著者
Yuanzhi Ke Masafumi Hagiwara
出版者
The Japanese Society for Artificial Intelligence
雑誌
人工知能学会論文誌 (ISSN:13460714)
巻号頁・発行日
vol.33, no.4, pp.D-I23-1-8, 2018-07-01 (Released:2018-07-02)
参考文献数
25
被引用文献数
1

Although word embeddings are powerful, weakness on rare words, unknown words and issues of large vocabulary motivated people to explore alternative representations. While the character embeddings have been successful for alphabetical languages, Japanese is difficult to be processed at the character level as well because of the large vocabulary of kanji, written in the Chinese characters. In order to achieve fewer parameters and better generalization on infrequent words and characters, we proposed a model that encodes Japanese texts from the radical-level representation, inspired by the experimental findings in the field of psycholinguistics. The proposed model is comprised of a convolutional local encoder and a recurrent global encoder. For the convolutional encoder, we propose a novel combination of two kinds of convolutional filters of different strides in one layer to extract information from the different levels. We compare the proposed radical-level model with the state-of-the-art word and character embedding-based models in the sentiment classification task. The proposed model outperformed the state-of-the-art models for the randomly sampled texts and the texts that contain unknown characters, with 91% and 12% fewer parameters than the word embedding-based and character embedding-based models, respectively. Especially for the test sets of unknown characters, the results by the proposed model were 4.01% and 2.38% above the word embedding-based and character embedding-based baselines, respectively. The proposed model is powerful with cheaper computational and storage cost, can be used for devices with limited storage and to process texts of rare characters.
著者
Ming YANG Masafumi HAGIWARA
出版者
Japan Society of Kansei Engineering
雑誌
International Journal of Affective Engineering (ISSN:21875413)
巻号頁・発行日
vol.15, no.2, pp.125-134, 2016 (Released:2016-06-30)
参考文献数
29
被引用文献数
1

In this paper, we propose an automatic Waka generation system with custom database, based on texts given by the user. The proposed text-based system has better compatibility with Waka poem, improving consistence and logicality of generated poems. Kansei information is also considered to make poems natural and closer to the emotions the user wants to express. Presented by interactive generation experiments, the proposed system can generate Waka poems reflecting stylistic and grammatical requirements. Meanwhile, the poems are also with related meanings and emotions to the original text and some poeticness.
著者
Maho HOTOGI Masafumi HAGIWARA
出版者
日本感性工学会
雑誌
International Journal of Affective Engineering (ISSN:21875413)
巻号頁・発行日
pp.IJAE-D-14-32, (Released:2015-07-21)
参考文献数
39
被引用文献数
6

In this paper, first, analyses of local community mascot characters are carried out to obtain knowledge to be the popular ones. Next a mascot characters automatic creation system is proposed using these findings. As for the analyses, we used 200 mascot characters. Many interesting findings could be obtained such that dark-round eyes tend to contribute to be popular as a mascot character. The proposed system utilizing these findings can create a mascot character reflecting a user's image inputted by affective words. Many parts having degrees of affective images are prepared in the proposed system and are combined to form a character using Rough Sets Theory and an Interactive Genetic Algorithm framework. We performed evaluation experiments. Many mascot characters satisfying user's image were created and remarkable results were obtained through subjective evaluations. For example, by using the extracted rules from the analyses and Interactive Genetic Algorithm, the proposed system can create much favorable mascot characters. Moreover, we found some interesting tendencies of the color used to paint characters. One of them is that gentle characters are often painted in colors with low saturation.