Receive the weekly sampler of posts and "Resource of the Week".
Subscribe »

Enter your
email address:

My Account »


Bookmark and Share

Testimonial?
If you find ResourceShelf useful, please supply a testimonial »








Home > ResourceBlog > Article

« All ResourceBlog Articles

 

Bookmark and Share   Feed

Thursday, 21st December 2006

Research: Where Do Good Query Terms Come From?

Where do good query terms come from?.
by Gheorghe Muresan and Dmitri Roussinov (2006)
Source: Proceedings 69th Annual Meeting of the American Society for Information Science and Technology (ASIST) via E-LIS

From the Abstract:

This paper describes a framework for investigating the quality of different query expansion approaches, and applies it in the HARD TRECexperimental setting. The intuition behind our approach is that each topic has an optimal term-based representation, i.e. a set of terms that best describe it, and that the effectiveness of any other representation is correlated with the overlap that it has with the optimal representation.
Indeed, we find that, for a wide number of candidate topic representations, obtained through various query-expansion approaches, there is a high
correlation between standard effectiveness measures (R-P, P@10, MAP) and term overlap with what is estimated to be the optimal representation. An important conclusion of comparing different query expansion approaches is that machines are better than humans at doing statistical calculations and at estimating which query terms are more likely to discriminate documents relevant for a given topic. This explains why, in the HARD track of TREC 2005, the overall conclusion was that interaction with the searcher and elicitation of additional information could not over-perform automatic procedures for query improvement. However, the best results are obtained from hybrid approaches, in which human relevance judgments are used by algorithms for deriving terms representations. This result suggest that the best approach in improving retrieval performance is probably to focus on implicit relevance feedback and novel interaction models based on ostention or mediation, which have shown great potential.

Source: E-LIS


Category:

Views: 704




blog comments powered by Disqus

« All ResourceBlog Articles

 

Read about the FreePint FamilyFreePint Family

A family of resources to help information workers be more effective, raise the value of information in their organisations and contribute to success. Read more »


FeedLatest Family Articles:


Click to view the article Quilting big data threads
Thursday, 24th May 2012

Recently I have found myself cooing over visualisation maps (and heat maps) of health and well being resources. The content rich data is overlayed with mapping technologies, and some interesting themes and patterns are emerging.


Click to view the article The fallacy of information overload
Wednesday, 23rd May 2012

A lot of the talk around social media in the last year has been around information overload. Social media has provided us with new and exciting ways to create content. But it has also meant learning new ways to manage and engage with social media tools. Are we teetering on the edge of an information overload precipice?


Click to view the article Information overload: fact, fantasy or filter failure?
Wednesday, 23rd May 2012

Information overload is a figment of your imagination. Or a failure of your filter. Or a symptom of your technological submissiveness. Depends on who you ask.


Click to view the article Newsdesk: tracking millions of pieces of information a day
Tuesday, 22nd May 2012

What if you had to sort through 3.5 million articles and social media posts a day and try to pull out the most relevant items for your organisation? What if you then had to cobble it all together into something readable for your top groups and executives in your organisation?


Click to view the article Alacra Compliance adds managerial oversight
Tuesday, 22nd May 2012

Alacra Compliance saves time by aggregating information from both free and fee-based sources and enabling users to conduct an accurate federated search across these sources (coined “simultaneous search” by Alacra).


All Family Articles »
Family Articles by Category »


Tell us what you're working on,
and we'll talk to you about how FreePint can help »


FreePint Family Testimonials

"Fabulous resource to learn of unique tools and insights. Very useful." Manager, Futures and Forecasting, Virginia, USA

More testimonials »






Subscribe

Subscribe to the ResourceShelf Newsletter and receive the weekly sampler of posts and Resource of the Week.

Find out more »

ResourceShelf sponsored by:

Article Categories

All Article Categories »

Archive

All Archives »