In this work we explore the potential role of music album cover arts for the task of predicting the overall rating of music albums and we investigate if one can judge a music album by its cover alone. We present the results of our participation to the Linked Data Mining Challenge at the Know@LOD 2016 Workshop, which suggest that the the cover album alone might not be su cient for the rating prediction task.
each album as \good" when the critics' score for it is greater than 80 and \bad" when lesser than 60.
To answer our question, we learn a SVM model that classi es music albums either as \good" or \bad" and we train the model only using the cover art of each album. The feature extraction from the cover art is based on image embeddings. 2
Our proposed approach for album classi cation consists of three main steps. Given a collection of music albums, we rst obtain the image of their cover art. Then, using o -the-shelf tools we obtain a feature vector representation of the images. We then learn a classi er to label each album as \good" or \bad", only exploiting the feature space obtained from its cover art. Figure 1 depicts the proposed pipeline. In this work we use the Know@LOD 2016 challenge dataset. It consists of 1,600 music albums. Each item provides: { album name { artist name { album release date { DBpedia4 URI for the album { the classi cation of the album as \good" or \bad".
The organisers split the dataset as 80% (1,280 instances) for training, and 20% (320 instances) for the test.
In our experiment we deliberately only want to exploit the cover art of the
albums, therefore we only use only the DBpedia URIs of the albums to obtain
their cover images. First, by using Rapidminer LOD extension [
image of the cover art. Then, by using the Mediawiki API6, we download all the images.
The resulting image set consists of 1558 images, with 2 images missing (the path obtained from dbp:cover did not correspond to any Wikipedia resource).
The dataset is available at https://github.com/petrovskip/know-lod2016, together with the extracted feature vectors and process used (explained in section 3). 2.2
Classi cation approach.
We learn a SVM model that classi es music albums either as \good" or \bad".
Starting from our image set, we use image embeddings to obtain a feature space.
Feature set. We use the Ca e deep learning framework [
Learning process We use all obtained features (without any ne-tuning) to train the C-SVM classi er (a wrapper implementation of libSVM classi er on Rapidminer) with a linear kernel and the default parameters. 3
We evaluated our approach using 10-fold cross validation on the training set and we obtained an accuracy of 58.03%. The resulting confusion matrix is presented in Table 1. The accuracy of the test set reported by the challenge system is of 60.3125%.
The low accuracy of our approach seems to suggest that the image alone is not a good predictor of the overall rating of the musical album. Nevertheless it would
true good true bad class precision pred. good 385 282 57.72% pred. bad 254 356 58.36% class recall 60.25% 55.80% be interesting to investigate if ne-tuning of the features or the combination pf other factors could lead to better results. 4
This paper presents our submission to the Linked Data Mining Challenge at the Know@LOD 2016 workshop. We proposed an approach that classi es music albums into \good" or \bad" based solely on their cover art. We trained a SVM classi er with the feature vector calculated from the album's cover art to solve the prediction problem of music album classi cation. While our approach has some interesting results, our experiment hints that only using album covers as features is not the best t for the task of the challenge.