學術出版;;Publications

Permanent URI for this communityhdl:11536/1

收集所有交通大學研究人員與學生個人或團體所發表之各項學術著作，涵蓋期刊論文、教師專書、學位論文等項。;;Collections of Publications of NCTU Researchers and Students Like Journal Papers and Thesis.

Browse

🏛️國立陽明交通大學機構典藏

Search Results

Now showing 1 - 10 of 65

A Robust Algorithm for Predicting Attacks Using Collaborative Security Logs
(2020-05-01)
Rezapour, Amir
;
Tzeng, Wen-Guey
;
資訊工程學系
;
Department of Computer Science
As networks become ubiquitous in our daily lives, users rely more on networks for exchanging data and communication. However, numerous new and sophisticated attacks that endanger security of users have been reported. In practice, blacklisting illicit sources has been a fundamental defense strategy in recent years. In this paper, we propose a predictor that is based on the observations from a centralized log-sharing infrastructure. Our observations include the direct relation between attackers and victims, victim similarities, and attacker correlations. We compile a customized blacklist for each Dshield.org contributor using a weighted function of direct and indirect relations between victims and attackers. This list not only offers a significantly higher prediction ratio, but also includes source addresses with potentially higher threats. We evaluate our predictor using two months of malicious activities acquired from Dshield.org. The experimental results demonstrate a significant improvement over previous algorithms.
Deep Bayesian Data Mining
(2020-01-01)
Chien, Jen-Tzung
;
交大名義發表
;
National Chiao Tung University
This tutorial addresses the fundamentals and advances in deep Bayesian mining and learning for natural language with ubiquitous applications ranging from speech recognition [7, 55] to document summarization [8], text classification [5, 75], text segmentation [18], information extraction [50], image caption generation [69, 72], sentence generation [25, 46], dialogue control [22, 76], sentiment classification, recommendation system, question answering [58] and machine translation [2], to name a few. Traditionally, "deep learning" is taken to be a learning process where the inference or optimization is based on the real-valued deterministic model. The "semantic structure" in words, sentences, entities, actions and documents drawn from a large vocabulary may not be well expressed or correctly optimized in mathematical logic or computer programs. The "distribution function" in discrete or continuous latent variable model for natural language may not be properly decomposed or estimated. This tutorial addresses the fundamentals of statistical models and neural networks, and focus on a series of advanced Bayesian models and deep models including hierarchical Dirichlet process [61], Chinese restaurant process [4], hierarchical Pitman-Yor process [60], Indian buffet process [35], recurrent neural network (RNN) [26, 41, 48, 65], long short-term memory, sequence-to-sequence model [59], variational auto-encoder (VAE) [44], generative adversarial network (GAN) [36], attention mechanism [27, 56], memory-augmented neural network [39, 58], skip neural network [6], temporal difference VAE [40], stochastic neural network [3, 47], stochastic temporal convolutional network [1], predictive state neural network [31], and policy neural network [49, 74]. Enhancing the prior/posterior representation is addressed [53, 62]. We present how these models are connected and why they work for a variety of applications on symbolic and complex patterns in natural language. The variational inference and sampling method are formulated to tackle the optimization for complicated models [54]. The word and sentence embeddings, clustering and co-clustering are merged with linguistic and semantic constraints. A series of case studies, tasks and applications are presented to tackle different issues in deep Bayesian mining, searching, learning and understanding. At last, we will point out a number of directions and outlooks for future studies. This tutorial serves the objectives to introduce novices to major topics within deep Bayesian learning, motivate and explain a topic of emerging importance for data mining and natural language understanding, and present a novel synthesis combining distinct lines of machine learning work.
Using Cache Optimization Method to Reduce Network Traffic in Communication Systems Based on Cloud Computing
(2019-01-01)
Kang, Lanlan
;
Chen, Ruey-Shun
;
Chen, Yeh-Cheng
;
Wang, Chung-Chei
;
Li, Xingguan
The overwhelming increase of population will lead to an increase network traffic usage. In most products, local cache mechanism is implemented for the purpose of reducing the network traffics. However, Due to storage space limitations, lot of times, cache will get purged when there are lots of queries sent and replaced by other newest queries. It is not an efficient by using traditional methodology to build or forming cache. In order to reduce the network traffic usage, the paper propose a solution, which utilizes data mining technique with clustering concept, by gathering the current feedback data that have from our SPN (Smart Protection Network). we are able to form these data in groups with similarity, and by deploying these data to client side, to achieve the reduction of traffic usage. In prototype, this design can really reduce network traffic more than 8%.
