Feedback wanted: A new way to consume news

[ABDULLAH GHANI]

I have been working on a news aggregator platform Zaream. It basically collects news from multiple sources/platforms across the globe and labels them on the basis of source, genre, and topics. The main idea is to visualize an event (a news) happening in any part of the world from multiple sources and give a brief idea of the event collecting information from multiple sources.
Fake news has become a major issue in this era of technology and needs to be solved. Zaream provides API for news collected from various channels. Accumulation of similar news and Analysis (both semantic and syntactic) are in beta stage.

Need Feedback for the idea..

Link : Zaream - News api at scale

 

Don't like ads? Remove them while supporting the forum: Subscribe to Fastlane Insiders.

[Kak]

Interesting philosophical question for you... Is “fake” objective?

This sounds like Flipboard, Apple News, Google Now, and RSS feeds. How is this different? Better?

 

[ABDULLAH GHANI]

Interesting philosophical question for you... Is “fake” objective?

This sounds like Flipboard, Apple News, Google Now, and RSS feeds. How is this different? Better?

Fake News is one of the objectives. The main idea is an accumulation of similar news or news belonging to the same topic as shown by different media channels. Unlike other existing applications, the idea is to create relations between multiple events without human intervention thereby producing an unbiased view of events

 

[Jello]

Fake News is one of the objectives.

I don't think Kak meant if fake news is your objective to filter out. But rather was asking you the philosophical question if and when fake news is considered fake in the eyes of the reader of the news.

 

Don't like ads? Remove them while supporting the forum: Subscribe to Fastlane Insiders.

[ABDULLAH GHANI]

I don't think Kak meant if fake news is your objective to filter out. But rather was asking you the philosophical question if and when fake news is considered fake in the eyes of the reader of the news.

Got the point. So the perspective of fake news is the false rumor spread by paid media and has got nothing to do with consideration of an individual for a news. That's a totally philosophical approach to my idea. Need to think more about this. Open to any suggestions.

 

[njord]

Fake News is one of the objectives. The main idea is an accumulation of similar news or news belonging to the same topic as shown by different media channels. Unlike other existing applications, the idea is to create relations between multiple events without human intervention thereby producing an unbiased view of events

But its "humans" uploading the videos how do you sensor it (18+ content)? and what happens if some dude uses multiple IP adresses varify fake news?

 

[ABDULLAH GHANI]

I don't think Kak meant if fake news is your objective to filter out. But rather was asking you the philosophical question if and when fake news is considered fake in the eyes of the reader of the news.

Fake news is a never-ending thing. And it's not related to the objective proposed here. Zaream provides news API by collecting news from multiple online sources (all are recognized).

News will be rated based on the reputation of the source and the occurrence of similar news across multiple sources. This idea is still in development. And your idea of detecting fake news is completely different and can't be achieved using the proposed idea.

Please try Zaream - News api at scale to get a better Idea

 

Don't like ads? Remove them while supporting the forum: Subscribe to Fastlane Insiders.

[jon.M]

Unlike other existing applications, the idea is to create relations between multiple events without human intervention thereby producing an unbiased view of events

Except that you're aggregating news that are written by totally biased humans. You might give an unbiased view over what's trending in the media. But you aren't giving an unbiased, true view on world events. Those are two separate things.

Are you doing sentiment analysis to bring out the most objective articles or something? How exactly are you doing that?

How would you rate the reputation of, let's say Washington Post? How fake-newsy would you say an article about Jeff bezos would be, if it was written by them?

 

[ABDULLAH GHANI]

See I am just providing similar news from multiple sources. It's up to you to rely on or reject a particular source.

If some channel reports news of Jeff Bezos for a particular topic, I am aggregating similar news for the same topic of Bezos . And my algo (still in development) will just rate news based on the source (recognised globally) and occurrences (for now) of same news across providers. Each news sources are rated (weighed) based on some publically accepted figures (either no. of active readers or ranking on some metrics, ). Need to figure out the criteria. I am using NER for topic segregation and Sentiment analysis for sentiment

See this fake news detection is a good topic for discussion and I am still figuring out ways for proper detection.

 

[ABDULLAH GHANI]

News provided here is not meant for the end user to consume. Its a platform for developers and researchers who can subscribe to certain genres or location or sources for news. Custom filters and proper tagging are still in the pipeline of development. Need inputs from this forum for improving and streamlining the idea.

 

Don't like ads? Remove them while supporting the forum: Subscribe to Fastlane Insiders.