Data Mining - Concepts and Techniques.

The objective of this paper is to give a comprehensive introduction to applied cryptography with an engineer or computer scientist in mind. The emphasis is on the knowledge needed to create practical systems which supports integrity, confidentiality, or authenticity. Topics covered includes an introduction to the concepts in cryptography, attacks against cryptographic systems, key use and handling, random bit generation, encryption modes, and message authentication codes. Recommendations on algorithms and further reading is given in the end of the paper. This paper should make the reader able to build, understand and evaluate system descriptions and designs based on the cryptographic components described in the paper.

[서평]「Applied Cryptography」

Matrix Factorization Techniques for Recommender Systems

Self-Directed Learning, A Guide for Learners and Teachers

In this paper, the analysis of recent advances in genetic algorithms is discussed. The genetic algorithms of great interest in research community are selected for analysis. This review will help the new and demanding researchers to provide the wider vision of genetic algorithms. The well-known algorithms and their implementation are presented with their pros and cons. The genetic operators and their usages are discussed with the aim of facilitating new researchers. The different research domains involved in genetic algorithms are covered. The future research directions in the area of genetic operators, fitness function and hybrid algorithms are discussed. This structured review will be helpful for research and graduate teaching.

/pdf/a-review-on-genetic-algorithm-past-present-and-future-2o5y8ru178.pdf

A review on genetic algorithm: past, present, and future

The content that websites provide were static in Web10, the clients could only acquire information from the website This changed when Web20 was released, websites had the capability to allow clients to comment, share songs/videos and interact with other clients As time goes by, more and more people draw attention from the public by sharing interesting content Attention can be easily transformed to income by adding advertisement to the content, so now there are a lot of talents that make a living out of the content they provide on various platforms over the internet Because of policy or high page views, the content provider often uploads contents to multiple platforms, which is very time consuming and a waste of bandwidth This paper proposes a content managing system that not only will help talents publish their work and status to multiple platforms in an efficient fashion, but also provides backup and maintenance mechanisms

A System of Publishing and Managing Contents in Multiple Web2.0 Websites

Software Defect Prediction (SDP) aims to assess software quality by using machine learning techniques. Recently, by proposing the connectivity-based unsupervised learning method, Zhang et al. have been proven that unsupervised classification has great potential to apply to this problem. Inspiring by this idea, in our work we try to replicate the results of Zhang et al.'s experiment and attempt to improve the performance by examining different techniques at each step of the approach using unsupervised learning methods to solve the SDP problem. Specifically, we try to follow the steps of the experiment described in their work strictly and examine three other clustering methods with four other ways for feature selection besides using all. To the best of our knowledge, these methods are first applied in SDP to evaluate their predictive power. For replicating the results, generally results in our experiments are not as good as the previous work. It may be due to we do not know which features are used in their experiment exactly. Fluid clustering and spectral clustering give better results than Newman clustering and CNM clustering in our experiments. Additionally, the experiments also show that using Kernel Principal Component Analysis (KPCA) or Non-Negative Matrix Factorization (NMF) for feature selection step gives better performance than using all features in the case of unlabeled data. Lastly, to make replicating our work easy, a lightweight framework is created and released on Github.

Unsupervised methods for Software Defect Prediction

The newest version of Globus Toolkit (GT4) adopts services-oriented architecture to provide grid environment based on Web Services. However, Globus Toolkit does not guarantee to reliably send and receive messages during messages passing between Web Services. Furthermore, the messages communication mechanism which Globus Toolkit Java Web Services Core provides is based on the changes of resource properties. In other words, it regards messages as resources properties but that is unreasonable for the perspectives of the programmers. In this research, we integrate GT4 Java WS core and our persistent version of JMS middleware (PFJM) to design reasonable programming styles and provide convenient and useful tools for Web Services development users by wrapping PFJM into PFJM Web Services (PFJM WS). Finally, we give a throughput test of messages communication respectively for GT4 Java WS core and PFJM WS. In the report, we can see that PFJM WS has a higher performance than GT4 Java WS core.

/pdf/extending-globus-toolkit-java-ws-core-to-support-reliable-1aj0bk54oc.pdf

Extending globus toolkit Java WS core to support reliable grid messagisng service

A decentralized commit protocol is proposed for distributed database systems. This protocol is based on the hypercube network topology and is (log/sub 2/N)-2 resilient to site failures, where N is the number of sites in the system. In addition, the number of messages sent among N sites is O(N log/sub 2//sup 2/N) which is only a factor of log/sub 2/N over the message complexity lower bound O(NlnN). >

A Resilient Decentralized Commit Protocol

C/C++ is one of the most common programming languages in introductory computer science courses. For students with varying levels of digital literacy, traditional teaching strategies that begin with C/C++ syntax and concepts appear inappropriate. Moreover, different programming languages are required for specific industrial purposes. To complement traditional teaching methods, we design a learner-centered constrictive strategy. Other than C/C++, students begin with two programming languages. Rather than conventional lectures, students learn to program by modifying, decomposing, and reassembling example codes. The midterm assesses their learning effectiveness. After the midterm, they can choose to take the collegiate programming examination (CPE) or not. If they pass CPE, they can pick between 1) regular class and final exam, or 2) no class after midterm, find a topic for a final project, finish it and present it in the last class. We invited 21 students from a class of 90. Eight students had no prior coding experience. Even though these eight students obtained lower scores in the midterm, they all had significant improvements in the final exams. Among sixteen students, six of whom had no prior coding experience, acknowledged that they learned useful skills in the class. Our results show that students with varying coding backgrounds can benefit from our proposed teaching strategy.

Shyan-Ming Yuan

Papers

A System of Publishing and Managing Contents in Multiple Web2.0 Websites

Unsupervised methods for Software Defect Prediction

Extending globus toolkit Java WS core to support reliable grid messagisng service

A Resilient Decentralized Commit Protocol

Modify, Decompose and Reassemble: Learner-Centered Constructive Teaching Strategy for Introductory Programming Course in College