Between Obvious and Interesting

This post is a continuation of Good Recommendations, Using Set Theory to analyse Recommendation relationships, Variation within Preferences and Predicting Preferences.

I’ve offered the idea that good recommendations lie on a scale between Obvious and Interesting. Taken to their extreme, the full line could actually run from Boring through to Random, with Obvious and Interesting somewhere in between. ‘Boring’ recommendations could be said to exist where the Advisor’s Preference Set is entirely made up of the Common Set (i.e. the Advisee knows at least as much as the Advisor). ‘Random’ recommendations could be said to exist where there is no Common Set at all (i.e. A disjoint B )

I’ve also attempted to explain my thoughts on the kinds of relationships that could exist between an Advisor and an Advisee, and ways in which they could be elaborated. I think I’ve done that, albeit in a not very scientific way. So there’s a lot of room for refinement, and there are some gaps to be filled (specifically around calculating variation within a set of preferences and analysing effects of different weightings of those preferences) and some of my assumptions are a little more tenuous than maybe they should be, but I think this could provide the basis for some interesting results.

I’m also keen to explore the possibility that subjective relationships (rather than behavioural relationships) between Things would produce better/more interesting recommendations and routes for discovery. For my MSc project, I’m intending to focus on the Music Domain, analysing music blogs to deduce relationships between artists to augment recommendations from existing services such as last.fm.

Sounds like: “Uncle Fucka”

I ‘learned’ Obla-dee-obla-daa by The Beatles’ at the weekend. And a sterling work of musicianship it was. In my learning journey, I have noticed how the verse sounds remarkably similar to the Uncle Fucka song, as popularised by the South Park movie, ‘South Park: Bigger Longer & Uncut’. McCartney must be proud.

I’ve also learned that there is an alternative pronounciation of capo. Having always assumed this to be a one way street, straight up cap-oh, I was slightly disorientated to hear my guitar tutor offer an alternative pronounciation: cape-oh. I’m not sure it actually matters all that much. Tomorrow, I think we’re progressing to Wonderwall. I’m quite looking forward to it.

In other news, I recieved Richard Swift Collection: Volume One in the post today. It’s the first CD I’ve bought since the Jens CD – which has been almost never off my playlist, pushing Mr Lekman right to the top of my last.fm profile. Incidentally, I’m also pleased to report that a sudden recent gush of playing has pushed Radiohead into third place, overtaking ABBA who only really feature that prominently because I kept playing SOS, and forgetting to flick shuffle on before leaving it running.

I’ve decided that what I play doesn’t actually reflect what I like all that much very well. Which is a bit of a paradox, really. This guy has written a lot about the difference between Last.fm and Pandora, which I played about at the weekend for the first time, having deferred doing so assuming that they were more or less the same.

It makes for a good read – I had a similar idea of writing something down that compared all these music-comparision-suggest-something-you’ll-love services, throwing Amazon’s releational-purchasing “people who bought this, also bought that” into the bag too.

I’m generally coming around to the conclusion that it’s not very relevant for a computer to recommend something you’ll die for to listen to again. With a few exceptions, most people who know me can’t recommend me something I’ll fall head over heals in love with, so the chances of having some automated process that can is reaching the near impossible. I got thinking about how I’d got into Jens – who I now rate right up there with some of my favourite artists – and it really was a chance happening across an MP3 blog over Christmas. Jens’ closest match via Pandora is Belle and Sebastian, who although I like, I’m hardly their biggest fan, and it’s not like they’re unchartered land in my musical world.

As Steve Krause pointed out, there’s obviously a need for a mix of both the social upping of Last.fm, and the genetic musical matching of Pandora. However, for true new music discovery, I’m finding that it’s ultimately a matter of hard graft and exposure to more musical avenues. And by that I mean catching more MP3 blogs, podcasts and the like in addition to my usual diet. Something that filters out the noise would be great though…