Logo Freelancer Bagaimana Ia BerfungsiLayari Pekerjaan
Log Masuk Daftar
Siarkan Projek
Profile cover photo
Anda kini mengikuti
Ralat mengikuti pengguna.
Pengguna ini tidak membenarkan pengguna untuk mengikuti mereka.
Anda telah mengikuti pengguna ini.
Pelan keahlian anda hanya membenarkan 0 mengikuti. Naik taraf di sini.
Berjaya menyahikut
Ralat tidak mengikuti pengguna.
Anda berjaya mengesyorkan
Ralat mengesyorkan pengguna.
Sesuatu telah berlaku. Sila segar semula halaman dan cuba lagi.
E-mel berjaya disahkan.
Avatar Pengguna
$10 USD / jam
Bendera UNITED STATES
philadelphia, united states
$10 USD / jam
Sekarang jam 12:41 PTG di sini
Menyertai Jun 2, 2012
0 Syor

Wenbin Z.

@iqlowest

 0.0 (0 ulasan)
0.0
0.0
0%
0%
$10 USD / jam
Bendera UNITED STATES
philadelphia, united states
$10 USD / jam
N/A
Pekerjaan Disiapkan
N/A
Mengikut Bajet
N/A
Tepat Pada Masa
N/A
Kadar Upah Semula

University of Pennsylvania

Ready for Helping :-)
Freelancer C Programmers United States
Lihat semua C Programmers dalam United States

Hubungi Wenbin Z. tentang pekerjaan anda

Log masuk untuk membincangkan sebarang butiran melalui sembang.

Portfolio

We propose an anomaly detection system based on user profiling analysis to promote its performance. It presents a method to extract user's behavior by carefully choosing features of the user's data. It improves performance of Principle Component Analysis (PCA) algorithm by sampling and online PCA methods. In this way it fits the demands of anomaly detection systems and thus improves the efficiency and stability to achieve a better user profiling. In addition, we apply our model into applications of databases and web browsers. Using data mining methods we successfully deduce the parameters in our model and then deploy it in real environment.
User Profiling Analysis based Anomaly Detection System
We propose an anomaly detection system based on user profiling analysis to promote its performance. It presents a method to extract user's behavior by carefully choosing features of the user's data. It improves performance of Principle Component Analysis (PCA) algorithm by sampling and online PCA methods. In this way it fits the demands of anomaly detection systems and thus improves the efficiency and stability to achieve a better user profiling. In addition, we apply our model into applications of databases and web browsers. Using data mining methods we successfully deduce the parameters in our model and then deploy it in real environment.
User Profiling Analysis based Anomaly Detection System
We propose an anomaly detection system based on user profiling analysis to promote its performance. It presents a method to extract user's behavior by carefully choosing features of the user's data. It improves performance of Principle Component Analysis (PCA) algorithm by sampling and online PCA methods. In this way it fits the demands of anomaly detection systems and thus improves the efficiency and stability to achieve a better user profiling. In addition, we apply our model into applications of databases and web browsers. Using data mining methods we successfully deduce the parameters in our model and then deploy it in real environment.
User Profiling Analysis based Anomaly Detection System
Developed a lightweight single sign-on system based on plug-ins. Facilitated users to access multiple services via one authentication by combining browser, BHO and DOM technology into an integrated system. We have built a SSO model based on B / S structure.This model uses plug-in technology. Client starts local enterprise applications and interacts with SSO server through plug-in to complete the automatic login.There are two major concerns in the designing of this system: security and convenience. For security, data in the databases are stored encrypted. Data transmission between the two ends is secured by SSL protocol. For convenience, the browser plug-in is under unified management and distribution, thus toreduce the costs of maintaining and updating the client. And the system does not require additional operations that users can easily get started.
Single Sign-On System
We have developed a system to automatically categorize comments as positive or negative based on SVM. Predicted the rating of a review given its text and other properties from Amazon. In this project, we tried a lot of machine learning approaches to address sentiment analysis problems on Amazon. We first eliminated words that are non-related to ratings, most of which are objective words. We used methods including frequency-based word-probability, stemming, PCA and so on. Then we added many useful additional features, like data, title and bigram features. Furthermore, we tried and compared standard machine learning approaches. During the competition, we constantly utilized cross validation to evaluate our performance. After trying SVM, Adaboost, logistic regression, knn and our own kernel, we found that a combination of several features after performing PCA using liblinear could give the us best performance: RMSE of 0.8529.
Sentiment Analysis System for Amazon Views
In this project, we built a cloud-base search engine called MiniGoogle. Data of the system are crawled and computed based on distributed system. Final search result returned will sorted by page rank, value of TF/IDF and other 5 features. Users also have access to a friendly user interface which provides concise way of searching and visualization of search history. Search results can be returned in around 0.5 seconds. The project involves four parts, each of which is loosely coupled with the others: Crawler Indexer/TF-IDF Retrieval Engine PageRank Search Engine and User Interface
MiniGoogle
In this project, we built a cloud-base search engine called MiniGoogle. Data of the system are crawled and computed based on distributed system. Final search result returned will sorted by page rank, value of TF/IDF and other 5 features. Users also have access to a friendly user interface which provides concise way of searching and visualization of search history. Search results can be returned in around 0.5 seconds. The project involves four parts, each of which is loosely coupled with the others: Crawler Indexer/TF-IDF Retrieval Engine PageRank Search Engine and User Interface
MiniGoogle
In this project, we built a cloud-base search engine called MiniGoogle. Data of the system are crawled and computed based on distributed system. Final search result returned will sorted by page rank, value of TF/IDF and other 5 features. Users also have access to a friendly user interface which provides concise way of searching and visualization of search history. Search results can be returned in around 0.5 seconds. The project involves four parts, each of which is loosely coupled with the others: Crawler Indexer/TF-IDF Retrieval Engine PageRank Search Engine and User Interface
MiniGoogle
In this project, we built a cloud-base search engine called MiniGoogle. Data of the system are crawled and computed based on distributed system. Final search result returned will sorted by page rank, value of TF/IDF and other 5 features. Users also have access to a friendly user interface which provides concise way of searching and visualization of search history. Search results can be returned in around 0.5 seconds. The project involves four parts, each of which is loosely coupled with the others: Crawler Indexer/TF-IDF Retrieval Engine PageRank Search Engine and User Interface
MiniGoogle
In this project, we built a cloud-base search engine called MiniGoogle. Data of the system are crawled and computed based on distributed system. Final search result returned will sorted by page rank, value of TF/IDF and other 5 features. Users also have access to a friendly user interface which provides concise way of searching and visualization of search history. Search results can be returned in around 0.5 seconds. The project involves four parts, each of which is loosely coupled with the others: Crawler Indexer/TF-IDF Retrieval Engine PageRank Search Engine and User Interface
MiniGoogle

