Browsing by Author "Lee, Chia-Ying"

Now showing 1 - 8 of 8

Early MEG markers for reading Chinese phonograms: Evidence from radical combinability and consistency effects
(2014-12-01)
Hsu, Chun-Hsien
;
Lee, Chia-Ying
;
Tzeng, Ovid J. -L.
;
生物科技學系
;
Department of Biological Science and Technology
Studies using functional magnetic resonance imaging have indicated that activities in the left inferior frontal cortex and left temporoparietal regions are associated with orthographic neighborhood size. To elucidate the temporal dynamics of reading-related cortical activities, we manipulated two types of neighborhood properties for Chinese phonograms, phonetic combinability and consistency. By using source analysis techniques in combination with magnetoencephalography, the results demonstrated a combinability effect in the right fusiform gyrus at similar to 170 ms, which may reflect perceptual expertise in processing Chinese orthography. During 200 ms to 250 ms, the left anterior insula showed larger activity in reading small combinability characters than in reading large combinability characters, and the left inferior parietal cortex showed greater activity in reading low consistency characters than in reading high consistency characters. These results indicate that the left anterior insula cortex and left inferior parietal cortex may play important roles in the early stages of reading Chinese phonograms. (C) 2014 Elsevier Inc. All rights reserved.
Gate controlled field emission triode and process for fabricating the same
(2009-08-13)
Tseng, Tseung-Yuan
;
Lee, Chia-Ying
;
Li, Seu-Yi
;
Lin, Pang
;
Tseng, Tseung-Yuen
This invention relates to a process for fabricating ZnO nanowires with high aspect ratio at low temperature, which is associated with semiconductor manufacturing process and a gate controlled field emission triode is obtained. The process comprises providing a semiconductor substrate, depositing a dielectric layer and a conducting layer, respectively, on the semiconductor substrate, defining the positions of emitter arrays on the dielectric layer and conducting layer, depositing an ultra thin ZnO film as a seeding layer on the substrate, growing the ZnO nanowires as the emitter arrays by using hydrothermal process, and etching the areas excluding the emitter arrays, then obtaining the gate controlled field emission triode.
Gate controlled field emission triode and process for fabricating the same
(2007-12-13)
Tseng, Tseung-Yuen
;
Lee, Chia-Ying
;
Li, Seu-Yi
;
Lin, Pang
This invention relates to a process for fabricating ZnO nanowires with high aspect ratio at low temperature, which is associated with semiconductor manufacturing process and a gate controlled field emission triode is obtained. The process comprises providing a semiconductor substrate, depositing a dielectric layer and a conducting layer, respectively, on the semiconductor substrate, defining the positions of emitter arrays on the dielectric layer and conducting layer, depositing an ultra thin ZnO film as a seeding layer on the substrate, growing the ZnO nanowires as the emitter arrays by using hydrothermal process, and etching the areas excluding the emitter arrays, then obtaining the gate controlled field emission triode.
Research on Steady-State Visual Evoked Potentials in 3D Displays
(2015-01-01)
Chien, Yu-Yi
;
Lee, Chia-Ying
;
Lin, Fang-Cheng
;
Huang, Yi-Pai
;
Ko, Li-Wei
Brain-computer interfaces (BCIs) are intuitive systems for users to communicate with outer electronic devices. Steady-state visual evoked potential (SSVEP) is one of the common inputs for BCI systems due to its easy detection and high information transfer rates. An advanced interactive platform integrated with liquid crystal displays is leading a trend to provide an alternative option not only for the handicapped but also for the public to make our lives more convenient. Many SSVEP-based BCI systems have been studied in a 2D environment; however there is only little literature about SSVEP-based BCI systems using 3D stimuli. 3D displays have potentials in SSVEP-based BCI systems because they can offer vivid images, good quality in presentation, various stimuli and more entertainment. The purpose of this study was to investigate the effect of two important 3D factors (disparity and crosstalk) on SSVEPs. Twelve participants participated in the experiment with a patterned retarder 3D display. The results show that there is a significant difference (p-value<0.05) between large and small disparity angle, and the signal-to-noise ratios (SNRs) of small disparity angles is higher than those of large disparity angles. The 3D stimuli with smaller disparity and lower crosstalk are more suitable for applications based on the results of 3D perception and SSVEP responses (SNR). Furthermore, we can infer the 3D perception of users by SSVEP responses, and modify the proper disparity of 3D images automatically in the future.
