著者

Na WANG Xianglian ZHAO

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences (ISSN:09168508)

巻号頁・発行日

vol.E106-A, no.6, pp.886-895, 2023-06-01

---

The application of time-series prediction is very extensive, and it is an important problem across many fields, such as stock prediction, sales prediction, and loan prediction and so on, which play a great value in production and life. It requires that the model can effectively capture the long-term feature dependence between the output and input. Recent studies show that Transformer can improve the prediction ability of time-series. However, Transformer has some problems that make it unable to be directly applied to time-series prediction, such as: (1) Local agnosticism: Self-attention in Transformer is not sensitive to short-term feature dependence, which leads to model anomalies in time-series; (2) Memory bottleneck: The spatial complexity of regular transformation increases twice with the sequence length, making direct modeling of long time-series infeasible. In order to solve these problems, this paper designs an efficient model for long time-series prediction. It is a double pyramid bidirectional feature fusion mechanism network with parallel Temporal Convolution Network (TCN) and FastFormer. This network structure can combine the time series fine-grained information captured by the Temporal Convolution Network with the global interactive information captured by FastFormer, it can well handle the time series prediction problem.

著者

Guo-dong Xu Lei Cai Yi-shu Ni Shi-yi Tian Ying-qi Lu Li-na Wang Lian-lian Chen Wen-ya Ma Shao-ping Deng

出版者

The Pharmaceutical Society of Japan

雑誌

Biological and Pharmaceutical Bulletin (ISSN:09186158)

巻号頁・発行日

vol.41, no.7, pp.1024-1033, 2018-07-01 (Released:2018-07-01)

参考文献数

54

被引用文献数

16

---

Acarbose and voglibose are the most widely used diabetes drugs as glycosidase inhibitors. In this study, the use of these two inhibitors significantly increased the content of starch in large intestine, and altered the concentration of short-chain fatty acids (SCFAs) by affecting the intestinal microbiota. However, there are some differences in the intestinal microbiome of the two groups of mice, mainly in bacteria such as Bacteroidaceae bacteroides and Desulfovibrionaceae desulfovibrio. The productions of acetate and propionate in caecum in voglibose group were significantly higher than those in acarbose group and two kinds of glycosidase inhibitors were close in the production of butyrate in caecum. The Tax4Fun analysis based on Kyoto Encyclopedia of Genes and Genomes (KEGG) data indicated that different productions of acetate and propionate between acarbose group and voglibose group may be related to 2-oxoisovalerate dehydrogenase and pyruvate oxidase. In addition, in-vitro experiments suggested that voglibose had less effect on epithelial cells than acarbose after direct stimulation. According to the recent researches of SCFAs produced by intestinal microbiota, our comparative study shown higher concentration of these beneficial fatty acids in the lumen of voglibose-treated mice, which implied a lower level of inflammation.