Current recommender systems exploit user and item similarities by collaborative filtering. Some advanced methods also consider the temporal evolution of item ratings as a global background process. However, all prior methods disregard the individual evolution of a user's experience level and how this is expressed in the user's writing in a review community. In this paper, we model the joint evolution of user experience, interest in specific item facets, writing style, and rating behavior. This way we can generate individual recommendations that take into account the user's maturity level (e.g., recommending art movies rather than blockbusters for a cinematography expert). As only item ratings and review texts are observables, we capture the user's experience and interests in a latent model learned from her reviews, vocabulary and writing style. We develop a generative HMM-LDA model to trace user evolution, where the Hidden Markov Model (HMM) traces her latent experience progressing over time -- with solely user reviews and ratings as observables over time. The facets of a user's interest are drawn from a Latent Dirichlet Allocation (LDA) model derived from her reviews, as a function of her (again latent) experience level. In experiments with five real-world datasets, we show that our model improves the rating prediction over state-of-the-art baselines, by a substantial margin. We also show, in a use-case study, that our model performs well in the assessment of user experience levels.
Modelling the real world complexity of music is a challenge for machine learning. We address the task of modeling melodic sequences from the same music genre. We perform a comparative analysis of two probabilistic models; a Dirichlet Variable Length Markov Model (Dirichlet-VMM) and a Time Convolutional Restricted Boltzmann Machine (TC-RBM). We show that the TC-RBM learns descriptive music features, such as underlying chords and typical melody transitions and dynamics. We assess the models for future prediction and compare their performance to a VMM, which is the current state of the art in melody generation. We show that both models perform significantly better than the VMM, with the Dirichlet-VMM marginally outperforming the TC-RBM. Finally, we evaluate the short order statistics of the models, using the Kullback-Leibler divergence between test sequences and model samples, and show that our proposed methods match the statistics of the music genre significantly better than the VMM.
We describe a computer system that provides a real-time musical accompanimentfor a live soloist in a piece of non-improvised music for soloist and accompaniment. A Bayesian network is developed thatrepresents the joint distribution on the times at which the solo and accompaniment notes are played, relating the two parts through a layer of hidden variables. The network is first constructed usingthe rhythmic information contained in the musical score. The network is then trained to capture the musical interpretations ofthe soloist and accompanist in an off-line rehearsal phase. During live accompaniment the learned distribution of the network is combined with a real-time analysis of the soloist's acoustic signal, performedwith a hidden Markov model, to generate a musically principledaccompaniment that respects all available sources of knowledge. A live demonstration will be provided.
We present a description of two small audio/visual immersive installations. The main framework is an interactive structure that enables multiple participants to generate jazz improvisations, loosely speaking. The first uses a Bayesian Network to respond to sung or played pitches with machine pitches, in a kind of constrained harmonic way. The second uses Bayesian Networks and Hidden Markov Models to track human motion, play reactive chords, and to respond to pitches both aurally and visually.
We believe that collaborative filtering is well described by a probabilistic model in which people and the items they view or buy are each divided into (unknown) clusters and there are link probabilities between these clusters. EM is an obvious method for estimating these models, but does not work because it cannot be efficiently constructed to recognize the constraint that a movie liked by two different people must be in the same movie class each time. K-means clustering is fast but ad hoc. Repeated clustering using K-means clustering or a "soft clustering" version of K-means may be useful, but usually does not improve accuracy. Clustering movies or people on other relevant attributes can help - and does help for the case of CD purchase data. Gibbs sampling works well and has the virtue of being easily extended to much more complex models, but is computationally expensive. We are currently developing more efficient Gibbs sampling methods for collaborative filtering problems, extending our repeated clustering and Gibbs sampling code to incorporate multiple attributes, and applying them to more real data sets.