Will The Predictive Search Engine That Google And The CIA Funded Successfully Predict The Elections?

Companies have used search engines and social media data to follow trends and attempt to predict the future for years. Recorded Future has drawn the most attention amongst the many firms attempting to master predictive search because they received funding from two high-profile entities: Google and the CIA. Will the company live up to the hype?

We’ll all find out next month during the US elections.

Focusing on the Elections

A special section dedicated to the upcoming Senate elections is being closely watched by those interested in using this technology. Currently, the Boston-based firm has fewer than 100 customers according to founder Christopher Ahlberg. If his service can accurately predict the outcome of the various races, particularly the close ones, his customer base will likely increase dramatically.

The Senate Elections section focuses on a handful of key indicators and tracks polls and news for 19 of the most “interesting” elections. The algorithm attempts to gauge sentiment and momentum over the last month. Despite claims of real-time accuracy, the data available to the public appears to be about a week behind.

How it Works

By scouring news sources, blogs, and social media sites, Recorded Future analyzes data in real time. Every instance of data is weighed based upon several factors such as historical accuracy – a financial site whose articles tend to accurately predict future stock trends will have more “clout” than one that is often wrong.

In a white paper, Record Future described their technology as:

Recorded Future has developed an analytics engine, which goes beyond search, explicit link analysis and adds implicit link analysis, by looking at the “invisible links” between documents that talk about the same, or related, entities and events. We do this by separating the documents and their content from what they talk about – the “canonical” entities and events (yes, this model is heavily inspired by Plato and his distinction between the real world and the world of ideas).

“That makes it possible for me to look for specific patterns, like product releases expected from Apple in the near future, or to identify when a company plans to invest or expand into India,” says Ahlberg.

Molding the Future

Deciphering the mountains of data available on the Internet is the key to Recorded Future’s and their competitors’ success. Is it sentiment that molds the news or is it news that molds the sentiment? Which came first, the chicken or the egg? The symbiotic relationship between sentiment and news leads to the next phase of future-analytics technology…

If we can predict the future, can we mold it as well by manipulating the data? Perhaps that’s why the CIA is so interested (and Google, for that matter).

Come November, we may either start hearing more about Recorded Future or it may sink back into obscurity.

Here’s a glimpse of the service:

https://www.youtube.com/watch?v=Un8toBItXdo