Modeling of the Latent Embedding of Music using Deep Neural Network

While both the data volume and heterogeneity of the digital music content is huge, it has become increasingly important and convenient to build a recommendation or search system to facilitate surfacing these content to the user or consumer community. Most of the recommendation models fall into two primary species, collaborative filtering based and content based approaches. Variants of instantiations of collaborative filtering approach suffer from the common issues of so called "cold start" and "long tail" problems where there is not much user interaction data to reveal user opinions or affinities on the content and also the distortion towards the popular content. Content-based approaches are sometimes limited by the richness of the available content data resulting in a heavily biased and coarse recommendation result. In recent years, the deep neural network has enjoyed a great success in large-scale image and video recognitions. In this paper, we propose and experiment using deep convolutional neural network to imitate how human brain processes hierarchical structures in the auditory signals, such as music, speech, etc., at various timescales. This approach can be used to discover the latent factor models of the music based upon acoustic hyper-images that are extracted from the raw audio waves of music. These latent embeddings can be used either as features to feed to subsequent models, such as collaborative filtering, or to build similarity metrics between songs, or to classify music based on the labels for training such as genre, mood, sentiment, etc.

Probabilistic Classification using Fuzzy Support Vector Machines

In medical applications such as recognizing the type of a tumor as Malignant or Benign, a wrong diagnosis can be devastating. Methods like Fuzzy Support Vector Machines (FSVM) try to reduce the effect of misplaced training points by assigning a lower weight to the outliers. However, there are still uncertain points which are similar to both classes and assigning a class by the given information will cause errors. In this paper, we propose a two-phase classification method which probabilistically assigns the uncertain points to each of the classes. The proposed method is applied to the Breast Cancer Wisconsin (Diagnostic) Dataset which consists of 569 instances in 2 classes of Malignant and Benign. This method assigns certain instances to their appropriate classes with probability of one, and the uncertain instances to each of the classes with associated probabilities. Therefore, based on the degree of uncertainty, doctors can suggest further examinations before making the final diagnosis.