Summarizer aims to shorten your daily news by leveraging AI

in blockchain •  3 years ago 

image.png

Digital newspapers are one of the online media that are widely used in the search and dissemination of information in the era of convergence. In addition to factors originating from readers and media organizations that increase the popularity of digital newspapers, the development of internet technology has also influenced the development of digital newspapers, such as: the development of online media, the development of mobile devices and smartphones, mobile internet access, cellular network technology (wireless). broadband), and the development of social media networks.

There are many platforms where people can get information, both through offline media such as newspapers and online media such as Google News. But in modern times like today, most people use online media to find out more quickly and easily. They can get the information they need based on their preferences. Users can receive messages directly on their smartphone or via email with a subscription. This method is believed to be more efficient than searching for news on online platforms.

However, there are other ways that are believed to be more efficient to get different information faster. Depending on the category you want, this is Summarizer, a fully automated newspaper that allows users to spend less time reading the news. Using intelligent algorithms, Summarizer can provide users with a variety of optimized content that can be read by users.

It is a fully computerized newspaper.
Summarizer uses artificial intelligence to make your daily news shorter. The bot scours the internet for news, summarizes them, and then categorizes them.

What exactly is a SUMMARIZER?
Summarizer is only available to $SMR owners. You don't have to pay anything to read Summarizer's content; all you have to do is hold $SMR. You can sell your $SMR back to the market at any time if you decide to stop reading Summarizer.

Summary function
Summarizer is a platform that uses AI to summarize daily news for users around the world. With a family of bots, Summarizer searches the web for messages, processes them, and makes them available to users through a modern and clean interface. But that's not all, there are several other Summarizer features:

  • AI Powered : Using intelligent AI, Summarizer will make daily messages shorter and easier for users to read. Summarizer uses TextRank to optimize message content.

  • Elegant and fast: Summarizer's interface is modern and fast-paced and designed to make readers feel comfortable reading the news and not be distracted by unimportant content.

  • Telegram Support: With telegram bot, Summarizer can quickly send messages to users via telegram.

  • No Payment Required: User can enjoy daily news without paying even $1. Users only need to have $SMR tokens and can enjoy the content for free.

  • Privacy: Unlike other messaging platforms, Summarizer has no tracking code that monitors what users read. On the other hand, Summarizer ensures that user privacy is completely secure.

The algorithm we use
TextRank is an unattended algorithm for automatic text summarization that can also be used to get the most important keywords in a document. The algorithm applies a variation of PageRank over a graph created specifically for the summary task. This results in a ranking of elements in the graph: the most important elements are those that better describe the text. This approach allows TextRank to generate summaries without the need for a training corpus or labeling and allows the use of algorithms with different languages.

For automated summarizing tasks, TextRank models any document as a graph using sentences as nodes. A function to calculate the similarity of sentences is needed to construct the edges between them. This function is used to give weight to the edges of the graph, the higher the similarity between sentences, the more important the edges between them in the graph. In the Random Walker domain, as is often used in PageRank, we can say that we tend to move from one sentence to another if they are very similar.

TextRank determines the similarity relationship between two sentences based on the content they share. This overlap is calculated simply as the number of common lexical tokens between them, divided by their respective lengths to avoid promoting long sentences.

The function returned in the original algorithm can be formalized as:
Definition 1. Given Si , Sj two sentences represented by a set of n words which in Si are denoted by Si = wi , wi , …, wi . The similarity function for Si, Sj can be defined as:

image.png

The result of this process is a solid graphic that represents the document. From this graph, PageRank is used to calculate the significance of each node. The most significant sentences are selected and presented in the same order as they appear in the document as a summary. These ideas are based on changing the way the spacing between sentences is calculated to give weight to the edges of the graph used for PageRank. This similarity measure is orthogonal to the TextRank model, so it can be easily integrated into the algorithm. We found several variations of this to produce significant improvements over the original algorithm.

BM25 BM25 / Okapi-BM25 is a rating function which is widely used as an advanced task for Information Retrieval. BM25 is a variation of the TF-IDF model using a probabilistic model.

Definition 2. Given two sentences R, S, BM25 is defined as:

image.png

where k and b are parameters. We use k = 1.2 and b = 0.75. avgDL is the average sentence length in our collection.

The definition of this function implies that if a word appears in more than half of the collection's documents, it will have a negative value. Since this can cause problems at a later stage of the algorithm, we use the following correction formula:

image.png

where takes values ​​between 0.5 and 0.30 and avgIDF is the average IDF for all terms. Another corrective strategy was also tested, setting = 0 and using a simple modification of the classic IDF formula.