A Distributed Algorithm for Fast Mining Frequent Patterns in Limited and Varying Network Bandwidth Environments
(2019-05-01)
Lin, Chun-Cheng
;
Li, Wei-Ching
;
Chen, Ju-Chin
;
Chung, Wen-Yu
;
Chung, Sheng-Hao
Data mining is a set of methods used to mine hidden information from data. It mainly includes frequent pattern mining, sequential pattern mining, classification, and clustering. Frequent pattern mining is used to discover the correlation among various sets of items within large databases. The rapid upward trend in data size slows the mining of frequent patterns. Numerous studies have attempted to develop algorithms that operate in distributed computing environments to accelerate the mining process. FLR-mining (Fast, Load balancing and Resource efficient mining algorithm) is one of the fastest methods of mining with efficient consideration of load balancing and resources. FLR-mining can automatically determine the appropriate number of computing nodes. However, FLR-mining and existing methods assume that the network bandwidth is constant. In practical distributed and many-task computing systems, this assumption fails because there are packet collisions caused by many mining tasks that run in a simultaneous manner. Therefore, a method that can consider the varying network bandwidth is necessary. In this study, we propose a method that can rapidly mine frequent patterns under the varying network bandwidth. The proposed method can also determine the appropriate number of computing nodes to efficiently utilize computing resources and achieve load balancing. Through empirical evaluation, the proposed method is shown to deliver excellent performance in terms of execution efficiency and load balancing.
Churn prediction and social neighbour influences for different types of user groups in virtual worlds
(2019-06-01)
Liu, Duen-Ren
;
Liao, Hsiu-Yu
;
Chen, Kuan-Yu
;
Chiu, Yi-Ling
;
資訊管理與財務金融系註：原資管所+財金所
In virtual worlds (VWs), users have more VW games alternatives, whereas VW companies consequently suffer from high customer turnover rate and low customer loyalty. Therefore, building a churn prediction model to facilitate subsequent churn management and customer retention is important. The churn behaviours and the impact of social neighbour influences to customer churn may be different for different types of users. Accordingly, we segment users into stable, unstable, and solitary user groups according to their social contact behaviours in VWs. Novel segmentation-based churn prediction approaches are proposed for churn prediction in VWs by building prediction models for each type of user groups and considering the effect of social neighbour influences for different user groups. The proposed approaches are evaluated by conducting experiments with a dataset collected from a VW platform. The experimental results show different churn prediction performances under different user groups. The segmentation-based churn prediction approaches perform better than do general approaches without considering user groups. Moreover, the results also reveal that social neighbour influences have a positive impact on stable and unstable users. The proposed work contributes to investigating the social neighbour influences on churn prediction for different types of user groups in VWs.
A Votable Concept Mapping Approach to Promoting Students' Attentional Behavior: Based on the Evidence of Mining Sequential Behavioral Patterns
(2017-01-01)
Sun, Jerry Chih-Yuan
;
Hwang, Gwo-Jen
;
Lin, Yu-Yan
;
Yu, Shih-Jou
;
Pan, Liu-Cheng
This study presented a votable concept mapping approach to promoting students' attentional behavior based on the evidence of mining sequential behavioral patterns. Twenty-nine students from an Educational Research Methodology course were recruited as participants. In the first week, a group-polling method was introduced in class; in the second week, participants were asked to draw concept maps using pen and paper (PnP concept mapping); and in the third week, the polling system and concept maps were integrated (votable concept mapping) and applied. The results showed that the votable concept mapping method was effective to stimulating students' attention during class. It was therefore suggested that teachers adopt methods integrating concept maps and polling tools to stimulate students' attention and thereby promote a positive cycle of attentional behavior in the classroom.
Using data mining to provide recommendation service
(2007-01-01)
Chen, Ruey-Shun
;
Tsai, Yung-Shun
;
Yeh, K. C.
;
Sun, Chia-Ming
;
Yu, D. H.
