changeset 54:c0111eb7298d

rebuttal

--- a/musicweb.bib	Mon Jun 13 12:45:07 2016 +0100
+++ b/musicweb.bib	Thu Jun 16 00:12:40 2016 +0100
@@ -2,7 +2,7 @@
 %% http://bibdesk.sourceforge.net/
 
 
-%% Created for George Fazekas at 2016-05-01 03:12:50 +0100 
+%% Created for George Fazekas at 2016-05-01 03:12:50 +0100
 
 
 %% Saved with string encoding Unicode (UTF-8)

--- a/musicweb.tex	Mon Jun 13 12:45:07 2016 +0100
+++ b/musicweb.tex	Thu Jun 16 00:12:40 2016 +0100
@@ -163,9 +163,9 @@
 
 
 \section{Introduction}\label{sec:introduction}
-In recent years we have witnessed an explosion of information, a consequence of millions of users producing and consuming web resources. Researchers and industry have recognised the potential of this data, and have endeavoured to develop methods to handle such a vast amount of information: to understand and manage it, to transform into knowledge. Multimedia content providers have devoted a lot of energy to analysing consumer preference, in an effort to offer customised user experiences. Music stream services, for instance, carry out extensive analysis trying to identify patterns in user's listening habits, and researchers are striving to refine multimedia recommendation algorithms. There are two main approaches to music recommendation\cite{Song2012}: the first is known as \emph{collaborative filtering}\cite{Su2009}, which recommends music items based on the choices of similar users.  The second model is based on audio content analysis, or \emph{music information retrieval}. The task here is to extract low to high-level audio features such as tempo, key, metric structure, melodic and harmonic sequences, instrument recognition and song segmentation, which are then used to measure music similarity\cite{Aucoutourier2002}, to carry out genre classification or to identify the mood of the song\cite{Kim2010}. Music discovery websites such as Last.fm\footnote{http://www.last.fm}, Allmusic\footnote{http://www.allmusic.com} or Pandora\footnote{http://www.pandora.com} have successfully developed hybrid systems which combine both approaches.
-There are, however, limitations in both approaches to music recommendation. Most users participating in (or whose data is used to analyse) collaborative filtering listen to a very small percentage of the music available, the so called ``short-tail'', whereas the much larger ``long-tail'' remains mainly unknown\cite{Celma2010}. These systems will show a bias towards music that is already consumed by many listeners. Suggesting already popular music will increase the likelihood of it being recommended to new users, thus creating a \emph{rich-club phenomenon}\cite{Zhou2004} or what is known as \emph{cumulative advantage}. Also, content analysis of audio features is mainly applied to songs: systems can recommend similar tracks, but generally know nothing about similar artists. Many music listeners follow artists because of their style and would be interested in music from similar artists. It is very hard to pinpoint what exactly makes two artists ``similar'': very often notions of similarity are based on social and cultural issues, rather than a precise definition of style.
- To many music lovers discovering new music, or music they weren't aware of, is an integral part of enjoying a musical experience, and appreciate expanding their musical taste. Lee and Price\cite{Lee2015} identify seven different \emph{personas} which tipify music service consumption. Two such personas, for instance, the ``active curator'' and the ``music epicurean'' characteristically spend a long time hunting for new music, whereas the ``wanderer'' enjoys the discovery process itself, trying out new things with an open mind.  Automatic musical discovery is a very challenging problem\cite{Jennings2007}. There are many different ways in which people are attracted to new artists: word of mouth, their network of friends, music magazines or blogs, songs heard in a movie or a T.V. commercial, they might be interested in a musician who has played with another artist or been mentioned as an influence,  etc. The route from listening to one artist and discovering a new one would sometimes seem very disconcerting were it to be drawn on paper. A listener is not so much following a map as exploring new territory, with many possible forks and shortcuts. Music discovery systems generally disregard this kind of information, often because it is very nuanced and difficult to parse and interpret.
