Browsing by Issue Date, starting with "2018-02-22"

Now showing 1 - 20 of 11281

Latent Dirichlet mixture model
(2018-02-22)
Chien, Jen-Tzung
;
Lee, Chao-Hsi
;
Tan, Zheng-Hua
;
電機工程學系
;
Department of Electrical and Computer Engineering
Text representation based on latent topic model is seen as a non-Gaussian problem where the observed words and latent topics are multinomial variables and the topic proportionals are Dirichlet variables. Traditional topic model is established by introducing a single Dirichlet prior to characterize the topic proportionals. The words in a text document are represented by a random mixture of semantic topics. However, in real world, a single Dirichlet distribution may not faithfully reflect the variations of topic proportionals estimated from the heterogeneous documents. To address these variations, we propose a new latent variable model where latent topics and their proportionals are learned by incorporating the prior based on Dirichlet mixture model. The resulting latent Dirichlet mixture model (LDMM) is constructed for topic clustering as well as document clustering. Multiple Dirichlets provide a solution to build structural latent variables in learning representation over a variety of topics. This study carries out the inference for LDMM according to the variational Bayes and the collapsed variational Bayes. Such an unsupervised LDMM is further extended to a supervised LDMM for text classification. Experiments on document representation, summarization and classification show the merit of structural prior in LDMM topic models. (C) 2017 Elsevier B. V. All rights reserved.
Recent advances in machine learning for non-Gaussian data processing
(2018-02-22)
Ma, Zhanyu
;
Chien, Jen-Tzung
;
Tan, Zheng-Hua
;
Song, Yi-Zhe
;
Taghia, Jalil
WATER-SOLUBLE PEPTIDE FLUORESCENCE MATERIAL
(2018-02-22)
Hsin-Chieh LIN
;
Shu-Min HSU
A water-soluble peptide fluorescence material having a structure of formula (I):$$$$In formula (I), n is an integer greater than or equal to 1, R1 is independently selected from hydrogen or a nitrogen-containing functional group, R2 is independently selected from hydrogen or alkyl, and A1 is polymerized by at least one amino acid monomer and having a structure of formula (II):$$$$In formula (II), m is an integer greater than or equal to 1, and R3 in each, of the amino acid monomers of A1 is independently selected from hydrogen, alkyl, aralkyl, alkylthioaalkyl, hydroxyaralky, heteroaralkyl, carboxylalkyl, or guanidinylalkyl, A2 is —OR5 or —N(R4)2, and R4 is independently selected from hydrogen, alkyl, aralkyl, alkylthioaalkyl, hydroxyaralky, heteroaralkyl, carboxylalkyl, guanidinylalkyl, monoglycosyl, biglycosyl, or oligosaccharyl, and R5 is hydrogen, alkyl, aralkyl, alkylthioaalkyl, hydroxyaralky, heteroaralkyl, carboxylalkyl, or guanidinylalkyl.
The deformation behavior and fracture toughness of single crystal YSZ(111) by indentation
(2018-02-25)
Yen, Cheng-Yo
;
Jian, Sheng-Rui
;
Tseng, Yu-Chin
;
Juang, Jenh-Yih
;
電子物理學系
The nano-scale deformation behaviors and indentation-induced fracture features of the (111)-oriented yttrium-stabilized zirconia single crystal (YSZ(111)) were investigated by Berkovich and micro-Vicker indentations, respectively. The load-displacement curves in the Berkovich nano-indentation experiments evidently exhibited indentation-induced single "pop-in" phenomenon during loading, indicating that the nano-scale deformation in the YSZ(111) crystals is due primarily to the activities of dislocation nucleation and propagation. Based on this scenario, the number of nanoindentation-induced dislocation loops giving rise to the pop-in event was estimated to be around 2 x 10(5) with a critical radius of similar to 2 nm. The hardness and Young's modulus of YSZ(111) single crystal obtained by the continuous contact stiffness measurements (CSM) mode were 22.3 +/- 1.1 GPa and 270.6 +/- 8.5 GPa, consistent with those reported previously in the literature. In addition, the fracture toughness of YSZ(111) single crystal was estimated to be about 1.4-1.6 MPa m(1/2). (C) 2017 Elsevier B.V. All rights reserved.
