Summarized news in your pocket



You need to become a Contributor to join the discussion - Find out how.
Varun ShenoyMaker@varunshenoy · High School Developer and Researcher
Hey Product Hunters! I'm Varun Shenoy, and this is what I've been working on for the past couple of months. Essentially, it's a personalized news summarization app for iOS. It leverages core natural language processing technology built into the iOS operating system to create fluent summaries with complementary visuals in the form of infographics. There is no back-end and all the code runs on the device, which means that it will never go down. It won me a scholarship to WWDC 2016, and I am truly excited to share it with everyone. It's definitely something I am going to work on for a while, as it has been a great learning opportunity for me. I built it because I love the idea of summarized news, yet most of the apps on the App Store that claimed to do so were often lacking and were not cohesive enough. Being a member of my high school's Speech & Debate team also propelled my interest in summarized text. We have to stay up to date on current events, and summarized news can help us when we don't have enough time to read long and often drawn out articles. Being a freshman in high school, I will be constantly releasing new updates to make Summit even more efficient and accurate. An Apple Watch companion app is in the works as well. I am going to be around for the rest of the day, so feel free to ask me any questions regarding the software. I'm a one man show, and thus any feedback would be much appreciated. Thank you!
Sajeesh Philip@philipsajeesh · Founder: &
@varunshenoy Really good job Varun :)
Vladislav Arbatov@vladzima · CTO @ P2L.TV
@varunshenoy Hey Varun! Very good job, congrats on that. It's actually similar to what we've been developing recently (with an ML standalone backend though) and I like your approach with built-in iOS technology.
Varun ShenoyMaker@varunshenoy · High School Developer and Researcher
Tweets by Musolé 🇿🇲 🇺🇸 🇦🇴@musolek · Marketing+SocialMedia+Crypto | #Travel
@varunshenoy love this. Eventually could you add a dark mode to this? Thanks
Varun ShenoyMaker@varunshenoy · High School Developer and Researcher
@musolek Of course! I have added it to my list of features for future updates. Thanks for checking Summit out!
Lesile Liu@luiyezheng · student, writer on
Really good job Varun! I can't remember what I was doing when I was 14 year old :-)
Varun ShenoyMaker@varunshenoy · High School Developer and Researcher
Thanks @luiyezheng!
Robert Paraschiv@robertpara · Product Manager
@varunshenoy you da man Varun. When I was 14 I was mainly focused on playing Duke Nukem 3D. I would personally love the ability to extract summaries from a connected Pocket account in a future release. Great job dude!
Varun ShenoyMaker@varunshenoy · High School Developer and Researcher
@robertpara Thanks for the donation! It truly means a lot! Pocket integration and an action extension are on the list of features I will be implementing in the future. As a side note, I still spend a good chunk of my time playing video games, mainly Fallout and Borderlands. Thanks for checking Summit out!
Vivek Sancheti@evivz · Co-Founder CryptoGround.Com
Hey the design and concept looks cool overall. But one question you say "infographics and maps" whats that? I havent seen it anywhere in the app.
Varun ShenoyMaker@varunshenoy · High School Developer and Researcher
@evivz The infographics and maps are provided underneath the summary. The infographics include ring charts, money highlights, and many more when certain information is extracted from the text. The maps are shown when there is a key location found in the text. Thanks for checking out Summit!
Apoorv Saini@apoorvsaini
It is just taking few sentences out of the news articles. Summarisation is a bit more complex.
Varun ShenoyMaker@varunshenoy · High School Developer and Researcher
@apoorvsaini There are actually two types of summarization, extractive and abstractive. Abstractive summaries are very difficult to do as it requires an extremely good language parser and it is simply too difficult. Extractive is a lot more common. It generally involves creating a bag of words with n-grams and then applying weights.