Skip to main content

Wolfram Alpha

See All Stories
Site default logo image

Wolfram, creator of Siri’s knowledge base, releases impressive new Image Identification tech

Wolfram-Image-identification

Wolfram Research, the company behind the Wolfram Alpha knowledge base that Apple’s Siri taps into, is today releasing new artificial intelligence tech to answer the question, “What is this a picture of?”

Wolfram is showing off what the Wolfram Language Image Identification Project is capable of using this web app accessible on desktop and mobile devices. You simply drag any photo into the app and the image identification AI attempts to detect what it is (as pictured above).

Now I’m excited to be able to say that we’ve reached a milestone: there’s finally a function called ImageIdentify built into the Wolfram Language that lets you ask, “What is this a picture of?”—and get an answer…. It’s a nice practical example of artificial intelligence. But to me what’s more important is that we’ve reached the point where we can integrate this kind of “AI operation” right into the Wolfram Language—to use as a new, powerful building block for knowledge-based programming.

And the new ImageIdentify function will be accessible to developers to create APIs or apps that tap into the feature through Wolfram Language“And if one had lots of photographs, one could immediately write a Wolfram Language program that, for example, gave statistics on the different kinds of animals, or planes, or devices, or whatever, that appear in the photographs.”

Wolfram has a lengthy, interesting read on the backstory of the project and more about how everything works behind the scenes.

Apple confirms its ‘Applebot’ is indexing the web for Siri and Spotlight

Site default logo image

applebot

Rumors that Apple might be creating its own search engine started doing the rounds last fall, when webmasters started seeing web crawler visits from IP addresses starting with 17 – the IP address block owned entirely by Apple. Apple has now officially confirmed in a support document that it is operating its own web crawler called Applebot.

Applebot is the web crawler for Apple, used by products including Siri and Spotlight Suggestions. It respects customary robots.txt rules and robots meta tags. It originates in the 17.0.0.0 net block.

While the wording is clearly intended to suggest that this is just business as usual, both the fact that Apple is running its own web crawler at all, and the somewhat vague wording, are interesting … 
Expand
Expanding
Close

Siri vs Google search in 1600-question street test, speed test

Site default logo image

[youtube=http://www.youtube.com/watch?v=kDsOtdRtG0Q&start=23]

As noted by Phillip Elmer-DeWitt at Fortune, Apple analyst Gene Munster published a note to clients today that contained the results of a Siri vs. Google search 1600-question showdown.

While it is not exactly a test of how well the companies’ various voice services stack up against one another (since Google Search queries were typed-in and not spoken), but it is a good indication of just how viable Siri is as an everyday mobile search product and alternative to Google. In the test, both Google and Siri were asked 800 questions in a quiet location. Another 800 questions were asked among the loud street traffic in Minneapolis. The results, according to Fortune:
Expand
Expanding
Close

Siri represents nearly 25 percent of Wolfram Alpha queries after four months

Site default logo image

Apple’s digital secretary named Siri, an iPhone 4S exclusive, is now responsible for nearly 25 percent of all searches conducted on Wolfram Alpha, an answer-engine developed by Wolfram Research.

As you know, Apple collaborated with Wolfram Alpha on Siri (Microsoft is another licensee), and took advantage of algorithms powering Mathematica, another Wolfram Research product. It lets users type in complex factual queries, and then Wolfram Alpha computes accurate answers from its structured data containing hundreds of datasets.

The Cupertino, Calif.-based Company fancies Wolfram Alpha’s curated knowledge database, which is a nice fit for the iPhone 4S’s factual question answering feature. According to the New York Times’ Steve Lohr, Siri accounts for a quarter of all Wolfram Alpha queries after four months:

Siri accounts for about a quarter of the queries fielded by Wolfram Alpha, whose staff has grown to 200.

Google should be worried, as this could be another sign of Siri users becoming accustomed to retrieving factual answers from Wolfram Alpha and not Google. For example: Telling Siri to “Google the iPhone” launches Safari with Google search results accompanied by text-based adverts, but just asking “How many days are there until Easter” produces a formatted answer from Wolfram Alpha with no advertising whatsoever. This is also important knowing that a quarter of all searches on mobile devices are conducted through voice commands.


Expand
Expanding
Close

Siri now lets users access the entire Best Buy product catalog

Site default logo image

As noted by RazorianFly, Siri can now help users retrieve the best deals for consumer electronics and gadgets sold at Best Buy. How does she handle this, you might ask? Well, Siri gets some of its answers from the acclaimed answer-engine Wolfram|Alpha.

Wolfram Research announced yesterday they are leveraging data from Best Buy’s public application programming interface, allowing users to browse more than 35,000 appliances and consumer electronics products sold by Best Buy.

Third-party applications and services that integrate with Siri automatically benefit, so Siri is now able to deliver answers to your product-related queries sourced from Best Buy’s vast database.

In addition, taking into account Wolfram|Alpha’s clever decision-making and analysis engine and Siri’s natural-language interface, you can ask her to, let’s say, list plasma TVs larger than 50-inches. Even though Best Buy-sourced answers are laid out as any other information Siri sources from the web, Apple could -in the future- take advantage of Best Buy’s public APIs to produce rich results sporting product images, categories, more meta data and so forth.


Expand
Expanding
Close