Search APIs - Creating Your Own Custom Search Engine

By: Reshmi Prakashan
Submitted: 2008-04-22 13:12:40
Print this article | Tell a friend | For publisher |

Rating:

 

So, what is a search engine and what does it do? (and – can I run my own search engine on my domain – an analysis)

Search engine is a program or a tool that searches for the information on the Internet based on the words which user insert into the search box. After the words are inserted into the search box and when you click on search button the search engine looks into its own database to find out whether it contains those information which the user is actually looking for.

There are many major search engines of the web. Why are these considered to be "major" search engines? Because they are either well known or well used like Google and Yahoo. But then, there are lots of other search engines as well, for eg: DogPile is a search engine that is based on MetaSearch (a search engine pulling out results from other search engines) also among those who are “TRUE” search engines (Search engines having their own database/index are search engines like Exalead & Dhoondho http://www.dhoondho.com.

Among the newer search engines, Dhoondho is a fast growing search engine in India, similar to Exalead in France. It has a very useful feature for developers called as a “Search API” (http://www.dhoondho.com/run/your/own/search/engine/run_my_own_search_engine.html ). Using this Software Developers/companies can use the vast Internet knowledge of Dhoondho, to perform Internet related searches & then act upon the information received. Using this custom search engines can be built, documents can be searched for, all in a very convenient XML API.

Let’s look at a sample search engine & see why lots of people like gogoquick.com (http://www.gogoquick.com) are creating their own custom Search engines

Search engines such as Google, Yahoo(Yahoo was a directory & now a search engine) & Dhoondho search engine uses a web crawler as its web-indexing robot, which is known as spiders, or crawler, which comb the Internet looking for documents and their web addresses. However, different search engines have their own ways of finding relevance. For eg: Google uses – Page rank, I’m not sure what the others use though but Dhoondho seems to be using something called the “Relevance Rank” More details here - http://www.dhoondho.com/about/relevence/rank/relevence_rank_algorithm.html

Let’s say that if I would want to create my own search engine what options do I have? I can try the following –

NOTE: Tests are as per what we felt during our trial of the above APIs. There is no guarantee that the same comparison will hold in the future or was in the past. Users are advised to carry out their own tests before settling on an API. The Author or the publisher of this article refuses to be blamed for any damages arising from any of the above APIs. This is put here as a disclaimer to absolve us of any damages arising of use f this information & to make all the Lawyers who want to sue us for providing our own views, un-happy.

Article source: Expert Articles