This research introduces personalized recommendation service into library services. Using the borrowing record of the library as basis, the association rules of data mining technique are used to look for book association by focusing on reader's borrowing mode, personal interest and trait in order to simplify the complexity of recommendation structure. The Bayesian network concept is used to build up a personalized book recommender system in order to generate different book recommendations, ranking from high to low, to help reader to locate book information most suitable to his requirement. Meanwhile we use user satisfaction questionnaire to understand the accuracy of recommended books and further to feedback information in order to help the post learning of Bayesian network parameter. This is for the perfection of the overall structure of recommender system so that readers could make use of the resource of the library more effectively and the value of the library system could be further improved.
Using Data Mining Technique to Improve Billing System Performance in Semiconductor Industry
(2018-01-01)
Chen, Yeh-Cheng
;
Chang, Yu-Teng
;
Kan, Y-S
;
Chen, R. S.
;
Wu, S. F.
The new billing approaches are manly to apply the integrated concept of data warehouse with relevant billing data; in addition, use the methods of mining association rule to sort out the Billing Quantities Pattern and then figure out the billing quantities. Moreover, employ the Decision Tree algorithm of data mining to find out the unit billing price. As a result, the new billing approach is made of the methods of data warehouse and date mining. This study is mainly focused on improving the operation of current billing system to establish the new functionality of the Billing quantities and Billing price. As for the benefit of these two new functions, it is not only able to lead into clients' billing systems, but it is also capable of upgrading the efficiency in rapid setup; especially for the enterprises that already possessed billing system internally but not yet implemented. In addition, it can also reduce the difference in revenue, shorten the process of issuing invoice, speed up the export operation, increase the export efficiency and provide the revenue data for integrating into the Executive Data System (EIS), Decision Support System (DSS) and Business Intelligent System (BIS) to allow enterprises making the right decisions promptly.
Data mining for enhanced driving effectiveness: an eco-driving behaviour analysis model for better driving decisions
(2017-01-01)
Hsu, Chia-Yu
;
Lim, Sirirat Sae
;
Yang, Chin-Sheng
;
科技管理研究所
;
Institute of Management of Technology
With the growing demand for energy efficient vehicles, automobile companies are constantly searching for better ways to study their customers' driving behaviour for effective new product design and development. One emerging driving behaviour among modern, eco-friendly drivers is the utilising of advanced vehicle technology for smarter, safer and more fuel-efficient driving. While many eco-driving studies focus on minimising fuel consumption, little attention is paid to how the behaviour of an individual driver and the type of vehicle used impact driving effectiveness. This study addresses this gap by proposing a novel overall drive effectiveness index that uses data mining for better driving decisions. Utilising data mining techniques, the index examines the impact of driving behaviour on driving effectiveness. A novel fuel consumption prediction model based on vehicle speed, engine speed and engine load was constructed. This decision-making support model accurately predicts real-time fuel consumption based on different driving behaviours, and hence, the driving effectiveness. Both the proposed index and fuel consumption model can be used to support decision-making in new product design and development.
A Systematic Approach to Correlation Analysis of In-Line Process Parameters for Process Variation Effect on Electrical Characteristic of 16-nm HKMG Bulk FinFET Devices
(2016-08)
Su, Ping-Hsun
;
Li, Yiming
;
電信工程研究所
;
Institute of Communications Engineering
This paper reports a systematic method to discover and optimize key fabrication in-line process of 16-nm high-kappa metal gate bulk FinFET to improve device\'s performance and variability. The sensitivity analysis is utilized to prioritize key in-line process parameters which significantly boost device\'s performance and effectively reduce its variations. To extract hidden correlations among complex and a large number of in-line process parameters, data mining technique is applied to highlight and group-associated in-line process parameters. The source of variations of in-line process parameters in each group is revealed and the optimized solution is proposed to reduce its sensitivity to devices\' fluctuation. Results show the dual gate-spacer, the source/drain (S/D) proximity, the S/D depth, and the S/D implant are grouped to the same cluster and significantly affect the threshold voltage (V-t,V- sat), the on-state current (I-d,I- sat), and the off-state current (I-d,I- off), but the key variation source of these parameters is the thickness of the dual gate-spacer. By replacing dual spacers with single spacers, the fluctuation of threshold voltage is 30% dropped.

學術出版;;Publications

Browse

Filters

Settings

Sort By

Results per page

🏛️國立陽明交通大學機構典藏

Search Results