(Student)溫柏叡 : Note that I did not find the time on the subtitles, the teacher cast 2.3 to write, and then speak 3.3 start, so pay attention to start from 2.3.
Update Date: 2017-04-17
Category: 2017 Operating System
[2017-04-18]Operating System
Update Date: 2017-04-25
Category: 2017 Operating System
[2017-04-25]Operating System
Update Date: 2017-05-02
Category: 2017 Operating System
[2017-05-02]Operating System
Update Date: 2017-05-10
Category: 2017 Operating System
[2017-05-09]Operating System
Update Date: 2017-05-14
Category: 2017 Operating System
[2017-05-16]Operating System
Update Date: 2017-05-23
Category: 2017 Operating System
[2017-05-23]Operating System
Update Date: 2017-06-04
Category: 2017 Operating System
[2017-06-06]Operating System
Update Date: 2017-06-14
Category: 2017 Operating System
[2017-06-13]Operating System
Update Date: 2017-06-19
Category: 2017 Operating System
[2018-03-02]Operating System
Update Date: 2018-03-07
Category: 2018 Operating System
[2018-03-09] Operating System
Update Date: 2018-03-13
Category: 2018 Operating System
[2018-03-15] Operating System
Update Date: 2018-03-18
Category: 2018 Operating System
[2018-03-22] Operating System
Update Date: 2018-03-27
Category: 2018 Operating System
[2018-03-29] Operating System
Update Date: 2018-04-11
Category: 2018 Operating System
[2018-04-12] Operating System
Update Date: 2018-04-18
Category: 2018 Operating System
[2018-04-19] Operating System
Update Date: 2018-04-25
Category: 2018 Operating System
[2018-04-26] Operating System
Update Date: 2018-05-01
Category: 2018 Operating System
[2018-05-03] Operating System
Update Date: 2018-05-01
Category: 2018 Operating System
[2018-05-10] Operating System
Update Date: 2018-05-15
Category: 2018 Operating System
[2018-05-17] Operating System
Update Date: 2018-05-22
Category: 2018 Operating System
[2018-05-24] Operating System
Update Date: 2018-06-07
Category: 2018 Operating System
[2018-05-31] Operating System
Update Date: 2018-06-07
Category: 2018 Operating System
[2018-06-07] Operating System
Update Date: 2018-06-07
Category: 2018 Operating System
[2018-06-14] Operating System
Update Date: 2018-06-22
Category: 2018 Operating System
[2018-06-21] Operating System
Update Date: 2018-06-22
Category: 2018 Operating System
[2019-02-21] Operating System
Update Date: 2019-02-21
Category: 2019 Operating System
[2019-03-07] Operating System
Update Date: 2019-03-07
Category: 2019 Operating System
[2019-03-14] Operating System
Update Date: 2019-03-14
Category: 2019 Operating System
[2019-03-21] Operating System
Update Date: 2019-03-21
Category: 2019 Operating System
[2019-03-28] Operating System
Update Date: 2019-03-28
Category: 2019 Operating System
[2019-04-11] Operating System
Update Date: 2019-04-11
Category: 2019 Operating System
[2019-04-18] Operating System
Update Date: 2019-04-18
Category: 2019 Operating System
[2019-04-25] Operating System
Update Date: 2019-04-25
Category: 2019 Operating System
[2019-05-02] Operating System
Update Date: 2019-05-02
Category: 2019 Operating System
[2019-05-09] Operating System
Update Date: 2019-05-09
Category: 2019 Operating System
[2019-05-16] Operating System
Update Date: 2019-05-16
Category: 2019 Operating System
[2019-05-23] Operating System
Update Date: 2019-05-23
Category: 2019 Operating System
[2019-05-30] Operating System
Update Date: 2019-05-30
Category: 2019 Operating System
[2019-06-4] Operating System(當天錄影機故障請看其他學期影片)
Update Date: 2019-06-04
Category: 2019 Operating System
[2019-06-13] Operating System
Update Date: 2019-06-13
Category: 2019 Operating System
Multimedia Database Introduction Film
Update Date: 2017-01-23
Category: About us
2017 New Student Orientation - Introduction to MDB Lab
Update Date: 2017-04-15
Category: About us
Project Name
Project Name
Project Name
Project Name
Project Name
Parallelization of a Self-adaptive Harmony Search Algorithm on Graphics Processing Units
Abstract
In recent years, in order to reduce the execution time, some evolutionary algorithms that run on GPUs using Compute Unified Device Architecture (i.e., CUDA) have been proposed. In these evolutionary algorithms, they compared the execution time and precision of GPU versions with those of CPU versions. In this study, we parallelize a self-adaptive harmony search algorithm and compare with the existing evolutionary algorithms on the same GPU platform. The proposed algorithm is divided into four steps: initialization, improvising, sorting, and updating. In the experiments, we use eight well-known optimization problems to evaluate the proposed algorithm and the other existing algorithms. As a result, our algorithm achieves the best performances among all the algorithms on the single-objective optimization problems with more dimensions or populations.
