RavenPack Launches 'Alternative to Polling'

New York-based big data analytics provider RavenPack has launched a platform it describes as an 'alternative to polling' - offering projections and analysis for the forthcoming US presidential election.

The new approach uses sentiment analysis measured by thousands of news stories about each candidate's policies and personal life; historical voting patterns; and media attention received by each candidate, to forecast election results. RavenPack claims the model, built by its Data Science team, has correctly predicted the winning candidate in four out of the five last US presidential elections, with a confidence of greater than 75%.

Election data can be downloaded at no cost, and users can embed the full dashboard or widgets on their web site, and subscribe to regular insights on the US election. CEO Armando Gonzalez (pictured) comments: 'Our news-driven methodology offers an alternative angle to traditional forecasting approaches such as polls or surveys and summarizes the complexities of the current political and socio-economic environment in the United States. We thought the US presidential election was an ideal opportunity to showcase the power of alternative data'.

Web site: www.ravenpack.com .

All articles 2006-23 written and edited by Mel Crowther and/or Nick Thomas, 2024- by Nick Thomas, unless otherwise stated.