Given that the exact mechanism of poison delivery (and the scene of the crime) are still unknown, Itsu faced a potential PR nightmare in defending their Health and Safety compliance, the quality of their sushi and the security of their clientele.

Rather than hide and hope this would blow over, Itsu has stepped up to their own role in the drama. As they’ve closed this restaurant for decontamination, the board across the front says:

Itsu, Piccadilly (London W1J) closed for the investigation of the poisoning of former Russian spy Alexander Litvinenko

An interesting approach, capitalising on such a bizarre, James-Bond style set of events. (Do note the black swirly background.) By boosting their own “casual bystander” role in this affair, Itsu is painting themselves as just close enough to the the events to enjoy the romance of it all, but also excusing themselves (probably accurately, it appears) from any blame.

Their flamboyant sign should ensure that people talk, spreading their message of innocence. I was amused enough by their efforts to help them in their viral marketing (by this posting) and will be interested to see how their sales turn out.

Itsu’s website (which has no signs of the cloak-and-dagger theme of their Piccadilly boards) indicates that they expect to reopen in early 2007.

Work on this blog has been temporarily halted due to my holiday. Please feel free to browse and do leave comments, requests, concerns and suggestions. I will be happy to respond to them when we re-open on the 22nd of November. Thank you for your custom.

I’ve been comparing search engines, looking for something suitable for a large amount of unstructured data from a lot of repositories. Now, option 1 says it operates on the probabilistic model of information retrieval (a description of this model is in this paper: part 1 and part 2), though the implementers are extremely vague on exactly how they’re using it.

As far as I can tell, the probabilistic model creates a score for each document based on the probabilities of each of your search terms being in that document. Probability of term 1 being there + probability of term 2 being there (etc) = matching score, which you can then use to rank this document against other documents.

In this implementation, they then weight the search terms that are rarest in all the documents (so that if you’re in a law firm and search for “Smith litigation”, “Smith” will be more important than “litigation”. Your firm will probably have a lot using the term “litigation” so it won’t be as useful to pick out the docs you need). It then normalises for document length, balances repeated terms (so that searching for ’smith smith litigation’ doesn’t mean it looks for documents with “smith” twice as often) and trims words to their stems using something like the Porter Stemming algorithm.

Okay, now I’ll admit I’m learning. But this algorithm isn’t new: this ‘City model’ of the probabilistic was initially proposed by Robertson and Sparck Jones in 1976 (‘Relevance weighting of search terms’. Journal of the American Society for Information Science, 27, 129-146)  Is this still as good as we can get?

[Tune in next time: we'll weigh this up against probabilistic latent semantic analysis and I can finally get around to asking my question!]