Add to Watchlist

AutoDJ [DEMO #12]


Citation of segment
Embed Code
Purchasing a DVD Cite video

Formal Metadata

Title AutoDJ [DEMO #12]
Title of Series 2014 Fall NuPIC Hackathon
Number of Parts 19
Author London, George
License CC Attribution 3.0 Unported:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
DOI 10.5446/18069
Publisher Numenta Platform for Intelligent Computing (NuPIC)
Release Date 2014
Language English

Content Metadata

Subject Area Computer Science
Abstract "I'm extracting my track-by-track music listening history from and then seeing if NUPIC can predict what artist I'm going to listen to next given a sequence of my previous listens." By George London.
Europe over over at the start of the test set out to write a lot of time and duration of the magnitude of the recommendation was from the music recommendation is that almost all existing systems work by trying to
figure out of your preference for much of almost same you have the same reference time and what time is what you're feeling what you're doing so what they have been used to make use of recommendations and context-sensitive but what you think might be good for looking at sequence songs that was or partially sequenced readings like forms of family units series songs and be able to detect the change my between some sequence of songs and the figure on the place so it is actually a form that is this is the 1st time the human you nuclear but it managed to download entire by year was the history and philosophy of science and then and then I think that it is actually thousand times the real songs that wasn't half of the UK only that 1 is the tendency of this fact but you can see here in this sequence is this the best artists and you told you there was a what is the name of God is the name of the track and mining and both of these I just use the category in the learning system also Sheridan nearly optimal but it also known as the Princess Consort sequences which subsampled process and it predicts next artists turns out be mostly in the US in the same order sentence now it was not an otherwise you listen to the 2nd last over the last so that's I don't actually have to evaluate quite yeah that's reasonable religion as I think therefore actually making it more reasonable it would help to provide a better context in the process of being an artist names in the fact vector mood of songs approximates the spectrograms everything in the spectrogram of times the problem of course lectures on anybody's interested in this problem endures internal data structure resembles that you're reading these with the with the consequence of this is the musicians in the center solve
checks all different qualities of culture so whether it true some the like that you have to get on the 1 hand and I I have I have used in Europe in the cool and I think I think it's reflected in doing this it can always
exactly that and that was the
end of the exam will be in the form of the results with those 2
hours of work In the talk to you get a lot good the the the the the mixture of
time on the internet and of
Musical ensemble
Multiplication sign
Demo (music)
Software testing
Order of magnitude
Physical system
Series (mathematics)
Context awareness
Process (computing)
Multiplication sign
Context-sensitive language
Image resolution
Data mining
Category of being
Vector space
Order (biology)
Data structure
Figurate number
Reading (process)
Physical system
Form (programming)
Form (programming)
Multiplication sign
Demo (music)


  227 ms - page object


AV-Portal 3.8.2 (0bb840d79881f4e1b2f2d6f66c37060441d4bb2e)