Evaluation
We tested LCS, Cosine Sim, BM25, and BM25+ as different ways of weighing edges for TextRank graphs. The best results are obtained using BM25 and BM25+ with the corrective formula shown in equation 3. We reach

image.png

image.png

Tokenomics.
The distribution of SMR tokens will take place in 3 stages. Private sale for $0.008 per SMR. Join the whitelist to participate in private sales! The public sale will be made after the private sale at a rate of USD 0.01 per SMR. Launch on PancakeSwap, planned after private and public sale. Starting price: $0.012 per SMR. The project allocation is shown in the screenshot.

image.png

The token allocation is propagated to the Binance Smart Chain and we are integrating tokens with Summarizer via Web3. Our SMR token has passed the TechRate audit. The source code has also been published and tested by BSCScan.

Roadmap Summary
Summarizer aims to streamline your daily news by leveraging AI.

  • TextRank implementation – We implemented TextRank as the base block for the Summarizer. The basic idea of ​​TextRank is to score each sentence in a text, then use the top n sentences to make a summary.

  • Design Summarizer – We wanted it to be elegant & fast. Comes with dark & ​​light mode, removes all unnecessary elements, optimized for speed.

  • Build an army of bots – Run by bots, but for the human Summarizer run by a family of bots. There are crawler-bot, summa-bot, editor-bot, delivery-bot, optimizing-bot, repairing-bot, etc.

  • Beta Summarizer – Let it run its course. We've been running it in beta mode for over a year. It has processed more than 60k articles, by itself.

  • Integrate with SMR Token – We are implementing SMR on the Binance Smart Chain and integrating tokens with Summarizer via Web3 Summarizer will soon be exclusive to SMR holders.

  • Summarizer on Android – Even though the website is quite fast & sleek, we wanted to take advantage of the native Android features for Summarizer readers. The app will be available for Android users in just a few days.

  • Summarizer on iOS – We are building iOS concurrently with Android app iOS users will experience Summarizer fully soon.

  • SMR Landing Page – We created this to offer users information about our token launch in the most transparent manner.

  • Marketing Plan for Launch – To spread the word about Summarizer widely, we plan our marketing strategy with various activities in all sectors. From articles about bitcoin talk, establishing a presence on social networks like Twitter, Reddit, building Telegram communities, setting up airdrop & content giving systems, running ads on cryptocurrency related websites, being listed on coinhunt & coinsniper, etc.

  • Audited by TechRate – Our token, SMR, has been audited by TechRate. The source code has also been published and verified on BSCScan.

  • Distribution of SMR Tokens – Distribution will be carried out in three phases. Private Sale, at a rate of $0.008 per SMR. The Public Sale, which will occur after the Private Sale, is $0.01 per SMR. The launch on PancakeSwap, happened after the Private & Public Sale. Starting Price: $0.012 per SMR

  • Sustainable Marketing Activities – From articles about bitcoin talk, building presence on social networks like Twitter, Reddit, building Telegram community, setting up airdrop & content giving system, running ads on cryptocurrency related websites, listed on coinhunt & coinsniper, etc.

  • Tracked by CoinGecko & CoinMarketCap – We have fully built the SMR token profile on BSCScan and have submitted our application to CoinGecko and CoinMarketCap

  • Make it exclusive to SMR holders – After the distribution phase, Summarizer content will be exclusive to SMR holders.

  • List of Exchanges – List of SMR on the first few exchanges.

  • Berkeley Model Implementation – This update will improve summary quality and clarity. The model compresses the source document text based on the constraints of constituent decomposition and RST discourse decomposition. In addition, it can improve the clarity of the summary by re-expressing pronouns whose antecedents will be removed or are not clear.

  • Summarizer Tailored for Each User –Each user on Summarizer will be able to customize their experience and content that appears in the UI. Choose and redesign the Summarizer layout & category list based on your interests.

  • Bridging SMR to other chains using Anyswap – Using Anyswap, we will bridge SMR to Ethereum, Polygon, and Harmony blockchains. This will increase the accessibility of Summarizer to other chain users. And by making SMR available on multiple chains, it will create more trading pairs & trading volume for SMR on DEXs like Uniswap and SushiSwap.

  • Open Summarizer technology to other news publishers – We plan to open our technology to other news publishers. Using Summarizer, news publishers can easily create summaries for their articles in bulk, conveying those summaries to their readers via newsletters and news feeds. They can even create their own version of the Summarizer website with just a few clicks. Profits generated from this activity will be used to buy back & burn SMR.

Summary Team

  • Brandon Thomas – Frontend Developer
  • Chris Miller – Blockchain Developer
  • Joy Stewart – Communications Manager
  • Julie Hardin – Marketing Manager
  • Mike Cook – Graphic Designer
  • Robert Hoover – Backend Developer
  • Steve Willis – Software Engineer

Detailed Information:
Official website – https://summarizer.co/
Tokenomics website – https://token.summarizer.co/
Telegram – https://t.me/SummarizerOfficial
Twitter – https://twitter.com/SummarizerC
Medium – https://medium.com/@summarizer
Reddit – https://www.reddit.com/user/Summarizer_Official

Author : Amild
Bitcointalk : https://bitcointalk.org/index.php?action=profile;u=2583828
My BSC Wallet ; 0xbf00577895715883E63C6694D33dA51b1cDEBDa8

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!