Learn More
In this paper, we propose a new Bayesian model for fully unsupervised word seg-mentation and an efficient blocked Gibbs sampler combined with dynamic programming for inference. Our model is a nested hierarchical Pitman-Yor language model, where Pitman-Yor spelling model is embedded in the word model. We confirmed that it significantly outperforms previous(More)
We present a nonparametric Bayesian method of estimating variable order Markov processes up to a theoretically infinite order. By extending a stick-breaking prior, which is usually defined on a unit interval, " vertically " to the trees of infinite depth associated with a hierarchical Chinese restaurant process, our model directly infers the hidden orders(More)
For a prediction problem for high-dimensional discrete sequences, we propose a solution using online change point analysis by Particle Filters combined with probabilistic text models LDA and DM. Problem: How long context should we use? – Hidden state of multinomial distributions may differ – Beyond " Bag of Words " assumption Dirichlet Dirichlet Mixtures:(More)
Human gaze behavior while reading text reflects a variety of strategies for precise and efficient reading. Nevertheless, the possibility of extracting and importing these strategies from gaze data into natural language processing technologies has not been explored to any extent. In this research, as a first step in this investigation, we examine the(More)
This paper describes the NiCT-ATR statistical machine translation (SMT) system used for the IWSLT 2006 evaluation compaign. We participated in all four language pair translation tasks (CE, JE, AE and IE) and all two tracks (OPEN and CSTAR). We used a phrase-based SMT in the OPEN track and a hybrid multiple translation engine in the CSTAR track. We also(More)
Our goal is to characterize expressive dynamic components of the singing voice fundamental frequency (F 0) contours, such as Vibrato and Portamento, using a stochastic model. We propose a process of generating the F 0 contours and a statistical framework of the model parameter estimation. Experimental results show that our method successfully extracts the(More)
This paper presents a new class of tensor fac-torization called positive semidefinite tensor factorization (PSDTF) that decomposes a set of positive semidefinite (PSD) matrices into the convex combinations of fewer PSD basis matrices. PSDTF can be viewed as a natural extension of nonnegative matrix factoriza-tion. One of the main problems of PSDTF is that(More)