The following muttering is in response to a question that came up about just how "accurate" are YGAMS™?
Let me state right up front. The math and logic behind the YGAMS value was an original concept of mine and while it is simple and sound in theory, it was not a derivative work. I have created what I believe to be a simple balanced measurement of popularity based on Internet chatter that is not reliant on any single source of information. When first put in motion the YGAMS™ engine has no preconceived ideas or startup values. It only knows how to query its five reference sources: Yahoo, Google, ASK.com, Microsoft, and Search.com on the bequest of the user and return a convienent list of links onto a single screen, while at the same time creating a mathmatical data set that contributes to the interpretation of the YGAMS score often refered to as a Gamys (pronounced Gamies) which is a measurement of Internet chatter surrounding a word, name or idea in relationship to the big picture.
This current beta allows me to provide live links that are capable of displaying the results of Gamys date lists that represent sets of related Gamys, in the form of charts, graphs, and discriptive text. This allows the friends and relitives as well as the public to view these data sets in the form of live examples and now as game pieces in TheYGAMSgame, a very challenging game. Unlike Trivia games, Gamys are based on what is, here and now, and not what was back then.
Maintaining sample consistency in the ever changing environment of the indexed Internet, while at the same time quantifying and logging what it has learned, has been both interesting and fun program to design. In an attempt to remain a good net-neighbor, previously calculated YGAMS are being cached for a day or more. The history charts are updated on a demand basis to reducing index and server load.
Every YGAMS calculation goes through the same normalizing process. This normalization process is based on the constantily changing internet and recalibrates when the log files rollover to a new day - every day. This helps to keep keeps the relative instability of the constantly changing indexes, in relationship with each other.
The bottom line is that YGAMS self calibrates once a day, based on what it has learned up to that point. In this way the YGAMS engine will keeps itself forever in tune with the constantly changing Internet. Even the garbage collection and housekeeping tasks are triggered by the search engines when they come around to crawl.
When I first started this project, I was surprised just how much the search results changed from day to day, as the spiders drop and reindex millions, if not billions of pages between them on some days. It became apparent that no one search engine could provide the unbiased big picture value I was looking for. It was through the analyses and normalization of five or more divergent results that would get me close to the relative popularity value I was looking for.
Please note that this is a work in progress, and will remain in beta for some time to come. I reserve the right to add and remove user accessible functions from the publicly accessible version over time as testing continues, and YGAMS™ as an information management product evolves.
Bob Bradlee - author and webmaster of YGAMS.COM™ - Paving the Gravel Path to the Information Super HighwaySM
You may find YGAMS History, along with a mutterings about YGAMS Scores, will help your understanding of the subject.
Screen modes: Text only, no graphs. - Phone-PDA Turse with Tables. - PC Verbose with Tables.
YGAMS™ © 2008 R.C.Bradlee
000\public@38.107.191.93 ~ 0s