Not Signed-In
Which clippings match 'Recommendation' keyword pg.1 of 1
04 JANUARY 2014

An introduction to recommender systems

1

TAGS

advogato.org • affinity analysisalgorithmic filtersAmazon.comautomatic predictions • collaborative filtering • collaborative filtering approach • correlationsdata matchingdata miningecho chamberfavourite things • FilmTrust • information filtering • information filtering system • information patterns • interests • Jennifer Golbeck • large datasetsLast.fm • moleskiing.it • Pandora Radiopersonal tastepersonalised suggestionsprediction • process of filtering • rating systemrecommendationrecommendation enginerecommendation platform • recommendation system • recommendation systems • recommender systems • relatednessrelationships between individualsserendipitysimilaritysimilarity machinesimilitude • trust metric • trust-based recommender systems • user datauser preferences

CONTRIBUTOR

Simon Perkins
22 MAY 2011

Mixcloud: joining the dots between radio shows, Podcasts and DJ mixes

"Quite simply, Mixcloud helps connect radio content to listeners more effectively. Mixcloud is re–thinking radio by joining the dots between radio shows, Podcasts and DJ mixes. We refer to them as Cloudcasts – audio shows that are stored in the 'cloud' and available to be streamed on–demand.

A Cloudcast is an extended audio show that is hosted in the 'cloud', in other words hosted somewhere centrally on the Internet rather than on your local hard–drive. Therefore, unlike Podcasts, Cloudcasts can be enjoyed via on–demand streaming without any waiting required. ...

Mixcloud is a legal platform for on–demand radio shows, Podcasts and DJ mixes which contain copyrighted music. The only licensing option available at present is a streaming format, and Mixcloud is committed to providing a legal option which supports artists and ensures they receive royalties for the shows and mixes which contain music. In addition, the site can keep track of what you (and your friends) listen to, so we can make better recommendations and provide a more personalised site tuned to exactly what you like to hear."

(Mixcloud)

TAGS

2011audio • audio shows • authorshipcloud computing • cloudcast • compilationcopyrightcustomdigital audioDJDJ mixes • favourite songs • joining the dots • listeners • mix of songs • Mixcloud • mixed tape • mixes • mixtapemusicon-demand • on-demand streaming • personal mixtape • personalised • playlistpodcast • podcasts • PPL • PRS for Music • radio • radio content • radio shows • recommendationremixremix cultureroyaltiesscriptible • streamed on-demand • streaming format • what you like to hear

CONTRIBUTOR

Simon Perkins
27 MARCH 2011

Wisdom of the Masses' perception of the web, can in some cases promote content perceived to be useless over content perceived as useful

"According to the 'long tail' principle, ICT innovations in content creation and distribution such as virtual inventories, Internet Protocol Television (IPTV) and other types of video on demand, music self–publishing in social networking sites and digital printing challenge old rationales that justified the adoption of mass–market models for the production and publication of cultural goods. These technologies dissolve the spatial and physical constraints which limited the range of creative content goods available in the market and open the gates for a flood of new (and old) media. In doing so they have created a new problem, of a navigational nature: in principle, diversity enables access to content goods better suited to a customer's preferences, but it also makes finding them more difficult (194).

The main reason for the success of Google's search services has been its ability to address Internet users' need for relevant resources, by adopting a scalable algorithm that establishes a webpage's rank according to its reputation. However, its user interface is still too rigid and makes it difficult, for example, to fully specify the type of content a user is looking for. Additionally, this technique, based on a 'Wisdom of the Masses' perception of the web, can in some cases promote content perceived to be useless over content perceived as useful, and be tampered with through search optimisation techniques such as link farming (195)."

(Juan Mateos–Garcia, Aldo Geuna and W. Edward Steinmueller, 2008, p.85)

194: In a context where information is abundant, attention becomes the scarce resource (Simon, H. A. 1971, 'Designing Organizations for an Information–Rich World', in Martin Greenberger, Computers, Communication, and the Public Interest, Baltimore, MD: The Johns Hopkins Pres).

195: i.e. exchanging reciprocal links with web sites in order to increase search engine optimization, as search engines often rank sites according to, among other things, the quantity of sites that link to them.

Fig.1 Perry Ogden (2003). 'Bono with Louis Le Brocquy'.

2). Fabienne Abadie, Ioannis Maghiros, and Corina Pascu (editors) 2008 'The Future Evolution of the Creative Content Industries: Three Discussion Papers', Office for Official Publications of the European Communities, EUR 23633 EN – 2008

1

TAGS

2007 • consumer demand • content creationcontent distributioncreative content • creative content sector • creative goods • cultural goods • cultural spacesdigital printingdiscoverabilitydiscussion paper • dominant business models • EPIS06 • ETEPS • Europe • European Perspectives on the Information Society • European Techno-Economic Policy Support • evolutionevolution of ICTGoogle IncICTinnovation • Institute for Prospective Technological Studies • Internet Protocol Television • IPTS • IPTV • Joint Research Centre • JRC • link farming • long tail • mass production • mass-market models • metadata • music self-publishing • navigationnavigation systemnetworkold mediapolicy makerspolicy makingpopular votepopularity rankingproductionpublicationrecommendationrepositoryreputation • scalable algorithm • searchsearch engine optimisationsearch servicesocial networking • strategic intelligence • The Institute for Prospective Technological Studies • video on demandvirtual inventories • webpage rank • wisdom of the masses

CONTRIBUTOR

Simon Perkins
14 FEBRUARY 2010

Get your Last.fm recommendations in Spotify

"Another improvement has been rolled out to Spotify recommendations engine Spotibot. The service, which aims to help users find music they like, but just don't know yet, has added Last.fm integration.

What that means is that instead of having to put in a song or band name you like, you can just get recommendations based on your recent listening instead. You just put in your Last.fm account name, or the account of anyone else for that matter, and you can generate a list of 5 to 30 songs that you're going to love. Well, maybe. ...

This is just the 'recommendations' bit of Last.fm at work. It also has 'your library', 'neighborhood' and 'loved tracks' functionality that could easily be added in the same way, so keep your eyes on Spotibot over the next month or so."

(Pocket–lint)

1

CONTRIBUTOR

Simon Perkins
11 JULY 2006

Last.fm: Recommending Music By People With Matching Tastes

Last.fm
Last.fm is the flagship product from the team that designed the Audioscrobbler system, a music engine based on a massive collection of Music Profiles. Each music profile belongs to one person, and describes their taste in music. Last.fm uses these music profiles to make personalised recommendations, match you up with people who like similar music, and generate custom radio stations for each person.

1

CONTRIBUTOR

Mia Thornton
Sign-In

Sign-In to Folksonomy

Can't access your account?

New to Folksonomy?

Sign-Up or learn more.