+In recent years we have witnessed an explosion of information, a consequence of millions of users producing and consuming web resources. Researchers and industry have recognised the potential of this data, and have endeavoured to develop methods to handle such a vast amount of information: to understand and manage it, to transform into knowledge. Multimedia content providers have devoted a lot of energy to analysing consumer preference, in an effort to offer customised user experiences. Music stream services, for instance, carry out extensive analysis trying to identify patterns in user's listening habits, and researchers are striving to refine multimedia recommendation algorithms. There are two main approaches to music recommendation\cite{Song2012}: the first is known as \emph{collaborative filtering}\cite{Su2009}, which recommends music items based on the choices of similar users.  The second model is based on audio content analysis, or \emph{music information retrieval}. The task here is to extract low to high-level audio features such as tempo, key, metric structure, melodic and harmonic sequences, instrument recognition and song segmentation, which are then used to measure music similarity\cite{Aucoutourier2002}, to carry out genre classification or to identify the mood of the song\cite{Kim2010}. Music discovery websites such as Last.fm\footnote{http://www.last.fm}, Allmusic\footnote{http://www.allmusic.com} or Pandora\footnote{http://www.pandora.com} have successfully developed hybrid systems which combine both approaches.\\
+There are, however, limitations in both approaches to music recommendation. Most users participating in (or whose data is used to analyse) collaborative filtering listen to a very small percentage of the music available, the so called ``short-tail'', whereas the much larger ``long-tail'' remains mainly unknown\cite{Celma2010}. These systems will show a bias towards music that is already consumed by many listeners. Suggesting already popular music will increase the likelihood of it being recommended to new users, thus creating a \emph{rich-club phenomenon}\cite{Zhou2004} or what is known as \emph{cumulative advantage}. Also, content analysis of audio features is mainly applied to songs: systems can recommend similar tracks, but generally know nothing about similar artists. Many music listeners follow artists because of their style and would be interested in music from similar artists. It is very hard to pinpoint what exactly makes two artists ``similar'': very often notions of similarity are based on social and cultural issues, rather than a precise definition of style.\\
+ To many music lovers discovering new music, or music they weren't aware of, is an integral part of enjoying a musical experience, and appreciate expanding their musical taste. Lee and Price\cite{Lee2015} identify seven different \emph{personas} which tipify music service consumption. Two such personas, for instance, the ``active curator'' and the ``music epicurean'' characteristically spend a long time hunting for new music, whereas the ``wanderer'' enjoys the discovery process itself, trying out new things with an open mind.  Automatic musical discovery is a very challenging problem\cite{Jennings2007}. There are many different ways in which people are attracted to new artists: word of mouth, their network of friends, music magazines or blogs, songs heard in a movie or a T.V. commercial, they might be interested in a musician who has played with another artist or been mentioned as an influence,  etc. The route from listening to one artist and discovering a new one would sometimes seem very disconcerting were it to be drawn on paper. A listener is not so much following a map as exploring new territory, with many possible forks and shortcuts. Music discovery systems generally disregard this kind of information, often because it is very nuanced and difficult to parse and interpret.\\
 MusicWeb is an application which offers the user the possibility of exploring editorial, cultural and musical links between artists. It gathers, extracts and manages musical metadata from many different sources and connects them in informative ways. This paper deals with the different ways in which musicweb collects these resources and shapes them into high-level information. We will first review various knowledege-based web resources available to musicweb. We will then introduce the application itself and detail the architecture to analyse and extract data. Before the final conclusions and discussion of future work we will analyse the experience of interfacing with the application and how users can explore and discover new musical paths.
 
 
@@ -337,7 +337,7 @@
 \section{Conclusions and Future work}\label{sec:conclusions}
 