The Explicit and Implicit Phonological Processing of Chinese Characters and Words in Taiwanese Deaf Signers
(2016-01)
Chiu, Yi-Shiuan
;
Kuo, Wen-Jui
;
Lee, Chia-Ying
;
Tzeng, Ovid J. L.
;
交大名義發表
This study investigates the phonological processing that Taiwanese deaf signers engage in when recognizing Chinese characters and words. The deaf participants\' orthography-phonology transformation (OPT) abilities were tested using an explicit Chinese homophone judgment task, and their implicit phonological activations were tested using lexical decision tasks. Chinese characters, whose phonetic radicals are not always reliable guides to pronunciation, are a useful tool for dissociating the influence of phonology and orthography. Experiment 1 manipulated sound-based phonological similarity (similar, dissimilar) and orthographical similarity (similar, dissimilar). Accuracy, sensitivity (d\'), and reaction times (RTs) were recorded for hearing participants, but only accuracy and sensitivity (d\') recorded for deaf participants, who are fluent signers of Taiwanese Sign Language (TSL). Additionally, the predictive abilities of log word frequency and the consistency values for homophone judgment performance were analyzed. Experiment 2-1 was designed to compare the effects of three primes (semantically related, sound-based phonologically related, and unrelated primes) on the performance of deaf and hearing participants on a lexical decision task. In Experiment 2-2, TSL phonologically related primes were compared with unrelated primes for deaf and hearing subjects. The results of Experiment 1 show that the accuracy of deaf subjects was poor (. 66) and was lower than that of hearing subjects (. 87). Particularly, the deaf subjects\' accuracy and sensitivity in the orthographically dissimilar condition were significantly lower than in the orthographically similar condition. For hearing subjects, log word frequency significantly predicted their accuracy and RTs, whereas the consistency values predicted only the RTs. For the deaf participants, the accuracy could be efficiently predicted by both log word frequency and consistency values, which reflect knowledge of OPT rules. We suggested that although deaf individuals had acquired knowledge of OPT rules, this knowledge was neither complete nor sufficiently robust to make homophone judgments. In Experiment 2-1, the results show that there was a semantic priming effect but no sound-based phonological priming effect for the deaf participants. The results further reveal that deaf people with a limited hearing ability did not automatically process the sound-based phonological representations under time-constrained conditions. In Experiment 2-2, there was an action-based phonological priming effect for deaf signers but not for hearing subjects, indicating that deaf signers automatically activated related action-based phonology to access the semantic meaning of words when reading Chinese. This study finds that deaf signers acquire OPT rules but that their OPT rules are not sufficiently robust or complete to allow them to make explicit phonological judgments and homophone judgments. Deaf signers automatically activated action-based representations rather than sound-based phonological representations when reading Chinese characters.
The Explicit and Implicit Phonological Processing of Chinese Characters and Words in Taiwanese Deaf Signers
(2016-01-01)
Chiu, Yi-Shiuan
;
Kuo, Wen-Jui
;
Lee, Chia-Ying
;
Tzeng, Ovid J. L.
