Liheng Jian

Learn More
Recent development in Graphics Processing Units (GPUs) has enabled inexpensive high performance computing for general-purpose applications. Compute Unified Device Architecture (CUDA) programming model provides the programmers adequate C language like APIs to better exploit the parallel power of the GPU. Data mining is widely used and has significant(More)
Abnormal remarks on World Wide Web, such as violence, threat, superstition, etc. may disturb the social order and public morality. Most traditional methods filter a page as long as it contains a keyword in a predefined blacklist. Such methods cannot provide a quantitative measure of how sensitive the content is. In this paper, we propose a utility-based Web(More)
Multiple Symbol Detection (MSD) is an important technique in digital signal processing. It estimates the sequence of the received signal by maximum-likelihood principle. Due to its high computational complexity, currently, MSD algorithms were implemented in specialized signal processing devices, such as Field Programmable Gate Arrays (FPGAs). As the rapid(More)
Due to the computational intensive nature, the current available WSN simulators, which are based on the traditional CPU computing architecture , cannot run in a linear scalability. In this paper, we propose and set up CU-Simulator, a parallel radio channel simulator to enhance the performance for simulating data packet transmission in WSNs using NVIDIA's(More)
To deal with the computational and memory intensive problem, we propose and implement CU-Simulator, a parallel radio channel simulator to enhance the performance for simulating data packet transmission in wireless sensor networks (WSNs) using GPU+CUDA. An efficient data structure is proposed for acceleration, called CUDA-quad-trees, which resides in the(More)
Data mining is widely used in various domains and has significant applications. However, current data mining tools cannot meet the requirement of applications with large‐scale databases in terms of speed. We propose three techniques to accelerate fundamental kernels in data mining algorithms on CUDA platform, scalable thread scheduling scheme for irregular(More)
With the rapid growth of information on the World Wide Web, recommender system has been receiving increasing attention. In academic literature recommendation applications, existing methods recommend papers merely based on their contents or cited frequencies, and none of them consider user's personalized requirements, such as authority, popularity, time,(More)
Abnormal remarks on the web, such as violence, threat, superstition, etc., may disturb the social order and public morality (referred as sensitive content). To provide a quantitative measure of the sensitivity of a webpage, we propose the concept of web content sensitivity which measures how sensitive a page is. We also propose a web content sensitivity(More)
  • 1