Learn More
Spam identification is crucial in implementing an effective email filtering system, while spam recognization has different properties comparing with normal text recognization. In this paper, we present three different classifiers with detailed analysis on various training data set of the given spam database. We then combine these classifiers into a mixture(More)
We design, implement and evaluate a parallel radix sort that simultaneously utilizes the CPU and GPU devices on the AMD Fusion APU. The parallel sort, referred to as Fusion Sort, partitions the sort keys between the CPU and GPU devices and utilizes the integrated memory system of the APU to avoid data copying between the devices. We identify three design(More)
A single fluorescent signal output often does not satisfy the detection accuracy requirements for bioactive compounds in biological systems. It is of great interest to develop a dual-channel turn-on fluorescence sensing system for self-validated detection. Herein, we report a self-quenching nanoparticle composed of dual two-photon dyes of naphthalimides,(More)
The chemo-, diastereo-, and enantioselectivities in proline and axially chiral amino sulfonamide-catalyzed direct aldol reactions between two enolizable aldehydes with different electronic nature have been studied with the aid of density functional theory (DFT) method. The potential energy profiles for the enamine formation between each aliphatic aldehyde(More)
Function inlining has been considered an important optimization method for compiled languages. In this thesis, we especially study the inlining benefits and present an object-oriented implementation of a source-level C language inliner. Various parsing and inlining topics, such as: message driven system, design patterns, parse-tree and symbol-table(More)
Multi-threshold image segmentation is a powerful image processing technique that is used for the preprocessing of pattern recognition and computer vision. However, traditional multilevel thresholding methods are computationally expensive because they involve exhaustively searching the optimal thresholds to optimize the objective functions. To overcome this(More)
  • 1