Sunday, March 06, 2005

Research Plan

Stage I
· Initial thoughts (Inception)
· Study of existing area
· Problem identification
· Scope for a research
· Research Questions
· High level Modelling of understanding


Stage II
· Identifying specific areas of interest in the problem domain
· High level modelling of specific area
· Identifying the major components and modelling components
o Modelling the templates
o Modelling the personalising filter
o Modelling the supporting Database


Stage III
· Modelling the interfaces
· Prototyping
· Testing



Stage IV
· Reviewing and Refining the models
· Possible extensions
· Case studies



State V
· Finalising the project
· Making a rough report
· Final thesis

AD-HOC IDEAS

(10th December 2004- while thinking about usability issues and expandability of the proposed system)

· Is the system need to support multi-users?
· Do we need make the roaming personal profiles by placing the profile database online?
·

26th December 2004

§ You can start with a thick client and may continue as an online client, as your search progress!
§ Track changes in web pages of interest to you or create a Persistent Search on a topic of interest to you (http://www.changedetect.com/)
§ Well what about a search result/ search job instead of a webpage of website?

Personal Web Search Agents

In the current world of Electronic Informatics, Info Search is more a relative term than a general one. General Search results have no relevance where people are looking for personalised, highly specific information.

One great thing ever happened in our civilisation is undoubtedly, the Internet – an abundant source of rich information covering almost all domains of knowledge humans are ever exposed to. Unfortunately present info search technologies; tools or services were not up to the mark in catering with the specific information needs of the users.

Some of the issues coming to my mind with the current online search practices are:

· Web search services normally returns huge number of results which makes it difficult for the user to get what he exactly after.

· Many of them can be quite irrelevant to the user (might make sense generally, but not user specific)

· Difficulties in putting your ideas (info seek thoughts) in search patterns – the web service might not understand what you mean!

· Services normally show quite generalised approach!

· Unexercised users might spend/waist considerable amount of time

searching specific information. In terms of today’s econo-sensitive business world, this means waist of human resources, hardware/software resource, energy and time – making it a pull back force in productivity.

Scope for a new search technique – potential refinement areas

· More specific user sensitive info search result returns

· Highly customised to user preferences and info search habits – “an effort to understand what user means from the search pattern given
instead of blindly searching for the pattern matching pages on Internet.”

· A learning strategy to cater with the dynamic search habits of the user.
Have great potential as sematic web evolves