This paper describes a dependency structure analysis of Japanese sentences based on the maximum entropy models. Our model is created by learning the weights of some features from a training corpus to predict the dependency between bunsetsus or phrasal units. The dependency accuracy of our system is 87.2% using the Kyoto University corpus. We discuss the contribution of each feature set and the relationship between the number of training data and the accuracy.

This paper describes a method of detecting grammatical and lexical errors made by Japanese learners of English and other techniques that improve the accuracy of error detection with a limited amount of training data. In this paper, we demonstrate to what extent the proposed methods hold promise by conducting experiments using our learner corpus, which contains information on learners' errors.

In this paper, we present a discriminative word-character hybrid model for joint Chinese word segmentation and POS tagging. Our word-character hybrid model offers high performance since it can handle both known and unknown words. We describe our strategies that yield good balance for learning the characteristics of known and unknown words and propose an errordriven policy that delivers such balance by acquiring examples of unknown words from particular errors in a training corpus. We describe an efficient framework for training our model based on the Margin Infused Relaxed Algorithm (MIRA), evaluate our approach on the Penn Chinese Treebank, and show that it achieves superior performance compared to the state-ofthe-art approaches reported in the literature.

No abstract

Mechanisms controlling the dissolved iron distribution in the North Pacific are investigated using the Biogeochemical Elemental Cycling (BEC) model with a resolution of approximately 1° in latitude and longitude and 60 vertical levels. The model is able to reproduce the general distribution of iron as revealed in available field data: surface concentrations are generally below 0.2 nM; concentrations increase with depth; and values in the lower pycnocline are especially high in the northwestern Pacific and off the coast of California. Sensitivity experiments changing scavenging regimes and external iron sources indicate that lateral transport of sedimentary iron from continental margins into the open ocean causes the high concentrations in these regions. This offshore penetration only appears under a scavenging regime where iron has a relatively long residence time at high concentrations, namely, the order of years. Sedimentary iron is intensively supplied around continental margins, resulting in locally high concentrations; the residence time with respect to scavenging determines the horizontal scale of elevated iron concentrations. Budget analysis for iron reveals the processes by which sedimentary iron is transported to the open ocean. Horizontal mixing transports sedimentary iron from the boundary into alongshore currents, which then carry high iron concentrations into the open ocean in regions where the alongshore currents separate from the coast, most prominently in the northwestern Pacific and off of California.