Social & Governance

Machine learning can be used in a variety of sectors including government, law enforcement, non-profit groups, and commercial businesses for social, ethical, and legal reasons. Companies like Amphos21, an environmental consultancy firm that specializes in the management of water resources, have benefited from using SpeedWise® Machine Learning (SML) to unveil hidden trends and patterns from data. Amphos21 was able to predict the evolution of water quality, earning an award for their efforts to come up with solutions to water management challenges.

Use Cases

Pollutants in Water

Using SML, they were able to build an advanced machine learning model that could be used to predict contamination in groundwater and surface waters. Normally, the process of testing the water for high levels of pollution is very time consuming, where water samples are manually analyzed in a lab. Building a predictive model using existing data with SML, Amphos 21 created a solution that can quickly identify pollution in water to help mitigate the risk of contaminated water infiltrating a wastewater treatment facility. As a result of winning first prize for their innovative and efficient solution, Amphos 21 will be awarded an opportunity to do a pilot test on the ground with the Catalan Water Agency.

Planning and Policymaking

Machine learning can be used to help agencies with planning and policies. Data from different sources can be used to better understand public opinion. For example, predicting how many new students will sign up for school next year can be done with data sources such as homes purchased in the school district area or demographics and population of new residents. Forecasting new students can further assist schools to better plan for resources and supplies.

Crime Analysis

Machine learning can help law enforcement better prepare and take protective measures. SML algorithms can be used to predict the location of where crime may occur. Historical data that includes features such as education, poverty, employment, time, and location can help train the model to more accurately forecast crime in an area.

Real Estate Home Value Appraisal

Machine learning can help agents accurately forecast home appraisal value. Typically, an agent or analyst will analyze key data points such as location, square footage, home upgrades and more for each home. This can be a time consuming process to do for each home individually. SML can be utilized by agents to quickly and accurately forecast home appraisal values for thousands of homes based on numerous variables.