A goal of statistical language modeling is to learn the joint probability function of sequences of words in a language. This is intrinsically difficult because of the curse of dimensionality: a word sequence on which the model will be tested is likely to be different from all the word sequences seen during training. Traditional but very successful approaches based on n-grams obtain generalization by concatenating very short overlapping sequences seen in the training set. We propose to fight the curse of dimensionality by learning a distributed representation for words which allows each training sentence to inform the model about an exponential number of semantically neighboring sentences. The model learns simultaneously (1) a distributed representation for each word along with (2) the probability function for word sequences, expressed in terms of these representations. Generalization is obtained because a sequence of words that has never been seen before gets high probability if it is made of words that are similar (in the sense of having a nearby representation) to words forming an already seen sentence. Training such large models (with millions of parameters) within a reasonable time is itself a significant challenge. We report on experiments using neural networks for the probability function, showing on two text corpora that the proposed approach significantly improves on state-of-the-art n-gram models, and that the proposed approach allows to take advantage of longer contexts.

This paper reports on the benefits of largescale statistical language modeling in machine translation. A distributed infrastructure is proposed which we use to train on up to 2 trillion tokens, resulting in language models having up to 300 billion n-grams. It is capable of providing smoothed probabilities for fast, single-pass decoding. We introduce a new smoothing method, dubbed Stupid Backoff, that is inexpensive to train on large data sets and approaches the quality of Kneser-Ney Smoothing as the amount of training data increases. 1

We propose a new hierarchical Bayesian n-gram model of natural languages. Our model makes use of a generalization of the commonly used Dirichlet distributions called Pitman-Yor processes which produce power-law distributions more closely resembling those in natural languages. We show that an approximation to the hierarchical Pitman-Yor language model recovers the exact formulation of interpolated Kneser-Ney, one of the best smoothing methods for n-gram language models. Experiments verify that our model gives cross entropy results superior to interpolated Kneser-Ney and comparable to modified Kneser-Ney. 1

Abstract. We present a prototype system, code-named Pulse, for mining topics and sentiment orientation jointly from free text customer feedback. We describe the application of the prototype system to a database of car reviews. Pulse enables the exploration of large quantities of customer free text. The user can examine customer opinion “at a glance ” or explore the data at a finer level of detail. We describe a simple but effective technique for clustering sentences, the application of a bootstrapping approach to sentiment classification, and a novel user-interface. 1

Many NLP tasks rely on accurately estimating word dependency probabilities P(w 1 |w 2 ), where the words w 1 and w 2 have a particular relationship (such as verb-object). Because of the sparseness of counts of such dependencies, smoothing and the ability to use multiple sources of knowledge are important challenges. For example, if the probability P(N ) of noun N being the subject of verb V is high, and V takes similar objects to V # , and V # is synonymous to V ## , then we want to conclude that P(N ## ) should also be reasonably high---even when those words did not cooccur in the training data. To capture

Abstract not found

We propose the first unsupervised approach to the problem of modeling dialogue acts in an open domain. Trained on a corpus of noisy Twitter conversations, our method discovers dialogue acts by clustering raw utterances. Because it accounts for the sequential behaviour of these acts, the learned model can provide insight into the shape of communication in a new medium. We address the challenge of evaluating the emergent model with a qualitative visualization and an intrinsic conversation ordering task. This work is inspired by a corpus of 1.3 million Twitter conversations, which will be made publicly available. This huge amount of data, available only because Twitter blurs the line between chatting and publishing, highlights the need to be able to adapt quickly to a new medium. 1

We discuss different strategies for smoothing the phrasetable in Statistical MT, and give results over a range of translation settings. We show that any type of smoothing is a better idea than the relativefrequency estimates that are often used. The best smoothing techniques yield consistent gains of approximately 1 % (absolute) according to the BLEU metric. 1

This paper investigates why the HMMs estimated by Expectation-Maximization (EM) produce such poor results as Part-of-Speech (POS) taggers. We find that the HMMs estimated by EM generally assign a roughly equal number of word tokens to each hidden state, while the empirical distribution of tokens to POS tags is highly skewed. This motivates a Bayesian approach using a sparse prior to bias the estimator toward such a skewed distribution. We investigate Gibbs Sampling (GS) and Variational Bayes (VB) estimators and show that VB converges faster than GS for this task and that VB significantly improves 1-to-1 tagging accuracy over EM. We also show that EM does nearly as well as VB when the number of hidden HMM states is dramatically reduced. We also point out the high variance in all of these estimators, and that they require many more iterations to approach convergence than usually thought. 1

Neural probabilistic language models (NPLMs) have been shown to be competitive with and occasionally superior to the widely-used n-gram language models. The main drawback of NPLMs is their extremely long training and testing times. Morin and Bengio have proposed a hierarchical language model built around a binary tree of words, which was two orders of magnitude faster than the nonhierarchical model it was based on. However, it performed considerably worse than its non-hierarchical counterpart in spite of using a word tree created using expert knowledge. We introduce a fast hierarchical language model along with a simple feature-based algorithm for automatic construction of word trees from the data. We then show that the resulting models can outperform non-hierarchical neural models as well as the best n-gram models. 1