Main aim of this paper is to present some notions on how results from commutative algebra and algebraic geometry could be used in representation and computation of maximum and minimum entropy (ME) models. We show that various formulations of estimation of ME models can be transformed to solving systems of polynomial equations in cases where an integer valued sufficient statistic exists. We give...

This paper describes the system we developed for PSB 2016 social media mining shared task on binary classification of adverse drug reactions (ADRs). The task focuses on automatic classification of ADR assertive user posts. We propose a weighted average ensemble of four classifiers: (1) a concept-matching classifier based on ADR lexicon; (2) a maximum entropy (ME) classifier with word-level n-gr...

In this paper, we present an approach to extract most relevant information from a (semi-)quantitative knowledge base, e.g., from a probability distribution. Relevance here is meant with respect to some appropriate inductive inference process, like maximum entropy inference (ME-inference) in probabilistics. So in particular, the method developed in this paper is apt to solve the inverse maxent p...

The maximum entropy (ME) model of a knowledge base R consisting of relational probabilistic conditionals can be defined referring to the set of all ground instances of the conditionals. The logic FO-PCL employs the notion of parametric uniformity for avoiding the full grounding of R. We present an implementation of a rule system transforming R into a knowledge base that is parametrically unifor...

This paper presents a maximum entropy based Chinese named entity recognizer (NER): Mencius. It aims to address Chinese NER problems by combining the advantages of rule-based and machine learning (ML) based NER systems. Rule-based NER systems can explicitly encode human comprehension and can be tuned conveniently, while ML-based systems are robust, portable and inexpensive to develop. Our hybrid...

We address the problem of predicting pauses between the words in a sentence, which is of considerable interest for text to speech systems. In doing so, we show that the performance of both a generative classifier (naive Bayes, NB) and a discriminative classifier (maximum entropy, ME) can be significantly enhanced by application of the generalised probabilistic descent (GPD) algorithm. The featu...

The Maximum Entropy principle (ME) is an appropriate framework for combining information of a diverse nature from several sources into the same language model. In order to incorporate long-distance information into the ME framework in a language model, a Whole Sentence Maximum Entropy Language Model (WSME) could be used. Until now MonteCarlo Markov Chains (MCMC) sampling techniques has been use...

K e y w o r d s C o n v e x hull, Entropy, Hankel matrix, Moment problem. 1. I N T R O D U C T I O N Every probability distribution has some uncertainty associated with it, and its entropy provides a quantitat ive measure of this uncertainty. Partial information given, for instance, in terms of averages about a random variate decreases its entropy. It thus appears interesting to provide an entr...

in this paper, we consider the determination methods of maximum entropy multivariate distributions with given prior under the constraints, that the marginal distributions or the marginals and covariance matrix are prescribed. next, some numerical solutions are considered for the cases of unavailable closed form of solutions. finally, these methods are illustrated via some numerical examples.