| || |
How are Search Engine Issues are tested in webTesting
How do we test the Search engine issues?
[This message has been edited by sanvibha (edited 09-22-2001).]
Re: How are Search Engine Issues are tested in webTesting
The best way would be to get the logic and rules of the search engine. You need to understand what search strings it allows and what operators (such as +, -, etc.) you can use in those search strings. You also have to determine what rules are used for things like relevance factors, if those are used at all.
It would be handy to know if you are talking about your own search engine (meaning a specific application of your own) or an existing search engine.
Also please note that the actual definition of "search engine testing" is: "Testing of the visibility of a particular Web site when using suitable search engines to locate the site with limited information." Thus if this is what you mean, you are also going to want to understand the means by which the visibility factors are generated by a search engine.
A good paper on this was given in the IEEE Computer magazine called "Estimating the Usefulness of Search Engines" by Weiyi Meng, King-Lup Liu, Clement Yu, and Wensheng Wu. The paper presents a statistical method to estimate the usefulness of a search engine for any given query where the "usefulness" of a search engine in this paper is defined to be a combination of the number of documents in the search engine that are sufficiently similar to the query and the average similarity of these documents. It also talks about placement of documents within search engines (such as how close to the top of the list you are or should be).
Another paper called "Interface with Text Data Mining for Web Search Engine" by Hiroyuki Kawano and Toshiharu Hasegawa talks about algorithms you can use derived associative keywords by an extended association algorithm. (Basically, a way to compute test cases.)