Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2020-09-28 Cooperative journals: 《计算机应用研究》
Abstract： Aiming at the problem that traditional clustering algorithm can't deal with multi view and high dimension data in big data, this paper proposed an intelligent weighted K-means clustering algorithm based on chaos particle swarm optimization algorithm. Firstly, it introduced the coupling degree between clusters to expand the similarity of clusters. Secondly, in order to eliminate the sensitivity of the initial clustering center, it used chaos particle swarm optimization algorithm to obtain the optimal initial clustering center, view weight and feature weight through global search. Then, it introduced an accurate perturbation strategy to improve the performance of chaos particle swarm optimization. Finally, experiments on Apache spark and single node show that the proposed method has better clustering performance under the condition of complex data sets with multiple views and high dimensions.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-04-01 Cooperative journals: 《计算机应用研究》
Abstract： The single image haze removal algorithm based on dark channel priori is currently the most efficient image defogging technology. However, when the image does not fully satisfy the dark channel priori, many artifacts and color distortion will appear in the processed image, so that the method should be modified according to the image situation. This paper proposed the hypothesis that the brighter the scene is, the less credible the dark channel priori will be; the lower the scene saturation is, the less credible the dark channel priori will be. Based on this assumption, an image’s dark channel confidence was designed to limit the dark channel value when the scene does not fully satisfy the dark channel prior. Furthermore, the image was post-enhanced to improve the visual effect. Three typical fog weather images were selected to verify the effectiveness of the proposed method. The experimental results show that the proposed algorithm performs better in preserving color and removing artifacts comparing to some current related defogging algorithms. A new dark channel confidence calculation method is designed to overcome the problem that the estimation of dark channel value will be too large when the image scene does not fully satisfy the dark channel priori. The proposed method improves the adaptability of the dark channel priori defogging model to different foggy scenes.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-04-01 Cooperative journals: 《计算机应用研究》
Abstract： The Goal-Oriented Requirement Language (GRL) focuses on the undetermined requirement, which has been widely used to capture initial requirement of the business system. The correctness of the GRL model is a key to influencing the development quality of the business system. Based on graph category, this paper proposes a model formalization approach to verify the correctness of the GRL model. Firstly, according to the meta-model of the GRL model, the morphism mechanism of the category theory has been applied to describe the relationship between goal node and task node, one goal node and the other goal node, one task node and the other task node. Then, the initial object and terminal object of the category model were added, and the neighborhood sequence was designed to represent the causation between the multiple goals and the task implementations. Finally, the correctness structure properties of the business objective model system were designed. The Web Payment system was applied to demonstrate the result of the formalization analysis and correctness verification. It shows that the graph category model can verify the correctness of GRL model and improve the quality of goal modeling.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-11-29 Cooperative journals: 《计算机应用研究》
Abstract： In the background of information overload, this is challenging to mine and organize meaningful concepts and their semantic connections from a set of related documents under the same topic in open information extraction. Thus, this paper proposed a multi-document conceptual graph model based on open-domain information extraction. Firstly, documents were ranked according to the improved TF-IDF weight of extracted topic words under the predefined topics, then the model relayed on a serious of methods, including coreference resolution, weight computation, open-domain information extraction method to extract numerous representative subject-predicate-object triples from multiple documents. For filtering out the noise of open-domain information approach itself and improving the accuracy of information extraction, this paper presented a fact filtering algorithm to retain only the most salient, compatible facts as well as a form of multiple conceptual subgraphs. Finally, in combined with the equivalent concepts and relationships across different subgraphs to connect into a fully connected conceptual graph with expressive topic ability. Experiments on Signal Media dataset illustrated that the proposed model has the ability to discern and effectively group the key information corresponds to specific topics within and across documents, and formed conceptual graph outperforms state-of-the-art the algorithms in terms of the coverage rate of topic concepts as well as the compatible facts. Besides, this model also has the important significance for the automatic Abstract: on.
Hits
Hits
Downloads
Comment