Automated Categorization of Search Results, a New Era?
Since the hakia Galleries have been on-line, we have received nothing but appraisals. Our proprietary approach to “Aspect Categorization” shines with examples in topics ranging from music to health. We currently cover more than a million popular queries.
hakia’s fully automated gallery production where the search results are categorized according to the query can be seen in the following demo link where 1425 different car brands and models are covered.
This is part of our ongoing effort to spread this capability to all search queries, effectively creating a new organization of the content on the entire Web, in a way as distinct as how Wikipedia invented its own style.
Microsoft’s recent news about KUMO and its screen-shots leave no doubt that some people are already convinced this is the way to progress in search.
Aspect categorization is different than what some search engines are already doing. For example, dividing the SERP into Web Results, Videos, News, Images, etc., is not aspect categorization. However, when the categories are related to the query, such as Obama’s Speeches and Quotes, Obama’s Fans, etc., (for the query Obama) then it is aspect categorization.
Aspect categorization in search is a tough business, it requires carefull off-line analysis to determine how the categories are going to be decided algorithmically, resources will be identified for crawling, and how the results will be detected to fit in.
The effectiveness of this approach in the broad search space is yet to be seen, and the users will have the last word as always. The tech bloggers and authors will be able to make their own judgment and recognize the limitations and imitations. In light of our patent application in progress, we are also anxious to see where all this leads to. Some exciting times ahead. Until then happy searching at hakia.
March 24th, 2009 at 3:46 pm
[...] hakia Blog » Blog Archive » Automated Categorization of Search … [...]