A Look At Yahoo's New Search Algorithm

Most e-commerce entrepreneurs spend at least some of their time trying to psych out search engines in order to figure out how their algorithms are being applied.

The better the algorithms are understood, the better their sites can be optimized for search engine rankings.

What are algorithms? Algorithms are the rules that a search engine uses as its spider crawls web pages, indexing and sorting them for retrieval. Some of the most familiar algorithms these days have to do with:

- The number and quality of incoming links. The higher the number and the better the quality of the links, the higher the ranking.

- The number of related terms on a page, known as latent semantic indexing. The more terms on a page that are relevant to the keywords used to describe the page, the higher the relevance and rank for the page.

- Unique content. Search engines are looking for unique content, not content that has been copied from site to site.

Google, Yahoo, MSN and other major search engines are constantly looking for new and better ways to make their results more meaningful. For example, MSN has recently launched it decision engine, Bing. Bing organises search results into meaningful categories, gives you options to go directly to sub domains within a website, and adds alternate searches on the same page.

Yahoo has recently begun discussing its new wave of search engine algorithm, which involves extracting semantic relationships from query logs. It is a mouthful. This approach has become a topic of conversation because its patent application, which was filed in 2007, has just been published.

Yahoo analysts realised that when a searcher received a page of search engine results, the searcher frequently skipped many of the result citations, clicking in on those that seemed to best match their needs. They theorized that if you looked closely at the search terms used and the search terms associated with their actual choices, you could get a lot closer to understanding what the searcher was looking for and how to satisfy the search need.

They identified three types of relationships:

1. Synonyms (close relationship) - Query terms that share a substantially equivalent set of clicked search results.

2. Lesser but included (inclusive relationship) - Where the set of clicked results for one query term is smaller in size than another, and those clicked URLs are substantially included in the clicked URLs for the second query.

3. Related (lesser relationship) - Where the clicked search results between two queries overlap, but not quite to the same level as the two relationships above - synonyms and lesser but included.

Yahoo believes that they can divine these relationships by studying clicking behavior and determining related search phrases and terms that might result in a better page result for the searcher.

Naturally, the same approach can be applied to linking paid ads to search terms and related searches as well.

Interestingly, while Google has been king for quite some time, Bing may give it a bit of a challenge. If Yahoo can bring their semantic query theory to an operational level, they may be in the running to upset one the two larger search engines or at least, to preserve their third place standing.