The Science of Crawl (Part 2): Content Freshness

The Science of Crawl (Part 2): Content Freshness

In our previous post we introduced a funnel for deduplicating web documents within a search index. The dual problems of exact duplicate and near duplicate web document identification are considered. By chaining together several methods with increasing specificity we identify a system which provides sufficient precision and recall with minimal computational tradeoffs.

In this post, we look at a second challenge of maintaining a continually evolving corpus of web documents: content freshness. Roughly, freshness can be broken down into two categories: search tuning and corpus freshness.

Read More

App Search by Action

App Search by Action

Mobile apps are built to do one thing really well, whether that’s listening to music, reading the news, ordering food, or booking a ride.  They are designed to help you take action quickly so that you can move on with your life.

One of the main features of our App Search API is the ability to find apps by the action they let a user take.  When you create a query, you can signal the intended action of the results.  

Read More

Implementing Deferred Deep Linking

Implementing Deferred Deep Linking

We have received a lot of questions about how to take someone to a specific part of an app after they download the app. In other words, you should be able to show someone a link to content within your app and then send them directly to that content (rather than to the “home page” of your app or to the app store download page) after they install.

We’ll explain how to set this process for your Android and iOS apps.

Read More

Wearables & Deep Links - Powered by the URX Search API

We built a live app that takes in voice commands spoken to an Android Watch and constructs them into search queries to the URX API. The API returns relevant content from other mobile apps and websites so that our app can find other apps and open them directly. Check out our video to see this in action:

In the demo, we send a voice command to the watch, which converts the speech to text and sends the text to our Android phone app. Our app constructs queries from this text and sends them to the URX API. These queries contain an action and a some keywords; in the first example, the action is “Listen” and the keywords are “Justin Timberlake”.

The URX API returns a set of search results upon receiving the query. We built our app to open the top search result, which may be a link to specific content within another app, like Spotify, or a link to web content, like the Slate article about the California drought.

This is just one of many ways that developers can build on the URX Search API. You can read all about how to construct a query, our search operators, and see which apps we’ve indexed in our developer documentation.

Join our beta today and start building on the URX API!

API Docs Now Available

Last month we opened our beta program for our App Search API, which lets developers find and link users directly to relevant content inside other apps. We really appreciate all the feedback we’ve gotten from developers and are excited to be releasing the first apps built on our platform soon.

Today, we’re releasing our API documentation to everyone on developers.urx.com. It includes everything you want to know about the API, our SDKs, and how to structure search queries for the most relevant results. It also includes a full example of how to implement a music search to an Ellie Goulding song on iOS, Android, and the web.

We're building a knowledge graph to categorize the content inside of apps and our first feature is supported actions. These let you request links that help you take a specific action such as listen, watch, or buy.

Please check out our docs and let us know what you think!  We’re handing out API keys now and would love to work with you - sign up for our beta program and and we’ll be in touch soon.

Andrew

Help Us Build the Deep Link Search Engine for Developers

URX is changing the way that users view mobile content through the world's first search API. Our engineering team is a talented group of data scientists, machine learning experts, and software engineers that are tackling the challenge of fixing the broken mobile experience head-on.  

We're now further expanding this team to fill several key positions.  Take a look below to see if you think you might be a good fit!   

Read More

The Deeplink Debrief - September 18

The Deeplink Debrief - September 18

We rounded up the latest industry news, highlights and announcements in mobile engagement and deep linking to bring you the top articles to read. Subscribe here to receive the latest updates and insights from URX (delivered to your inbox every other week).

Read More

Deepscape 2.0

Deepscape 2.0

Over the past several months we’ve seen a number of major announcements by technology giants and stealth companies alike about the future of deep links, discovery, search and the impact it will have on the broader industry.

We updated the "Deepscape" (or deep link industry landscape) to reflect an increase in deep linking marketing tools and solutions, a heightened focus around app discovery and re-engagement and the emergence (at least on a more public stage) of search and accessibility as it relates to in-app and app-to-app content and actions.

Read More

KDD Retro: Google Knowledge Vault and Topic Modeling

The URX data science team recently attended the 2014 Knowledge Discover and Data mining (KDD) conference in New York. The conference was a whirlwind, four day tour of the latest research in data mining. The theme of these sessions was utilizing text information to improve knowledge inference. This encompasses a wide variety of text-based problems: topic modelling, structured knowledge extraction, bioinformatics, targeted advertising, and many, many more. In the wake of KDD14, we wanted to summarize a couple of our favorite programs and encourage feedback.

Read More

The Science of Crawl (Part 1): Deduplication of Web Content

Here at URX we are building the world's first mobile app search API. Backing this API is a search engine containing a large corpus of web documents, meticulously maintained and carefully grown by crawling internet content. We've come to discover that building a functional crawler can be done relatively cheaply, but building a robust crawler requires overcoming a few technical challenges. In this series of blog posts, we will walk through a few of these technical challenges including content deduplication, link prioritization, feature extraction and re-crawl estimation. 

In this first installment, I will walk through the duplicate web content problem.

Read More