Keywords: evolutionary algorithms, CUDA, self-adaptive harmony search algorithms, parallel algorithms, optimization problems
Update Date: 2019-01-14
Author: Sun-Ho, Chou
Category: Algorithm
Optimizing Convolutional Neural Network Architecture Using a Self-adaptive Harmony Search Algorithm
Abstract In recent years, the advance of GPUs led to the development in neural networks and deep learning. However, it is difficult to find a good CNN architecture people desire. In the past, people had to manually find the CNN architecture, and this is quite time-consuming and labor-intensive. In this paper, we use a self-adaptive harmony search algorithm to find the optimized convolutional neural network architecture for image recognition. The system architecture is divided into two phases. In the first phase, we search for the most suitable layer length of a CNN. In the second phase, we fine-tune the architecture found in the first phase or a pre-trained architecture. In the experiments, three popular and well-known datasets are used to evaluate the proposed methods and the state-of-the-art CNN search methods. The experimental results show that our methods achieve competitive performances compared with the other methods on MNIST and CIFAR-10, and have the best performance among all the methods on Caltech-101. Keywords: convolutional neural networks, deep learning, hyper-parameter optimization, self-adaptive harmony search algorithms, transfer learning
Update Date: 2019-01-14
Author: Jung-Sheng, Liu
Category: Algorithm
Text Analysis and Detection on Fake News
Abstract
In general, the features of fake news are almost the same as those of real news, so it
is not easy to identify them. In this paper, we propose a fake news detection system using
a deep learning model. First, news articles are preprocessed and analyzed based on
different training models. Then, an ensemble learning model combining four different
models called embedding LSTM, depth LSTM, LIWC CNN, and N-gram CNN is
proposed for fake news detection. Besides, to achieve high accuracy of detecting fake
news, the optimized weights of the ensemble learning model are determined using the
Self-Adaptive Harmony Search (SAHS) algorithm. In the experiments, we verify that the
proposed model is superior to the state-of-the-art methods, with the highest accuracy
99.4%. Furthermore, we also investigate the cross-domain intangibility issue and achieve
the highest accuracy 72.3%. Finally, we believe there is still room for improving the
ensemble learning model in addressing the cross-domain intangibility issue.
Keywords: deep learning, fake news, natural language processing, text mining
Update Date: 2018-07-26
Author: Bo-Hong, Chen
Category: Deep learning
Image Retrieval Based on AND/OR-construction Models
Abstract
With the rapid development of the Internet, the number of images on the Internet
is increasing by millions. Finding desired images from numerous images has become
an important research topic. In this paper, we propose an image retrieval system
facilitating retrieval time and accuracy. Since the performance of image retrieval is
deeply influenced by image features and retrieval methods. Five different types of
features and five different methods are used to find the best combination for an image
retrieval system. First, we segment out the main object in an image and then extract its
features. Next, relevant features are selected from the original feature set for facilitating
image retrieval, using the SAHS algorithm. Then, five methods based on AND/ORconstruction
are proposed to build the image retrieval model, using the relevant features.
Finally, the experimental results not only show that our methods are more effective than
the other state-of-the-art methods, but also present some observations never explored
by the previous research.
Keywords: data mining, image retrieval, locality sensitive hashing, Kmeans, deep
learning
Update Date: 2018-07-26
Author: Yun-Xin, Xie
Category: Data mining
Solving Multi-Objective Optimization Problem Using a Self-Adaptive Harmony Search Algorithm
Abstract
In recent years, there have been many Multi-Objective Evolutionary Algorithms
(MOEAs) proposed to solve multi-objective optimization problems. These evolutionary
algorithms generate many solutions for iterations and move to the true Pareto optimal
region gradually. As expected, since the Harmony Search (HS) algorithm can also iterate
over a large number of solutions (in HM memory) and moves to the true Pareto optimal
region, we use it to solve multi-objective optimization problems. In this paper, the
proposed system architecture can be divided into two phases. In the first phase, we aim
to search feasible solution regions as widely as possible in the entire process. In the
second phase, we focus on searching optimized solutions stepwise in the feasible solution
regions. Since the proposed algorithm uses many parameters, we adjust some of them in
a self-adaptive way and call the algorithm self-adaptive. In the experiments, we use the
eleven well-known multi-objective problems to examine the proposed algorithm and
other existing algorithms, based on five performance indicators. As a result, our algorithm
achieves better performances than the others in IGD, HV, and Spread indicators.
Recently in exploiting green energy, solar power generation is a must-be trend and approach, especially for the countries with nature resource shortage. However, how to build solar power plants with the best power generation efficiency in limited spaces is always a crucial issue. In this paper, the approach of finding the optimum models of generating solar power is proposed to build solar power plants for different environments in Taiwan. First, we collect all the data from existing solar power farms, including 1) design methods of power generation, 2) actual power generation, and 3) surrounding environments. Then, after a series of preprocessing steps and system analysis on them, the optimal models of generating solar power could be mined out. Finally, in the experiments, we evaluate the system from five aspects regarding to input and output parameters. As a result, we observe that using the majority voting strategy improves the system accuracy and helps engineers build solar power plants with the maximum power generation.
Update Date: 2017-01-19
Author: Shun-Hao, Chang
Category: Data Mining
Implementation of a Medical-care Management System