Enhanced immersion cooling using two-tier micro- and nano-structures
(2018-02-25)
Hsu, Ya-Tzu
;
Li, Jia-Xiong
;
Lu, Ming-Chang
;
機械工程學系
;
Department of Mechanical Engineering
Continual increases in the functionality and miniaturization of electronic devices have resulted in a rapid increase in the power density of such devices. Thus, an efficient cooling technology is required to maximize heat dissipation and prevent electronic failure. Immersion cooling is a promising technique for the thermal management of high-power-density electronics. However, common working fluids in immersion cooling have high global warming potential, and the heat transfer performance of immersion cooling requires improvement to achieve efficient cooling of state-of-the-art high-power-density electronics. In this study, Novec 649, which has low global warming potential and a low boiling point, was applied as a working fluid for immersion cooling. A Si nanowire (SiNW) array, Si micropillar (SiMP) array, and Si nanowires on a Si micropillar (SiNW/MP) two-tier structure were employed to enhance boiling performance. The SiMP surface exhibited the highest critical heat flux (CHF) of 23.5 +/- 1.3 W/cm(2), whereas the SiNW surface exhibited the lowest CHF but a relatively high heat transfer coefficient (HTC). The SiNW/MP surface exhibited the highest HTC of 23611.7 +/- 1586.2 W/m(2) K and a relatively large CHF of 17.4 +/- 1.2 W/cm(2). Compared with a plain SiO2 surface, the CHF and HTC of the SiNW/MP two-tier structure could be enhanced by 30% and 455%, respectively. These results suggest that the SiNW/MP surface is effective for enhancing immersion cooling. (C) 2017 Elsevier Ltd. All rights reserved.
Relationship Between Earthquake b-Values and Crustal Stresses in a Young Orogenic Belt
(2018-02-28)
Wu, Yih-Min
;
Chen, Sean Kuanhsiang
;
Huang, Ting-Chung
;
Huang, Hsin-Hua
;
Chao, Wei-An
It has been reported that earthquake b-values decrease linearly with the differential stresses in the continental crust and subduction zones. Here we report a regression-derived relation between earthquake b-values and crustal stresses using the Anderson fault parameter (A) in a young orogenic belt of Taiwan. This regression relation is well established by using a large and complete earthquake catalog for Taiwan. The data set consists of b-values and A values derived from relocated earthquakes and focal mechanisms, respectively. Our results show that b-values decrease linearly with the A values at crustal depths with a high correlation coefficient of -0.9. Thus, b-values could be used as stress indicators for orogenic belts. However, the state of stress is relatively well correlated with the surface geological setting with respect to earthquake b-values in Taiwan. Temporal variations in the b-value could constitute one of the main reasons for the spatial heterogeneity of b-values. We therefore suggest that b-values could be highly sensitive to temporal stress variations.
Zinc oxide-based current spreading layer behavior on the performance of P-side-up thin-film red light emitting diodes
(2018-02-28)
Tseng, Ming-Chun
;
Wuu, Dong-Sing
;
Chen, Chi-Lu
;
Lee, Hsin-Ying
;
Horng, Ray-Hua
Transparent conductive layers deposited on a GaP window layer were used to fabricate high-brightness p-side-up thin-film AlGaInP light emitting diodes (LEDs). Zinc oxide (ZnO) and aluminum-doped zinc oxide (ZAO) with an Zn:Al cycle ratio of 20:1 were grown by atomic layer deposition for comparison. The ZAO thin films reduced the droop in the external quantum efficiencies of the LEDs as well as the junction temperature, which result in increases in the light output power and thermal stability of the LEDs. The efficiency droops of the LEDs with the ZnO and ZAO thin films were 41% and 15%, respectively. The junction temperature of the LED with the ZAO thin film can be reduced to 39.6 degrees C at an injection current of 700 mA, which is lower than that of the LED with the ZnO thin film (58.1 degrees C). The above results are promising for the development of ZAO thin films for applications in AlGaInP LEDs. (C) 2017 Elsevier B.V. All rights reserved.
