Browsing by Author "Chang, Ching-Chih"

Now showing 1 - 4 of 4

Development of a Multisite, Closed-loop Neuromodulator for the Theranosis of Neural Degenerative Diseases
(2018-01-01)
Chen, Hsin
;
Chang, Yen-Chung
;
Yeh, Shih-Rung
;
Hsieh, Chih-Cheng
;
Tang, Kea-Tiong
Stimulating specific brain regions has been found useful for treating neural disorders, such as the Parkinson's disease, epilepsy, depression, etc. However, how electrical stimulation modulates neural activities remains not fully understood. As animal models provide the advantage of recording and stimulating different disease-related regions simultaneously, this paper introduces the latest development of a multisite, closed-loop-controlled microsystem for investigating novel treatments on neural degenerative diseases with freely-moving rats. The algorithms for recognizing pathological neural activities automatically are also developed and realized in hardware, so as to control the stimulation in a closed loop and in real time. The pilot studies on the efficacy of treating the Parkinson's disease with closed-loop-controlled stimulation will be presented and discussed. Finally, the feasibility of modelling and probing how neural dynamics and connectivity are modulated by stimulation will be an important topic for future research.
Effects of interfacial sulfidization and thermal annealing on the electrical properties of an atomic-layer-deposited Al(2)O(3) gate dielectric on GaAs substrate
(2008-04-01)
Cheng, Chao-Ching
;
Chien, Chao-Hsin
;
Luo, Guang-Li
;
Yang, Chun-Hui
;
Chang, Ching-Chih
In this study we investigated the interfacial chemistry occurring between an atomic-layer-deposited Al(2)O(3) high-k film and a GaAs substrate and the impact of sulfidization and thermal annealing on the properties of the resultant capacitor. We observed that sulfide passivation of the Al(2)O(3)/GaAs structure improved the effect of Fermi level pinning on the electrical characteristics, thereby providing a higher oxide capacitance, smaller frequency dispersion, and reduced surface states, as well as decreased interfacial charge trapping and gate leakage currents. Photoemission analysis indicated that the (NH(4))(2)S-treated GaAs improved the quality of the as-deposited Al(2)O(3) thin film and preserved the stoichiometry of the dielectric during subsequent high-temperature annealing. This behavior was closely correlated to the diminution of GaAs native oxides and elemental arsenic defects and their unwanted diffusion. In addition, thermal processing under an O(2) atmosphere, relative to that under N(2), decreased the thickness of the Al(2)O(3) gate dielectric and relieved the gate leakage degradation induced by metallic arsenic; as a result, superior dielectric reliability was attained. We discuss the underlying thermochemical reactions that account for these experimental observations. (C) 2008 American Institute of Physics.
Electrical and material characterization of atomic-layer-deposited Al(2)O(3) gate dielectric on ammonium sulfide treated GaAs substrates
(2008)
Cheng, Chao-Ching
;
Chien, Chao-Hsin
;
Luo, Guang-Li
;
Chang, Ching-Chih
;
Kei, Chi-Chung
This paper has studied the electrical and interfacial properties of atomic-layerdeposited Al(2)O(3) thin film on ammonium-sulfide passivated GaAs. It was found that the Al(2)O(3) deposited at 300 degrees C relative to that at 100 degrees C showed the nearly four orders of magnitude reduction in gate leakage current at the capacitance-equivalent-thickness of 40 angstrom. The capacitance-voltage (C-V) characteristics displayed the higher oxide capacitance, reduced frequency dispersion and less charge trapping when GaAs receiving (NH(4))(2)S sulfide immersion; these improvements can be reasonably explained by the suppression of both native oxides and the resultant improved interface quality. Annealing as-deposited Al(2)O(3)/GaAs structures at high temperatures further reduces the Fermi level pinning effect on accumulation capacitance, however, causes an increase in C-V frequency dispersion and gate leakage current. We suggested that these phenomena are strongly associated to the amount of As-related defects resided at the dielectric/ substrate interface during thermal desorption.
Multi-Source Learning for Sales Prediction
(2017-01-01)
Tsai, Kuen-Han
;
Wang, Yau-Shian
;
Kuo, Hsuan-Yu
;
Tsai, Jui-Yi
;
Chang, Ching-Chih
With the convenience and popularity of Internet, the sales on e-commerce platforms have grown significantly. This exponential growth generates massive chunks of data, which provides the opportunity to utilize the historical data for predicting the customers' behaviors and can thus offer better services. One of the major tasks is to correctly estimate the coming sales of products since the e-retailers can reserve the products in a smart way. However, even with massive data, it is still challenging to estimate the sale amount of each product due to 1) the complicated language structures, 2) difficulties in integrating the features, and 3) missing values. To address these issues, we propose a framework for predicting the sales, which contains two phases: 1) feature extraction from product attributes and reviews, and 2) tensor decomposition for multi-source learning. The experimental results show that our framework outperforms baselines by 73%.