Here are links to 10 posts I wrote this year about things that I found in patents and white papers from the search engines. I’ve included links to the patent filings and whitepapers as well, and included the date that each post was published in parentheses after the links to those posts.
I broke them down into the following categories:
– Local Intent in Search
– Ambiguous Locations
– Synergy Between Local and Mobile
– Sentiment Analysis in Local Reviews
I tried to include a short introduction to each section as well, to provide an idea of why I thought those were important.
Local Intent in Search
On December 15th of this year, Yahoo started showing Yahoo local results within organic search results for query terms that they thought might have a local intent, even when that query didn’t have a geographic term included with it, according to a Yahoo blog post, Get More Personally Relevant Results When You Search for Local Businesses.
I noticed patent filings from Google and Yahoo, and a Yahoo whitepaper over the course of the year that gave us some hints on how each of the search engines might determine whether there might be a local intent in a query that doesn’t include a location. That kind of local intent associated with a query might trigger the appearance of local search results in organic web searches. Understanding why a query might be determined to have a local intent, without a location, has been helpful.
1. Google Local Search, Categories, and the What and Where of Local Map Listings (April 21st, 2009)
Google patent filing – Interpreting Local Search Queries
2. How a Search Engine Might Determine Whether a Search Involves a Geographical Intent (May 18th, 2009)
Yahoo whitepaper – Discovering Usersí Specific Geo Intention in Web Search (pdf)
3. How Search Engines Might Divine the Intent behind Regional Queries vs. Global Queries (December 14th, 2009)
Yahoo patent filing – Identifying Regional Sensitive Queries in Web Search
When a local search contains some ambiguous geographic information such as a landmark or area name such as “Space Needle” or “Times Square,” instead of an actual street, city, or state name, it might attempt to associate that limited geographical information with an actual location. People do search for hotels or restaurants or other types of businesses near landmarks. I wrote about a Google patent filing that explained how they might keep track of this information, and score locations when there is more than one landmark or area with the same name (such as “Times Square”). This was also the first place I had seen Google mention that they might include user-added data such as “my maps” submissions to help index what might be found at different locations.
1. Google Geocoding, Ambiguous Locations, and My Maps Submitted Data (July 13th, 2009)
Google patent filing – Geocoding Multi-Feature Addresses
Synergy Between Local and Mobile
In addition to announcing the launch of Google Latitude in February of 2009, Google Maps Navigation for Android in October of 2009, and the release of an official Google phone early next year, Google started taking some other steps that draw mobile phones and local search closer together.
Three Google patent applications published this year describe how barcodes in business windows can be used to help you learn more about a business, how mobile phone cameras may be used with product search and may use GPS to let you look at the online catalogs for stores you are presently shopping within, and how you can rate local businesses by phone while you’re still shopping.
If you’re following local search but skimming past information about tie-ins with mobile devices, you’re missing out on a very important element of local search.
1. Google Barcodes and Place Rank Transforming Local Search (December 10th, 2009)
Google patent filing – Machine-Readable Representation of Geographic Information
2. How Google Might Let you Shop by Camera Phone (December 28th, 2009)
Google patent filing – Image Capture for Purchases
3. Google Approach to Making Online Ratings Easier… (October 2nd, 2009)
Google patent filing – Ratings Using Machine-Readable Representations
Sentiment Analysis in Local Reviews
Google held their second annual Searchology event in May of 2009, and part of the big news was the addition of some “smart snippets” in search results that could contain things like ratings for restaurants. Another major announcement was the addition of sentiment analyis in reviews shown to searchers, in a new options section that provides reviews, and within the reviews that are shown in local search. Three Google patent filings published this year, and a Google White paper from 2008 provided some possible insights into how Google goes about understanding the sentiment of content found in reviews, how it might address sentiment in other domains, and how it might rate raters.
1. Opinion Summaries in Google Maps Reviews (August 4th, 2009)
Google patent filng – Aspect-Based Sentiment Summarization
2. Google’s New Review Search Option and Sentiment Analysis (June 12th, 2009)
Google White paper – Building a Sentiment Summarizer for Local Service Reviews (pdf),
Google Patent filing – Domain-Specific Sentiment Classification
3. How Google May Rate Raters (June 15th, 2009)
Google patent filing – Rating Raters
Bill’s bio if you want to know more about him:
Continue reading Loci 2009: Bill Slawski’s Important Patents of 2009