Microstructure evolution determined by the crystalline phases competition in self-assembled WO3-BiVO4 hetero nanostructures
(2018-02-28)
Song, Haili
;
Li, Chao
;
Chien Nguyen Van
;
Dong, Wenxia
;
Qi, Ruijuan
A series of self-assembled WO3-BiVO4 nanostructured thin films were grown on the (001) yttria-stabilized zirconia (YSZ) substrate at the substrate temperatures of 400 degrees C, 500 degrees C, 550 degrees C, 600 degrees C, 650 degrees C and 700 degrees C by a pulsed laser deposition method. The microstructures including crystalline phases, epitaxial relationships, surface morphologies and interface structures were investigated by a combination of x-ray diffraction, scanning electron microscopy and high-resolution transmission electron microscopy. The sample grown at 400 degrees C was amorphous due to the low driving forces for nucleation and diffusion. For the samples made at 500 degrees C, 550 degrees C and 600 degrees C, the monoclinic BiVO4 matrix epitaxially grew on YSZ, forming the matrix, where the WO3 nanopillars were embedded in with a specific orientation relationship among BiVO4, WO3 and YSZ. However, in thin films deposited at 650 degrees C and 700 degrees C, the WO3 grains randomly grew on the YSZ substrate, which dominated the microstructures of the resultant thin films. Quantitative analyses of the microstructures revealed that the lateral grain sizes of BiVO4 and WO3 increased and the volume fraction of BiVO4 in the thin films decreased with the increase of the deposition temperature. A three-regime growth mechanism of the WO3-BiVO4 composite thin film was proposed based on the growth dynamics determined by the competition between BiVO4 and WO3. Published by AIP Publishing.
Ultrafast intersystem crossing for nitrophenols: ab initio nonadiabatic molecular dynamics simulation
(2018-02-28)
Xu, Chao
;
Gu, Feng Long
;
Zhu, Chaoyuan
;
應用化學系
;
應用化學系分子科學碩博班
Ultrafast intersystem crossing mechanisms for two p- and m-nitrophenol groups (PNP and MNP) have been investigated using ab initio nonadiabatic molecular dynamics simulations at the 6SA-CASSCF level of theory. Trajectory surface hopping simulation has been performed within an intersystem crossing network constructed from two low-lying singlets (S-0 and S-1) and two low-lying triplets (T-1 and T-2). It is found that the dominant relaxation S-1 -> T-2 pathway accounts for 65.4% (85.0%) of the quantum yield with a time constant of 13.4 fs (22 fs) and the S-1 -> T-2 -> S-0 pathway accounts for 33.1% (13.5%) with a time constant of 275 fs ( 375 fs) for PNP ( MNP). In comparison with the previously studied excited-state proton transfer process for ONP, the dominant relaxation S-1 -> T-2 -> T-1 pathway accounts for 49.3% with a time constant of 40 fs and the S-1 -> T-2 -> T-1 -> S-0 pathway accounts for 47.5% with a time constant of 300 fs. The relaxation mechanisms and electronic structures of the intersystem crossings are in close relation with the relative motion between the torsion motions of the nitro-group and the hydroxyl group. The present simulation provides new physical insight for understanding ultrafast photochemical intersystem crossing dynamics.
