Search engine design and implementation

We are looking for new way of designing a search engine for the internet. Often the answers that you get from a search is not what you are looking for.

The thesis consists of three parts:

- design and implement a search bot that surfs the internet and collect relevant information from the sites visited.
This sub task can be very large, depending on what information we want to be able to search. In the first step we will only look at html documents in swedish or english, which will make this task relativly easy.

- design and implement a data structure or data base to store the information collected. The data structure or data base must be efficient when it comes to memory usage and search speed. It must also be flexible when it comes to search criteria, e.g. allow SQL commands. Also, the criteria that we want to use when searching the data base must be evaluated and possibly altered to deliver the desired result.
This is the difficult part of this thesis.

- design a web site that can act as the interface to the user. This part is flexible. It is not necessary to make an advanced solution, like using Tomcat and Java. A CGI solution using Perl or something simular will be good enough.

The thesis require you to design and implement these three parts. In the design process you will also be required to choose the appropriate tools and programming languages, and how advanced each sub tasks shall be to allow you to finish the project within the specified time frame.

In all parts of the project you will get the support that you need from our personnel.

We think that this is a good thesis for two students with a high level of knowledge, personal ambition and drive, probably from D or Y.