Opinion mining, context analysis, and global sustainable waste management

Abstract

Given the ever growing abundance of data now available in the world, we still know surprisingly little about some waste management processes and actions. Too many countries still have poor data or get it too late, while the data that exist fail to cover some issues in sufficient detail. Google data can help decision-makers overcome the data shortage and completeness problems they face working with traditional sources of waste management data. The research uses publicly accessible global and country-level traditional and Google data, as well as a 988-keyword waste management dictionary. We are developing 988 WASTENN neural networks and 988 WASTERE linear regression models to estimate the use of sustainable waste management (SWAM) keywords globally. The models can be used to predict SWAM trends in efforts to minimize waste. Our tools’ main advantage is mining accessible stakeholders’ opinions of ’big data’. On average, the developed WASTERE and WASTENN models have sufficient reliability (R2aver = 0. 636–0.991). The SWAM map and models of the world developed in this study analyze the links between SWAM and its macro-environment to provide evidence-based digital recommendations and interventions that will help to improve SWAM. We analyzed 85 countries, and our findings, based on this neural networks, statistical and multi-criteria analysis, suggest that a state’s sustainability, productivity, context, and waste management performance are interconnected: as the social, environmental, political/governance, cultural, and economic (SEPCE) indicators of a country improve, its waste metrics also improve and there is more analysis of waste questions. The article proposes evidence-based policy recommendations for SEPCE incentives.