Manabu Sassano

An Empirical Study of Active Learning with Support Vector Machines forJapanese Word Segmentation

We explore how active learning with Support Vector Machines works well for a non-trivial task in natural language processing. We use Japanese word segmentation as a test case. In particular, we discuss how the size of a pool affects the learning curve. It is found that in the early stage of training with a larger pool, more labeled examples are required to achieve a given level of accuracy than those with a smaller pool. In addition, we propose a novel technique to use a large number of unlabeled examples effectively by adding them gradually to a pool. The experimental results show that our technique requires less labeled examples than those with the technique in previous research. To achieve 97.0% accuracy, the proposed technique needs 59.3% of labeled examples that are required when using the previous technique and only 17.4% of labeled examples with random sampling.

Linear-time dependency analysis for Japanese

We present a novel algorithm for Japanese dependency analysis. The algorithm allows us to analyze dependency structures of a sentence in linear-time while keeping a state-of-the-art accuracy. In this paper, we show a formal description of the algorithm and discuss it theoretically with respect to time complexity. In addition, we evaluate its efficiency and performance empirically against the Kyoto University Corpus. The proposed algorithm with improved models for dependency yields the best accuracy in the previously published results on the Kyoto University Corpus.

Named entity chunking techniques in supervised learning for Japanese named entity recognition

This paper focuses on the issue of named entity chunking in Japanese named entity recognition. We apply the supervised decision list learning method to Japanese named entity recognition. We also investigate and incorporate several named-entity noun phrase chunking techniques and experimentally evaluate and compare their performance. In addition, we propose a method for incorporating richer contextual information as well as patterns of constituent morphemes within a named entity, which have not been considered in previous research, and show that the proposed method outperforms these previous approaches.

Virtual examples for text classification with Support Vector Machines

We explore how virtual examples (artificially created examples) improve performance of text classification with Support Vector Machines (SVMs). We propose techniques to create virtual examples for text classification based on the assumption that the category of a document is unchanged even if a small number of words are added or deleted. We evaluate the proposed methods by Reuters-21758 test set collection. Experimental results show virtual examples improve the performance of text classification with SVMs, especially for small training sets.

Learning with multiple stacking for named entity recognition

In this paper, we present a learning method using multiple stacking for named entity recognition. In order to take into account the tags of the surrounding words, we propose a method which employs stacked learners using the tags predicted by the lower level learners. We have applied this approach to the CoNLL-2002 shared task to improve a base system.

Combining Outputs of Multiple Japanese Named Entity Chunkers by Stacking

In this paper, we propose a method for learning a classifier which combines outputs of more than one Japanese named entity extractors. The proposed combination method belongs to the family of stacked generalizers, which is in principle a technique of combining outputs of several classifiers at the first stage by learning a second stage classifier to combine those outputs at the first stage. Individual models to be combined are based on maximum entropy models, one of which always considers surrounding contexts of a fixed length, while the other considers those of variable lengths according to the number of constituent morphemes of named entities. As an algorithm for learning the second stage classifier, we employ a decision list learning method. Experimental evaluation shows that the proposed method achieves improvement over the best known results with Japanese named entity extractors based on maximum entropy models.

Prediction of Prospective User Engagement with Intelligent Assistants

Intelligent assistants on mobile devices, such as Siri, have recently gained considerable attention as novel applications of dialogue technologies. A tremendous amount of real users of intelligent assistants provide us with an opportunity to explore a novel task of predicting whether users will continually use their intelligent assistants in the future. We developed prediction models of prospective user engagement by using large-scale user logs obtained from a commercial intelligent assistant. Experiments demonstrated that our models can predict prospective user engagement reasonably well, and outperforms a strong baseline that makes prediction based past utterance frequency.

Using a partially annotated corpus to build a dependency parser for japanese

We explore the use of a partially annotated corpus to build a dependency parser for Japanese. We examine two types of partially annotated corpora. It is found that a parser trained with a corpus that does not have any grammatical tags for words can demonstrate an accuracy of 87.38%, which is comparable to the current state-of-the-art accuracy on the Kyoto University Corpus. In contrast, a parser trained with a corpus that has only dependency annotations for each two adjacent bunsetsus (chunks) shows moderate performance. Nonetheless, it is notable that features based on character n-grams are found very useful for a dependency parser for Japanese.