to investigate the evolution of COVID-19-related menstrual health issues over time, as gleaned from tweets. Change modeling involves monitoring shifts in the acquired knowledge over time and creating innovative change measures or metrics to organize and potentially restructure this information. This approach to tracking changes offers a fresh, consolidated, and abstract perspective for examining the collected data, opening up new methods for analysis and inference. An additional valuable aspect of change modeling is its capacity to enhance the understanding of results, thereby increasing their practical utility. The outcomes of the framework presented here consist of innovative, straightforward, and easily understandable metrics capable of capturing changes from multiple perspectives. The primary benefit lies in the fact that these simple yet elegant change metrics simplify the decision-making process, providing a rapid and concise, one-glance overview of the metrics. The framework also serves as a foundation for generating only a limited selection of well-defined association links derived from discussions about menstruation on Twitter. This activity aligns with the goal of producing well-qualified, comprehensible, and contextually relevant knowledge, and provides advantages in terms of computational efficiency and storage space.

Statistics