Skip to main content

2010 | Buch

Music Recommendation and Discovery

The Long Tail, Long Fail, and Long Play in the Digital Music Space

insite
SUCHEN

Über dieses Buch

In the last 15 years we have seen a major transformation in the world of music. - sicians use inexpensive personal computers instead of expensive recording studios to record, mix and engineer music. Musicians use the Internet to distribute their - sic for free instead of spending large amounts of money creating CDs, hiring trucks and shipping them to hundreds of record stores. As the cost to create and distribute recorded music has dropped, the amount of available music has grown dramatically. Twenty years ago a typical record store would have music by less than ten thousand artists, while today online music stores have music catalogs by nearly a million artists. While the amount of new music has grown, some of the traditional ways of ?nding music have diminished. Thirty years ago, the local radio DJ was a music tastemaker, ?nding new and interesting music for the local radio audience. Now - dio shows are programmed by large corporations that create playlists drawn from a limited pool of tracks. Similarly, record stores have been replaced by big box reta- ers that have ever-shrinking music departments. In the past, you could always ask the owner of the record store for music recommendations. You would learn what was new, what was good and what was selling. Now, however, you can no longer expect that the teenager behind the cash register will be an expert in new music, or even be someone who listens to music at all.

Inhaltsverzeichnis

Frontmatter
Chapter 1. Introduction
Abstract
In recent years typical music consumption behaviour has changed dramatically. Personal music collections have grown, aided by technological improvements in networks, storage, portability of devices and Internet services. The number and the availability of songs have de-emphasised their value; it is usually the case that users own many digital music files that they have only listened to once, or not at all. It seems reasonable to suppose that with efficient ways to create a personalised order of users’ collections, as well as ways to explore hidden “treasures” inside them, the value of their music collections would drastically increase.
Òscar Celma
Chapter 2. The Recommendation Problem
Abstract
Generally speaking, the reason people could be interested in using a recommender system is that they have so many items to choose from—in a limited period of time—that they cannot evaluate all the possible options. A recommender should be able to select and filter all this information to the user. Nowadays, the most successful recommender systems have been built for entertainment content domains, such as: movies, music, or books.
Òscar Celma
Chapter 3. Music Recommendation
Abstract
This chapter focuses on the recommendation problem in the music domain. Section 3.1 presents some common use cases in music recommendation. After that, Sect. 3.2, discusses user profiling and modelling, and how to link the elements of a user profile with the music concepts. Then, Sect. 3.3 presents the main components to describe the musical items, that are artists and songs. The existing music recommendation methods (collaborative filtering, content, context-based, and hybrid) and the pros and cons of each approach are presented in Sect. 3.4. Finally, Sect. 3.5 summarises the work presented, and provides some links with the remaining chapters of the book.
Òscar Celma
Chapter 4. The Long Tail in Recommender Systems
Abstract
The Long Tail is composed of a small number of popular items, the well-known hits, and the rest are located in the heavy tail, those not sell that well. The Long Tail offers the possibility to explore and discover—using automatic tools; such as recommenders or personalised filters—vast amounts of data. Until now, the world was ruled by the Hit or Miss categorisation, due in part to the shelf space limitation of the brick-and-mortar stores. A world where a music band could only succeed selling millions of albums, and touring worldwide.
Òscar Celma
Chapter 5. Evaluation Metrics
Abstract
This chapter presents the different evaluation methods for a recommender system. We introduce the existing metrics, as well as the pros and cons of each method. This chapter is the background for the following Chaps. 6 and 7, where the proposed metrics are used in real, large size, recommendation datasets.
Òscar Celma
Chapter 6. Network-Centric Evaluation
Abstract
In this chapter we present the network-centric evaluation approach. This method analyses the similarity network, created using any recommendation algorithm. Network-centric evaluation uses complex networks analysis to characterise the item collection. Also, we can combine the results from the network analysis with the popularity of the items, using the Long Tail model.
Òscar Celma
Chapter 7. User-Centric Evaluation
Abstract
Up to now, we have presented a user agnostic network-based analysis of the recommendations. In this chapter we present a user-centric evaluation of the recommender algorithms. This user-based approach focuses on evaluating the user’s perceived quality and usefulness of the recommendations. The evaluation method considers not only the subset of items that the user has interacted with, but also the items outside the user’s profile. The recommender algorithm predicts recommendations to a particular user—taking into account her profile—and then the user provides feedback about the recommended items. Figure 7.1 depicts the approach.
Òscar Celma
Chapter 8. Applications
Abstract
This chapter presents two implemented prototypes that are related with the main topics presented in the book; music discovery and recommendation. The first system, named, Searchsounds, is a music search engine based on text keyword searches, as well as a more like this button, that allows users to discover music by means of audio similarity. Thus, Searchsounds allows users to dig into the Long Tail, by providing music discovery using audio content-based similarity. The second system, named FOAFing the Music, is a music recommender system that focuses on the Long Tail of popularity, promoting unknown artists. The system also provides related information about the recommended artists, using information available on the web gathered from music related RSS feeds.
Òscar Celma
Chapter 9. Conclusions and Further Research
Abstract
Research in recommender systems is multidisciplinary. It includes several areas, such as: search and filtering, data mining, personalisation, social networks, text processing, complex networks, user interaction, information visualisation, signal processing, and domain specific models, among others. Furthermore, current research in recommender systems has strong industry impact, resulting in many practical applications.
Òscar Celma
Backmatter
Metadaten
Titel
Music Recommendation and Discovery
verfasst von
Òscar Celma
Copyright-Jahr
2010
Verlag
Springer Berlin Heidelberg
Electronic ISBN
978-3-642-13287-2
Print ISBN
978-3-642-13286-5
DOI
https://doi.org/10.1007/978-3-642-13287-2

Neuer Inhalt