Spectroscopy of prospective interstellar ions and radicals isolated in para-hydrogen matrices
(2018-02-28)
Tsuge, Masashi
;
Tseng, Chih-Yu
;
Lee, Yuan-Pern
;
應用化學系
;
應用化學系分子科學碩博班
para-Hydrogen (p-H-2) serves as a new host in matrix-isolation experiments for an investigation of species of astrochemical interest. Protonated and mono-hydrogenated species are produced upon electron bombardment during deposition of p-H-2 containing a precursor in a small proportion. The applications of this novel technique to generate protonated polycyclic aromatic hydrocarbons (H(+)PAH), protonated polycyclic nitrogen heterocycles (H(+)PANH), and their neutral counterparts, which are important in the identification of interstellar unidentified infrared emission bands, demonstrate its superiority over other methods. The clean production with little fragmentation, ease of distinction between protonated and neutral species, narrow lines and reliable relative infrared intensities of the lines, and broad coverage of the spectral range associated with this method enable us to assign the isomers unambiguously. The application of this method to the protonation of small molecules is more complicated partly because of the feasible fragmentation and reactions, and partly because of the possible proton sharing between the species of interest and H-2, but, with isotopic experiments and secondary photolysis, definitive assignments are practicable. Furthermore, the true relative infrared intensities are critical to a comparison of experimental results with data from theoretical calculations. The spectra of a proton-shared species in solid p-H-2 might provide insight into a search for spectra of proton-bound species in interstellar media. Investigations of hydrogenated species involving the photolysis of Cl-2 or precursors of OH complement those using electron bombardment and provide an improved ratio of signal to noise. With careful grouping of observed lines after secondary photolysis and a comparison with theoretical predictions, various isomers of these species have been determined. This photolytic technique has been applied in an investigation of hydrogenated PAH and PANH, and the hydrogenation reactions of small molecules, which are important in interstellar ice and the evolution of life. The electronic transitions of molecules in solid p-H-2 have been little investigated. The matrix shift of the origins of transitions and the spectral width seem to be much smaller than those of noble- gas matrices; these features might facilitate a direct comparison of matrix spectra with diffuse interstellar bands, but further data are required to assess this possibility. The advantages and disadvantages of applying these techniques of p-H-2 matrix isolation to astrochemical research and their future perspectives are discussed.
Decentralized Synchronization of Uncertain Nonlinear Systems With a Reputation Algorithm
(2018-03-01)
Klotz, Justin R.
;
Parikh, Anup
;
Cheng, Teng-Hu
;
Dixon, Warren E.
;
機械工程學系
Decentralized synchronization approaches generally assume that communication of state information between neigh-boring agents is completely accurate. However, just one agent's communication of inaccurate information to neighbors can significantly degrade the performance of the entire network. To help abate this problem, we present a decentralized controller for a leader-follower framework which uses local information to vet neighbors and change consensus weights accordingly. Because updates of the consensus weights produce a switched system, switching control theory techniques are used to develop a dwell-time that must elapse between agents' successive weight updates. A Lyapunov-based stability analysis is presented which develops sufficient conditions for approximate convergence of follower agents' states to a leader agent's time-varying state. Simulation results are provided to demonstrate the performance of the developed techniques.
An Efficient Tree Search Algorithm for the Free Distance of Variable-Length Error-Correcting Codes
(2018-03-01)
Huang, Chun
;
Wu, Ting-Yi
;
Chen, Po-Ning
;
Alajaji, Fady
;
Han, Yunghsiang S.
We propose an efficient tree search algorithm for determining the free distance of variable-length error-correcting codes (VLECs). A main idea behind the algorithm is to structure all pairs of code word-concatenated sequences as a tree, in which we seek the pair of sequences that determine the free distance. In order to speed up the algorithm, we establish constraints that do not compromise optimality in determining the free distance. Experimental results on VLECs algorithmically constructed for the English alphabet show that our algorithm requires a considerably smaller number of bitwise distance computations and covers a much smaller number of tree nodes than Dijkstra's algorithm operating over the pairwise distance graph while being a hundred times faster in terms of execution time.
