Learn More
Firstly, the rapid design of Digital FIR filter was completed by using the Signal Processing Toolbox FDATool, the designed filter deal with a compound signal by filtering, in order to prove whether the designed Filter contented filtering. MATLAB programs and Simulink model of the filter were used to verify the filter performance in MATLAB. Experimental(More)
Traditional architecture design approaches suffer from the crosscutting features and behaviors, which scattered and tangled in the components and connectors, result in the final architecture solutions uneasy to evolve and reuse. In this paper, we propose an approach for multi-dimensional separating crosscutting concerns at architecture level, including: (1)(More)
Software architecture design is an important stage in the whole software development lifecycle. There are so many various concerns at architecture level which relate to all kinds of stakeholders and describe the all facets of problem domain. Unfortunately, they often tangle and scatter in the components and connectors, result in the final architecture(More)
Reusing software architecture, which is a kind of coarse-grained software resources at design time, is always a very difficult problem in the realm of software engineer. We consider that the lack of information and techniques, which support the process of reusing software architecture, may result in this problem. So this paper proposes an approach, called(More)
The Semantic Programming Language (SPL) employs concepts and its relationships in ontology to specify the data used within a process and implements data flow via semantic variables. Therefore, one of the most challenging aspects in Semantic Web Services orchestration realized by SPL is transforming ontological represented data into XML data with equivalent(More)
Image sentiment classification, which aims to predict the polarities of sentiments conveyed by the images, has gained a lot of attention. Most existing methods address this problem by training a general classifier with certain visual features, ignoring the discrepancies across domains. In this paper, we propose a novel weighted co-training method for(More)
  • 1