Answering Featured Snippets Timely, Using Sentence Compression on News

Timely Featured Snippets and Sentence Compression

How is a knowledge graph updated when some earth-shaking event takes place? Is a search engine manually editing information in that knowledge graph? It seems like an area that could be using a machine learning element to automate it and keep it up to date.

Another place that would benefit from machine learning would be generating featured snippets that answer questions people might ask at Google, and it appears that Google thought it might be useful there, too. A Wired Magazine article from Monday describes how a sentence compression algorithm behind these featured snippets might be used:

Google’s Hand-Fed AI Now Gives Answers, Not Just Search Results

At the heart of this approach is the crawling of a data store of news articles and other sources, with the help of a “massive team of PhD linguists it calls Pygmalion”, and the use of algorithms that are referred to as “sentence compression” algorithms that might generate answers to questions from sources such as that news sources for featured snippets.

Continue reading “Answering Featured Snippets Timely, Using Sentence Compression on News”

Ranking Events in Google Search Results

Ranking Events without Links

This summer, Google was granted a patent that describes how the search engine might rank events based upon data that might indicate the popularity of those events, without relying on things such as the number of links pointed to pages about those events. The patent involves ranking events that occur in physical locations.

Examples of the kinds of events talked about in this patent include ranking events like music concerts, art exhibits, and athletic contests, all happening for specified periods of times at specified physical locations, such as concert halls, galleries, stadiums, or museums.

Since many events in a geographic region can happen at the same time or at overlapping times, interested individuals may at times find it difficult to determine which events to attend. For example, individuals may be unaware that events of interest are scheduled to occur or may have difficulty identifying the most interesting events when multiple events are occurring.

This ranking events patent lays out a general process flow to describe how the method in the patent works. It starts with receiving data about a physical location, and events taking place there during a certain time period and computing signal scores for those events based upon things such as a mention of the event and a popularity score for the event based upon those signal scores.

Continue reading “Ranking Events in Google Search Results”

Google News Recommendations and the Google Knowledge Base

Google News Recommendations

I’ve seen posts from SEO by the Sea show up in search results with an “In the News” heading above them, even though my site hasn’t officially been accepted in Google News. Some blog posts that have been given that “In the News” treatment have been criticized lately. See: Google does a better job with fake news than Facebook, but there’s a big loophole it hasn’t fixed. It seems that this criticism is going to have an impact, with the “In the News” label taken away from Google Search Results:

Google is removing its ‘In the news’ label due to the fake news nightmare

Are there any other solutions? I do like when something I write is treated as newsworthy and is presented to a larger audience in a way that helps those posts stand out, but sometimes satire blog posts end up being treated that way as well. This article points out some other possible solutions:

Continue reading “Google News Recommendations and the Google Knowledge Base”

GS1 Web Vocabulary Schema Workshops in California

An Extension to a Web Vocaulary Schema from GS1

California bear flag

I noticed a blog post published yesterday, November 2, 2016, and it looked helpful: Use JSON-LD to add to your Website. Schema and structured data seem to be growing in importance on the Web, as we see more knowledge panels and rich snippets and product search results. I’ve been looking at Knowledge Panels in Site Audits. JSON-LD seems to be the favored Web vocabulary Schema by Google in adding structured data on your web pages. See: What is JSON-LD? A Talk with Gregg Kellogg.

If you do SEO and aren’t familiar with GS1, you probably should be. They invented the use of bar codes in shopping. They also came up with GTINS (Global Trade Item Numbers) which are used online at places such as eBay and Amazon, and Google Product Search. A recent blog post by GS1 Vice President Rich Richardson is also worth reading: Why bar code numbers matter.

In February, GS1 published an extension to a wb vocabulary Schema for products. Extensions like this are how Search and SEO are growing. The Schema blog told us about it in:

Continue reading “GS1 Web Vocabulary Schema Workshops in California”

Entities in the Google Knowledge Graph Search API for Google

Exploring The Google Knowledge Graph Search API


The Google Knowledge Graph Search API on a query for Google shows the following Entities and results scores for them. I thought they were diverse enough to be interesting and worth sharing. A couple of the ones listed seem odd, such as the Indian Action movie. “Thuppakki” and the Town in Kansas,”Topeka.” (It seems like there is a song titled, “Google Google” in the film Thuppakki, and in 2010 Topeka renamed itself “Google” to try to attract Google Fiber to the area.) We are told by Google that “Results with higher result scores are considered better matches.”

These are the Google Knowledge Graph Search API results on a search for Google:

Google “resultScore”: 292.863342
Google Chrome “resultScore”: 51.392109
X “resultScore”: 51.392109
Googleplex “resultScore”: 44.052853
Google China “resultScore”: 30.75222
Google Lively “resultScore”: 30.75222
DoubleClick “resultScore”: 29.141159
GV “resultScore”: 28.957876
Thuppakki “resultScore”: 28.693569
Google Store “resultScore”: 26.077885
“Google Japan” “resultScore”: 24.272602
DeepMind Technologies “resultScore”: 24.115602
Topeka “resultScore”: 23.718664
Rich Miner “resultScore”: 21.961121
Google Capital “resultScore”: 21.048887
Google Hacks “resultScore”: 21.003328
“Google Korea” “resultScore”: 20.818398
Barney Google and Snuffy Smith “resultScore”: 20.384176
Verily Life Sciences “resultScore”: 19.65727
Patrick Pichette “resultScore”: 19.614473

Continue reading “Entities in the Google Knowledge Graph Search API for Google”