PotteryGo: A Virtual Pottery Making Training System
(2018-03-01)
Chiang, Pei-Ying
;
Chang, Han-Yu
;
Chang, Yung-Ju
;
資訊工程學系
;
Department of Computer Science
【活動花絮】來成為熱情洋溢的資訊人
(國立交通大學資訊學院, 2018-03)
系學會會長郭毓梁
交大資工系學會是由資工系大二及大三的同學們組成，做為學生自治團體及全體同學與系上的聯繫溝通管道，致力於促進師長與同學間的溝通交流，也期望能透過舉辦別出心裁且有趣的活動，凝聚系上的感情與同學間的情誼。
Efficient and Adaptive Error Recovery in a Micro-Electrode-Dot-Array Digital Microfluidic Biochip
(2018-03-01)
Li, Zipeng
;
Lai, Kelvin Yi-Tse
;
McCrone, John
;
Yu, Po-Hsien
;
Chakrabarty, Krishnendu
A digital microfluidic biochip (DMFB) is an attractive technology platform for automating laboratory procedures in biochemistry. In recent years, DMFBs based on a microelectrode- dot-array (MEDA) architecture have been proposed. MEDA biochips can provide advantages of better capability of droplet manipulation and real-time sensing ability. However, errors are likely to occur due to defects, chip degradation, and the lack of precision inherent in biochemical experiments. Therefore, an efficient error-recovery strategy is essential to ensure the correctness of assays executed on MEDA biochips. By exploiting MEDA-specific advances in droplet sensing, we present a novel error-recovery technique to dynamically reconfigure the biochip using real-time data provided by on-chip sensors. Local recovery strategies based on probabilistic-timed-automata are presented for various types of errors. An online synthesis technique and a control flow are also proposed to connect local-recovery procedures with global error recovery for the complete bioassay. Moreover, an integer linear programming-based method is also proposed to select the optimal local-recovery time for each operation. Laboratory experiments using a fabricated MEDA chip are used to characterize the outcomes of key droplet operations. The PRISM model checker and three benchmarks are used for an extensive set of simulations. Our results highlight the effectiveness of the proposed error-recovery strategy.
Electrical coupling between ventricular myocytes and myofibroblasts in the infarcted mouse heart
(2018-03-01)
Rubart, Michael
;
Tao, Wen
;
Lu, Xiao-Long
;
Conway, Simon J.
;
Reuter, Sean P.
Recent studies have demonstrated electrotonic coupling between scar tissue and the surrounding myocardium in cryoinjured hearts. However, the electrical dynamics occurring at the myocyte-nonmyocyte interface in the fibrotic heart remain undefined. Here, we sought to develop an assay to interrogate the nonmyocyte cell type contributing to heterocellular coupling and to characterize, on a cellular scale, its voltage response in the infarct border zone of living hearts. We used two-photon laser scanning microscopy in conjunction with a voltage-sensitive dye to record transmembrane voltage changes simultaneously from cardiomyocytes and adjoined nonmyocytes in Langendorff-perfused mouse hearts with healing myocardial infarction. Transgenic mice with cardiomyocyte-restricted expression of a green fluorescent reporter protein underwent permanent coronary artery ligation and their hearts were subjected to voltage imaging 7-10 days later. Reporter-negative cells, i.e. nonmyocytes, in the infarct border zone exhibited depolarizing transients at a 1:1 coupling ratio with action potentials recorded simultaneously from adjacent, reporter-positive ventricular myocytes. The electrotonic responses in the nonmyocytes exhibited slower rates of de- and repolarization compared to the action potential waveform of juxtaposed myocytes. Voltage imaging in infarcted hearts expressing a fluorescent reporter specifically in myofibroblasts revealed that the latter were electrically coupled to border zone myocytes. Their voltage transient properties were indistinguishable from those of nonmyocytes in hearts with cardiomyocyte-restricted reporter expression. The density of connexin43 expression at myofibroblast-cardiomyocyte junctions was similar to 5% of that in the intercalated disc regions of paired ventricular myocytes in the remote, uninjured myocardium, whereas the ratio of connexin45 to connexin43 expression levels at heterocellular contacts was similar to 1%. Myofibroblasts contribute to the population of electrically coupled nonmyocytes in the infarct border zone. The slower kinetics of myofibroblast voltage responses may reflect low electrical conductivity across heterocellular junctions, in accordance with the paucity of connexin expression at myofibroblast-cardiomyocyte contacts.
