Tweetsense's algorithm combs through social media posts (such as tweets) about a political candidate and reads each post like a human would. A sentiment rating is given to each post, ranging from happy to sad. Emotions which don't express clear sadness or happiness, such as alertness or fear, are marked neutral. The sentiments of the 7 days (or any span) are then aggregated for each candidate and the overall approval ratings for all candidates are printed on a line graph next to each other. The number of posts analyzed for each candidate is around 2000 per day.

When possible, Tweetsense uses demographic weighing on the posts to more accurately reflect the population. We also use demographic data to show exactly where a candidate may be leading: For example, Senator A could have a 6% lead on female approval while Senator B has a 9% lead on the latino vote. The latest version is also using popular topic data to show the same comparisons, such as who is leading by x% in terms of econ...
Tweetsense's algorithm combs through social media posts (such as tweets) about a political candidate and reads each post like a human would. A sentiment rating is given to each post, ranging from happy to sad. Emotions which don't express clear sadness or happiness, such as alertness or fear, are marked neutral. The sentiments of the 7 days (or any span) are then aggregated for each candidate and the overall approval ratings for all candidates are printed on a line graph next to each other. The number of posts analyzed for each candidate is around 2000 per day.

When possible, Tweetsense uses demographic weighing on the posts to more accurately reflect the population. We also use demographic data to show exactly where a candidate may be leading: For example, Senator A could have a 6% lead on female approval while Senator B has a 9% lead on the latino vote. The latest version is also using popular topic data to show the same comparisons, such as who is leading by x% in terms of economic change.

The algorithm works by using a combination of free natural language processors as well as our own for identifying political jargon. It has also been tested for non-political terms such as Coke vs Pepsi with entertaining results.
More information

Investors

Advisors

Carl Heine
Admin
Carl Heine Entrepreneur Coach, Curriculum Developer

Recommendations

Neel Bhat

William is onto something big here. Tweetsense, his latest creation, will be a game changer to those gathering poll data and analyzing trends in the political scene. Our world is changing; no longer do we turn to the daily newspaper for our news or wait for a phone call or someone to come to our door to give them our opinion on a political candidate. We get our news and provide our opinions through the various forms of social media. Tweetsense is the perfect tool to harness the potential of gathering and analyzing this data in real-time at a fraction of the cost and effort as before. William's success in the several startup competitions he's recently taken part in is a testament to the potential Tweetsense has.

James Gerry

William is the founder of Tweetsense. He is an outstanding citizen and visionary. His new product, Tweetsense, is technically advanced and offers a meaningful way to analyze Tweets based on their content. This can provide crowdsourced, not survey, insight into what people are thinking about. It also has a predictive portion that can help analysts understand upcoming trend direction. This business has already won several start-up contests and is worthy of much more attention. I highly recommend Tweetsense as an exciting and viable new business venture.