;
交大名義發表
This study investigates the phonological processing that Taiwanese deaf signers engage in when recognizing Chinese characters and words. The deaf participants\' orthography-phonology transformation (OPT) abilities were tested using an explicit Chinese homophone judgment task, and their implicit phonological activations were tested using lexical decision tasks. Chinese characters, whose phonetic radicals are not always reliable guides to pronunciation, are a useful tool for dissociating the influence of phonology and orthography. Experiment 1 manipulated sound-based phonological similarity (similar, dissimilar) and orthographical similarity (similar, dissimilar). Accuracy, sensitivity (d\'), and reaction times (RTs) were recorded for hearing participants, but only accuracy and sensitivity (d\') recorded for deaf participants, who are fluent signers of Taiwanese Sign Language (TSL). Additionally, the predictive abilities of log word frequency and the consistency values for homophone judgment performance were analyzed. Experiment 2-1 was designed to compare the effects of three primes (semantically related, sound-based phonologically related, and unrelated primes) on the performance of deaf and hearing participants on a lexical decision task. In Experiment 2-2, TSL phonologically related primes were compared with unrelated primes for deaf and hearing subjects. The results of Experiment 1 show that the accuracy of deaf subjects was poor (. 66) and was lower than that of hearing subjects (. 87). Particularly, the deaf subjects\' accuracy and sensitivity in the orthographically dissimilar condition were significantly lower than in the orthographically similar condition. For hearing subjects, log word frequency significantly predicted their accuracy and RTs, whereas the consistency values predicted only the RTs. For the deaf participants, the accuracy could be efficiently predicted by both log word frequency and consistency values, which reflect knowledge of OPT rules. We suggested that although deaf individuals had acquired knowledge of OPT rules, this knowledge was neither complete nor sufficiently robust to make homophone judgments. In Experiment 2-1, the results show that there was a semantic priming effect but no sound-based phonological priming effect for the deaf participants. The results further reveal that deaf people with a limited hearing ability did not automatically process the sound-based phonological representations under time-constrained conditions. In Experiment 2-2, there was an action-based phonological priming effect for deaf signers but not for hearing subjects, indicating that deaf signers automatically activated related action-based phonology to access the semantic meaning of words when reading Chinese. This study finds that deaf signers acquire OPT rules but that their OPT rules are not sufficiently robust or complete to allow them to make explicit phonological judgments and homophone judgments. Deaf signers automatically activated action-based representations rather than sound-based phonological representations when reading Chinese characters.
Towards high performance data analytic on heterogeneous many-core systems: A study on Bayesian Sequential Partitioning
(2018-12-01)
Lai, Bo-Cheng
;
Wu, Tung-Yu
;
Chiu, Tsou-Han
;
Li, Kun-Chun
;
Lee, Chia-Ying
Bayesian Sequential Partitioning (BSP) is a statistically effective density estimation method to comprehend the characteristics of a high dimensional data space. The intensive computation of the statistical model and the counting of enormous data have caused serious design challenges for BSP to handle the growing volume of the data. This paper proposes a high performance design of BSP by leveraging a heterogeneous CPU/GPGPU system that consists of a host CPU and a K80 GPGPU. A series of techniques, on both data structures and execution management policies, is implemented to extensively exploit the computation capability of the heterogeneous many-core system and alleviate system bottlenecks. When compared with a parallel design on a high-end CPU, the proposed techniques achieve 48x average runtime enhancement while the maximum speedup can reach 78.76x. (C) 2018 Elsevier Inc. All rights reserved.
Unified Designs for High Performance LDPC Decoding on GPGPU
(2016-12-01)
Lai, Bo-Cheng Charles
;
Lee, Chia-Ying
;
Chiu, Tsou-Han
;
Kuo, Hsien-Kai
;
Chang, Chun-Kai
Modern GPGPU\'s have enabled massively parallel computing with programmability that can exploit the highly parallel nature of LDPC decoding. Previous works customized the design on a GPGPU towards specific execution attributes of a particular LDPC decoding matrix. Supporting different LDPC decoding matrices requires either substantial rework on the current program, or a brand new parallel design. This paper proposes two unified designs that can achieve high performance for both regular and irregular LDPC decoding on a GPGPU. The first design introduces a node-based scheme with a versatile translation array mechanism that can efficiently handle the complex data access patterns of different LDPC decoding matrices. The second design proposes an edge-based parallel paradigm that uses more intuitive data layout. More edges than nodes in a Tanner graph also give the edge-based design higher computation parallelism when there are limited concurrent codewords. With the proposed unified designs, designers can be ignorant of the types of LDPC matrices and achieve high performance LDPC decoding. The experiments on a GTX 470 GPGPU have demonstrated up to 134.56x runtime improvement, when compared with designs on a high-end CPU. The maximum throughput can reach 80.25 Mbps. When compared with the previous customized designs, the proposed systematic designs can reach better performance while relieving the effort of customization.