Folks:
We listed the following as some of the factors that should affect the page importance, and
desirable characteristcs of an importance metric.
If you can think of other factors that should also be taken into consideration when deciding page
importance, please do make a comment:
–It should be sensitive to
•The query
–Or at least the topic of the query..
•The user
–Or at least the user population
•The link structure of the web
•The amount of accesses the page gets
–It should be stable w.r.t. small random changes in the network link structure
–It shouldn't be easy to subvert with intentional changes to link structure
4 comments:
This might be included in the last bullet point, but I can also think of the relative trustworthiness of the page -- where this factor could come from all the trust computation and propagation mechanisms we have discussed like A&H etc
How about considering the age of the document - how long it has been in the web.. Older documents may be considered trust worthy, authoritative and valuable.
Moreover we can also consider the Uniqueness of the page.
Actually I would append to Sanjay's comment and say that while considering the age, the trustworthiness should depend also on the query: a news-specific query, for instance, would place more weight on recent pages.
I think the page importance 'sensitive to the user / user context' itself are too generic and we might want to break this down further.
For example, one such information is 'User's Location', and queries like, "mcdonalds burger", "mattress for sale", the importance of pages will have to take his location into consideration.
And the 'temporal aspect' is interesting too. The recent pages about the search entity, say a 'person name' could decide the page importance. Example for this could be, if there was a search on "Jim Gray" in the past 2-3 weeks, the news pages describing 'Jim Gray's disappearance' will become more important.
Post a Comment