Fidest – Agenzia giornalistica/press agency

Quotidiano di informazione – Anno 32 n° 250

BPU Holdings Uses Advanced Technology to Filter Fake News

Posted by fidest press agency su domenica, 5 agosto 2018

Fake news – it’s a term that has become all too familiar today. With the ever-increasing flow of information, trustworthy news has become increasingly debatable. The journal Science reports false news stories are 70% more likely to be re-tweeted than true stories, true stories take about six times longer to reach an audience and fake news is shared nearly 10 times more than true stories1. BPU Holdings uses Artificial Intelligence (AI) and advanced technology in their products and services to omit unreliable information to better diversify the quality of valuable information. ZimGO Polling and Neil’s News Bot use a distributed and scalable filtering processes to manage quality control of fake news to give their users an authentic experience.
ZimGO Polling is BPU’s political forecasting tool that uses Twitter social networks to connect to the empowered citizen in ways not possible with legacy tools. It is the first campaign tool to accurately assess emotional citizen sentiment that provides a competitive advantage for campaign professionals and candidates. It was determined the most valuable citizen information was found on Twitter. However, statistical processing was affected by the high volume of shared posts, re-tweets and spammers. To manage this, our AI interacts with the Twitter API to process streams of duplicates and unwanted noise. After the initial filtering, any re-tweets are discarded so the noise is highly reduced. A second measure is text filtering; distributed and scalable patterns are strained from the high numbers of duplicate tweets without jeopardizing the overall latency of stream results. If we notice the same text coming through, we assume it’s a duplicate. Going further, Similarity Deduplication is something currently being established to work on defining concepts to catch small nuances on a bot posting. In this third layer, we’re looking for small variations; for example, many bots post the same text except for the link and this third layer of filtering would trap those. A fourth level is provided by allowing people to tag posts as spam, which then further trains the AI to look for irrelevant posts.
Neil is BPU’s personal news bot. It is a personal news curator for artificial Intelligence (AI) that is a virtual assistant bringing reputable articles and interesting stories from reliable news sources. Unlike ZimGO Polling, it uses quality control with a white list of about 3,000 sources thoughtfully selected to provide our users with the best news on topics individuals have selected. Secondly, quality scoring is applied by manually checking the white list to ensure they are accurately providing the necessary and requested news the users search. This applies a second measure of filtering to avoid sources that are too short, lack text and contain an influx of links.


Inserisci i tuoi dati qui sotto o clicca su un'icona per effettuare l'accesso:

Logo di

Stai commentando usando il tuo account Chiudi sessione /  Modifica )

Google photo

Stai commentando usando il tuo account Google. Chiudi sessione /  Modifica )

Foto Twitter

Stai commentando usando il tuo account Twitter. Chiudi sessione /  Modifica )

Foto di Facebook

Stai commentando usando il tuo account Facebook. Chiudi sessione /  Modifica )

Connessione a %s...

%d blogger hanno fatto clic su Mi Piace per questo: