In this project, my teammates and I want to research a relatively new security vulnerability in bash: Shellshock. According our survey, this vulnerability is most exploited in Common Gateway Interface-based (CGI-based) web server, Dynamic Host Configuration Protocol (DHCP), and Limited OpenSSH servers. To present this vulnerability, we build up a website based on CGI and
Continue Reading “Cryptography and Network Security Final Project”
In the final project, our teammates decide to work on a open machine learning problem on Kaggle — Titanic: Survival Prediction. We try different models in Sklearn package and it turns out that Adaboost algorithm has the best performance.
In this homework, I am asked to use Hadoop to fulfill relatively big data manipulation. It turns out that I need to install VirtualBox with Windows on my Mac before I can install Hadoop on Windows system. After that, I write both Java and Python files to fulfill the required calculation.
In this project, I am given a problem, which can be solved with either general machine learning models: Support Vector Machine (SVM), Random Forest (RF), or Gradient Boosting Machine (GBM), or collaborative models: Matrix Factorization (MF) or Tensor Decomposition (TD). At the end, I decide to work on TD model and generate all required statistic
Continue Reading “Big Data: Language, Compile, and Application Homework 1”
My teammate and I decide to study a sub-field in virtual machine — memory management. Existing solutions can be categorized into different groups based on several standards: 1) memory reclamation or memory sharing in terms of mechanism, 2) memory utilization or memory isolation in terms of emphasis phase, and 3) whether there is modification in
In this project throughout the whole semester, we are asked to target on an open-sourced project with 100,000+ lines of codes, understand its design patterns, and extend some functionalities of it. At the end, we pick up a open-sourced music player, figure out how Android works from scratch, add a shortcut for that app in
In the last homework, I implement an algorithm to find out roots for any polynomial function with double-format coefficients using Newton method. The precision I can deal with in my current code is `1e-10`; however, with numerical guarantees, I believe more precise result is available.
In this project, I am going to implement sparse matrix multiplication in C++ with `blas` library. At the end, my implementation has very similar performance as MATLAB.
Related to last homework. I am asked to continue testing the performance of `blas` library. Triple nested for loop for matrix multiplication takes around 30 minutes to finish the job; however, the pre-built optimized blas can finish the same job in around 1 minute. Other `blas` implementations, such as ATLAS and openBlas, have similar performance.
There are two parts in this homework. Firstly, examine the behavior of minimum double value and maximum double value, i.e. make minimum value smaller and maximum value larger. Secondly, examine the performance for loop with different implementations. At the end, it turns out that MATLAB with built-in `blas` library has much better performance compared to
Recent Comments