DNA microarray provides a powerful basis for analysis of gene expression. Data mining methods such as clustering have been widely applied to microarray data to link genes that show similar expression patterns. However, this approach usually fails to unveil gene-gene interactions in the same cluster. Association rule mining and loglinear models have been used for this purpose, but their inherent...

Linear Discriminant Analysis (LDA) has been established as an important means for dimension reduction and decorrelation in speech recognition. The major points of criticism of LDA are that it uses an ad hoc and non-discriminative training criterion, and that the estimation is performed in a separate preprocessing step. This paper presents a new discriminative training method for the estimation ...

Experimental solubilities of four antiepileptic drugs, that is, lamotrigine, diazepam, clonazepam, and phenobarbital in propylene glycol + water mixtures at 298.15 K were reported. The solubility of drugs was increased with the addition of propylene glycol, and the maximum values are observed in neat propylene glycol. The Jouyban-Acree model was used to fit the experimental data, and the solubi...

String-to-string transduction is a central problem in computational linguistics and natural language processing. It occurs in tasks as diverse as name transliteration, spelling correction, pronunciation modeling and inflectional morphology. We present a conditional loglinear model for string-to-string transduction, which employs overlapping features over latent alignment sequences, and which le...

Following a brief presentation of marine catch trends in Southeast Asia, some biological peculiarities of the stocks upon which these fisheries rely are discussed. Two empirical loglinear models are presented allowing rough estimation of potential yield of small pelagic fishes fish from primary production, and of demersal fish from mean water depth and primary production. These models are appli...

Given a subset of data that differs from the rest, a user often wants an explanation as to why this is the case. For instance, in a database of flights, a user may want to understand why certain flights were very late. This paper presents ESCAPE, a sequential covering algorithm designed to generate explanations of subsets that take the form of disjunctive normal rules describing the characteris...

Randomized comparative trials are often used to assess the relative merits of two or more interventions aimed at having beneficial effects on the incidence of categorical outcomes. In simple applications chisquare tests can be used to analyze contrasts among proportions of incident events or risk ratios (relative risks). However, assessment of intervention differences may be obscured by outcome...

This technical report describes our statistical method for deriving an initial version of the domain model directly from textual specification written in natural language. The main idea of the method is to exploit the intrinsic relations between textual specifications and their domain models. To do this, an in-depth linguistic analysis on the text using the state-of-the-art NLP tools (based on ...