 % \subsection{Future work}
-MusicWeb is an emerging application being developed at a research centre to explore the possibilities of linked data-based music discovery and tested with a small user base of student and staff. % As such, its user base is at the moment limited to member of the research group (roughly 30 people, students and staff). Hence there are no major technical infrastructure requirements which cannot be supported by the university. This does not mean that we do not intend to utilise very large datasets. It was conceived and it is being developed mainly as a research tool. 
+MusicWeb is an emerging application being developed at a research centre to explore the possibilities of linked data-based music discovery and tested with a small user base of student and staff. % As such, its user base is at the moment limited to member of the research group (roughly 30 people, students and staff). Hence there are no major technical infrastructure requirements which cannot be supported by the university. This does not mean that we do not intend to utilise very large datasets. It was conceived and it is being developed mainly as a research tool.
 % actually C4DM is over 70 people!!
 Our aim is to gather in one application various different approaches to music discovery and how they can benefit from linked music metadata. Our next steps are directed toward evaluating its potential acceptance by end users. It would be of great value to us to find out which linking methods listeners find most appealing or interesting, and which they would use more often.
 As to the different methods of linking music metadata, our next steps will be:

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/rebuttal.txt	Thu Jun 16 00:12:40 2016 +0100
@@ -0,0 +1,17 @@
+Thank you very much for the reviews, they have been very useful and have given us great suggestions as to how to move forward. As suggested by one of the reviewers, we have been contemplating a journal article which would include a more thorough evaluation, including discovery span and user engagement, testing of individual components related to different modalities in our discovery application and testing the system holistically. This constitutes future work.
+
+Many of the comments can and would be addressed in a camera ready version of the paper: description of figures, a better diagram of the architecture, unifying naming criteria, etc. We would be very happy to discuss other technical details, which we thought might clutter the content of the paper, which had to be kept necessarily short. These include infrastructure, triple stores, rest services, databases, deployment, expected web traffic and performance benchmarking, concurrent data mining and big data handling.
+Also, we could write many pages about difficulties encountered, overcome as well as ongoing. There are many issues which we have tried to address: reliability of services, NLP entity extraction and topic modelling, identifying relevant texts (using metadata, tagging and non-curated raw text), as well as content based matching.
+There is an excellent suggestion to relate two artists through indirect connections. This is a very good comment and we should definitely take it into account in future work. However, our current focus is on finding sets of artists that are in the overlap of certain categories. In this sense, using direct links seems more appropriate. Using indirect links would result in navigation by multiple categories at once which may be less clear to the user.
+
+We would be to include more detailed descriptions of the methodology implemented, with more details provided on the specific algorithms and the rationale behind their use. In the case of computing artists' similarity, the main purpose of the content-based analysis is to establish musically overlapping factors in two artists’ repertoire rather than computing similarity directly.
+
+ We do feel, however, that there are certain misunderstandings which we perhaps failed to make clear:
+
+- The application is not a music recommendation system. The paper presents an emerging application for music artist discovery. In contrast with conventional recommendation system that provide recommendation by similarity, the focus here is on novelty and serendipity which are commonly identified as important requirements in music recommendation systems[Celma, 2010], presenting problems that are yet to be addressed successfully. Typical systems employ collaborative filtering or similarities in curated metadata sources. These approaches do not reach artists in the long tail of distributions computed from listening habits or preferences in social networks, while using curated metadata doesn’t scale to large catalogues and fails to reach new artists.
+In the camera ready version, a more thorough account on requirements gleaned from music recommendation will be provided together with specific problems in the domain of artist discovery. We will enumerate how our system meets some of these requirements and provide an assessment for cases where they aren’t yet successfully met. We believe that thorough testing is outside the scope of the present submission. The system’s ability to dynamically create interesting links between artists proves an initial hypothesis that Linked Data combined with text and music content processing can provide a faceted browsing experience that is relevant in music artist discovery.
+- The system demonstrates a first experiment in artist discovery allowing users to navigate the vast space of music artists by combining multiple modalities. The faceted browsing interface allows users to choose a direction most relevant to their information seeking task (i.e. cultural links, overlaps of certain musical factors between artists, typical mood, etc.). As mentioned earlier, MusicWeb is not a music recommendation system. Consequently, techniques applied in that domain are less relevant in our case and direct comparison with recommendation methods in algorithms would therefore be somewhat moot.
+
+Finally, Dimebag Darrell was the guitarist for Pantera.
+
+O. Celma. Music Recommendation and Discovery:The Long Tail, Long Fail, and Long Play in the Digital Music Space. Springer Verlag, 2010.