New machine learning model measures social impact of large corporate investments
Companies are growing more focused on the social impact of business operations, and how certain decisions might be received by others, according to a recent Deloitte report. The professional services network created a new Social Impact Measurement Model (SIMM), which uses machine learning technology to predict possible outcomes of major corporate investments in a community.
If an organization opened a new office or headquarters, for example, the SIMM would create a forecast across 142 social measures to determine the social impact for the four years following the investment, the report noted. These social measures include education, housing, family and migration, income and employment, transportation, and industry factors.
SEE: IT leader's guide to deep learning (Tech Pro Research)
"With the rise of the social enterprise-those organizations looking beyond revenue and profit to understand their impact on society-many of our clients are raising the profile of purpose-driven outcomes," said Janet Foutty, chair and CEO of Deloitte Consulting LLP, in a Thursday press release. "The Social Impact Measurement Model enables our clients to understand if their investments will pay social dividends, providing value to companies, communities, and local governments."
The SIMM is also able to analyze past investments, up to the past five years, and study the effects of those decisions, the report said. Corporations previously limited decisions to factors such as job creation, output, and income figures; however, adding the social impact aspect could help businesses make more educated decisions around what is worth spending money on.
Machine learning is typically used in big organizations to learn more about their customers and to create better customer experiences. But SIMM takes that sentiment to the next level by creating a more cognizant approach to how certain decisions would affect people outside the company's clientele base and also producing a machine learning use case for businesses that might not think they are candidates for machine learning.