Ulasan

Perubahan disimpan
Tiada ulasan untuk dilihat di sini!

Pengalaman

RA

University of Pennsylvania
Jan 2012 - Mei 2012 (4 bulan, 1 hari)
RA for the project "Accountable Virtual Machines" at University of Pennsylvania. Mainly responsible for the design of the deterministic replay component.

Internship

Huaxin Information Security Company, China
Mei 2010 - Ogo 2010 (3 bulan, 1 hari)
Gained experience in developing .NET security products and in configuring and maintaining network. Designed and implemented classification system for the protection of information as a final project.

Pendidikan

Master of Science

University of Pennsylvania, United States 2011 - 2012
(1 tahun)

Bachelor of Science

Northeastern University, China 2007 - 2011
(4 tahun)

Penerbitan

Distributed Center-Controlled Based Network Auditing Architecture Design

ISAI 2010
Distributed Center-Controlled Based Network Auditing Architecture Design

Lightweight Single Sign-On System based on Plug-in Technology

ISAI 2010
Developed a lightweight single sign-on system based on plug-ins. Facilitated users to access multiple services via one authentication by combining browser, BHO and DOM technology into an integrated system.

Hubungi Wenbin Z. tentang pekerjaan anda

Log masuk untuk membincangkan sebarang butiran melalui sembang.

Pengesahan

Pekerja Bebas Pilihan
Identiti Disahkan
Pembayaran Disahkan
Telefon Disahkan
E-mel Disahkan
Facebook Dihubungkan

Pensijilan

freelancer_orientation.png Freelancer Orientation 1 75%

Kemahiran Teratas

C Programming
Java
JavaScript
XML
.NET

Layari Pekerja Bebas Serupa

C Programmers in United States
C Programmers
Java Developers
Javascript Developers

Layari Pameran Serupa

C Programming
Java
JavaScript
XML
Pengguna Sebelumnya Pengguna Seterusnya
Jemputan berjaya dihantar!
Logo Freelancer Malaysia / Bahasa Malaysia Bantuan & Sokongan Pekerja bebas Kategori ProjekPeraduanPekerja BebasEnterprise KeahlianProgram Pekerja Bebas PilihanPengurusan ProjekPekerjaan TempatanPhoto AnywherePameranAPI untuk PembangunDisahkanAplikasi Desktop TentangTentang kamiBagaimana ia BerfungsiKeselamatanPelaburPeta LamanKisahDalam BeritaPasukanAnugerahSiaran AkhbarKerjaya Terma Dasar PrivasiTerma dan SyaratDasar HakciptaTatakelakuanYuran dan Caj Rakan KongsiEscrow.comLoadshiftWarrior Forum AplikasiLogo Apple App StoreLogo Mainan Google
Salin ke papan keratan gagal, sila cuba lagi selepas melaraskan kebenaran anda.
Disalin ke papan keratan.
Pengguna Berdaftar Jumlah Pekerjaan Disiarkan
Freelancer ® is a registered Trademark of Freelancer Technology Pty Limited (ACN 142 189 759)
Copyright © 2023 Freelancer Technology Pty Limited (ACN 142 189 759)
Memuatkan pratonton
Kebenaran diberikan untuk Geolocation.
Sesi log masuk anda telah luput dan telah dilog keluar. Sila log masuk sekali lagi.