Simulating painted appearance of BTF materials
(2018-03-01)
Chen, Ta-En
;
Huang, Tsung-Shian
;
Lin, Wen-Chieh
;
Chuang, Jung-Hong
;
資訊工程學系
Bidirectional texture function (BTF) provides a realistic representation for real-world materials; however, the representation does not allow users to append additional rendering effects to the BTF data in a simple and effective way. In this paper, we present an approach for generating the painted appearance on a material that is represented by a BTF. We first convert a BTF into a physical representation composed of a height-field and a spatially varying bidirectional reflectance distribution function (SVBRDF). This representation allows us to simulate the appearance of paints on the surface of a material by computing the reflectance and the height field of the painted material. During rendering, we combine an existing BTF rendering approach and the modified material data to simulate the painted appearance of the BTF materials. Our experiments show that the proposed approach can simulate the painted appearance of a material while preserving the characteristic appearance of the original BTF. Our approach is mainly suitable for less specular BTF materials that allow depth reconstruction.
The effects of dynamic industrial transition on sustainable development
(2018-03-01)
Tsai, Diana H. A.
;
經營管理研究所
;
Institute of Business and Management
This paper analyzes the dynamics of industrial development and structural changes, and its impacts on sustainable development. The weak form of the Porter Hypothesis is examined under industrial transition and structural changes. Explicitly incorporating the Markov regime switching mechanism in a productivity framework, we measure how sustainable development is affected and how firms are adjusted when facing industrial transition and structural changes. Applied to Taiwan manufacturing industries, the model is implemented to identify structural changes and to evaluate the viability of sustainable development under new constraints. This study suggests industries to adopt more sustainable practices which can promote and even improve industrial competitiveness. Such practices would empower economies to assess current structural changes and, based on the environmental implications, recommend future economic policy for sustainable development. (C) 2017 Elsevier B.V. All rights reserved.
The Application of Quality Function Deployment to Smart Watches - The House of Quality for Improved Product Design
(2018-03-01)
Ho, W. C.
;
Lee, A. W.
;
Lee, S. J.
;
Lin, Grace T. R.
;
科技管理研究所
Consumers are confused about the benefits of using smart watches due to the various insignificant functionalities included and the unclear value propositions promoted by smart watch manufacturers. In order to improve the quality of products and to provide features that can satisfy consumers' needs, technicians must understand the expectations of consumers. This paper recommends the use of quality function deployment (QFD), a method that integrates consumers' dynamic requirements with product designs, which can be represented using a House of Quality (HoQ) matrix diagram. Here, the unmet and uncovered needs of consumers relating to the smart watch are analyzed extensively using a questionnaire-based survey. The opinions of consumers and technicians are then examined, and a HoQ diagram is constructed to evaluate the relationship between the design attributes of smart watch and its technical attributes. Finally, the attributes that need to be improved are listed in a suggested order of priority.
渦輪編碼的數位資料之去穿刺方法與裝置及渦輪解碼器系統
(2018-03-01)
張錫嘉
;
林振揚
一種渦輪解碼器系統解碼由一系統碼及第一與第二同位檢查碼構成的L-長度數位資料，並包含一柵欄控制器，其獲得該第一/二同位檢查碼與原始系統碼的位元數比Ep/D並根據該數位資料的碼率產生一指示出一目標解碼柵欄的柵欄控制輸出，該目標解碼柵欄是由一渦輪解碼器選擇來執行解碼操作。一補零模組將零補進該系統碼，並根據Ep/D之值將一或多個零補進該第一/二同位檢查碼以使得該第一/二同位檢查碼的同位檢查位元與該(等)零-位元在位元位置上係呈週期性地排列。