Discriminative language modeling (DLM) has been shown to improve the accuracy of automatic speech recognition (ASR) systems, but it requires large amounts of both acoustic and text data for training. One way to overcome this is to use simulated hypotheses instead of real hypotheses for training, which is called semisupervised training. In this study, we compare six different perceptron algorith...

This paper describes experiments carried out at NiCT for the TREC 2009 Entity Ranking track. Our main study is to develop an effective approach to rank entities via measuring the “similarities” between supporting snippets of entities and input query. Three models are implemented to this end. 1) The DLM regards entity ranking as a task of calculating the probabilities of generating input query g...

Clustering in high-dimensional spaces is nowadays a recurrent problem in many scientific domains but remains a difficult task from both the clustering accuracy and the result understanding points of view. This paper presents a discriminative latent mixture (DLM) model which fits the data in a latent orthonormal discriminative subspace with an intrinsic dimension lower than the dimension of the ...

The methods for the interpretation of aerial images and maps are usually different although both describe the same landscape. The presented work shows that regarding remote sensing data and maps as different kinds of sensors allows a similar approach for both in the domain of landscape interpretation. The prior knowledge about the landscape objects is represented explicitly by semantic nets. Ba...

Great Lakes water levels have fluctuated over thousands of years. High water levels were a problem in the 1980s, but a recent sudden drop in Lakes Michigan and Huron has caused particular concern, in part because lower water levels are consistent with many global climate change scenarios. We examined water level data (1860-2006) representing Lakes Michigan and Huron to evaluate changes in both ...

In future mobile communication networks, there will be a shift towards higher carrier frequencies and highly integrated multiple antenna systems. The system performance will largely depend on the available RF hardware. As such, RF power amplifiers (PAs) with improved efficiency, linearity, and bandwidth are needed. Dynamic load modulation (DLM) is one of the most common PA efficiency enhancemen...

A disaster is a failure of resilience to an event. Mitigating the risks that a hazard will progress into a destructive event, or increasing the resilience of a society-at-risk, requires careful analysis, planning, and execution. The Disaster Logic Model (DLM) is used to define the value (effects, costs, and outcome(s)), impacts, and benefits of interventions directed at risk reduction. A Risk-R...

Cinobufagin (CB), a major bioactive component of the traditional Chinese medicine Chansu, has been reported to have potent antitumor activity. In this study, in vitro metabolism of CB among species was compared with respect to metabolic profiles, enzymes involved, and catalytic efficiency by using liver microsomes from human (HLM), mouse (MLM), rat (RLM), dog (DLM), minipig (PLM), and monkey (C...

In the present study, N-(alpha-methylbenzyl-)-1-aminobenzotriazole (MBA) and ketoconazole (KET) were identified as the inhibitors with selectivity toward dog CYP2B11 and CYP3A12, respectively. Their selectivity was evaluated using phenacetin O-deethylation (CYP1A), diazepam (DZ) N1-demethylation (CYP2B11), diclofenac 4'-hydrxylation (CYP2C21), bufuralol 1'-hydroxylation (CYP2D11), and DZ C3-hyd...