作者：
Chen, ZhuoLiu, GuanjunDepartment of Computer Science
Key Laboratory of the Ministry of Education for Embedded System and Service Computing Shanghai Electronic Transactions and Information Service Collaborative Innovation Center Tongji University Shanghai China
With the development of online payment, electronic transaction fraud events also take place often and result in huge financial losses. Recently, DenseNet as one of the most prominent Convolutional Neural Networks (CNN...
详细信息
Due to the mobility and frequent disconnections, the correctness of mobile interaction systems, such as mobile robot systems and mobile payment systems, are often difficult to analyze. This paper introduces three crit...
详细信息
LiDAR-based place recognition is an essential and challenging task both in loop closure detection and global relocalization. We propose Deep Scan Context (DSC), a general and discriminative global descriptor that capt...
详细信息
The construction of a smart city needs to be supported by good machine learning methods, random forest is a kind of technology which has been widely applied in dealing with regression or classification problems. The d...
详细信息
ISBN:
(纸本)9781450389099
The construction of a smart city needs to be supported by good machine learning methods, random forest is a kind of technology which has been widely applied in dealing with regression or classification problems. The difficult points to be solved include how to set random forest parameters. The number of subtrees and samples will affect the performance of the algorithm to a great extent. This paper proposes a forest parameter optimization method based on firework algorithm (FWA-RF). By using its explosion, mutation and selection mechanisms, the parameters are adjusted in self-adaptive man-ner to avoid falling into a local optimal state. Through the comparative experiments on 15 UCI databases, the accuracy of the improved algorithm is higher than that of the original RF classification by 1.55% to 26.67%, and is superior to decision tree, Bagging, Adaboost as well as RF based on genetic algorithm. The results indicate that this method can effectively find the range of effective adjustment parameters, make parameter optimization and adjustment easier, and provide technical support for the creation of smart city.
Obstacle avoiding is one of the most complex tasks for autonomous driving systems, which was also ignored by many cutting-edge end-to-end learning-based methods. The difficulties stem from the integrated process of de...
详细信息
Estimating geometric elements such as depth, camera motion, and optical flow from images is an important part of the robot's visual perception. We use a joint selfsupervised method to estimate the three geometric ...
详细信息
Mobile computing systems, service-based systems and some other systems with mobile interacting components have recently received much attention. However, because of their characteristics such as mobility and disconnec...
详细信息
As one of the fundamental tasks in computer vision, semantic segmentation plays an important role in real world applications. Although numerous deep learning models have made notable progress on several mainstream dat...
详细信息
Consistency degree calculation is established on the basis of known correspondence, but in real life, the correspondence is generally unknown, so how to calculate consistency of two models under unknown correspondence...
详细信息
Consistency degree calculation is established on the basis of known correspondence, but in real life, the correspondence is generally unknown, so how to calculate consistency of two models under unknown correspondence has become a problem. For this condition, we should analyze unknown correspondence due to the influence of different *** this paper we obtain the relations of transitions based on event relations using branching processes, and build a behavioral matrix of relations. Based on the permutation of behavioral matrix, we express different correspondences, and define a new formula to compute the maximal consistency degree of two workflow nets. Additionally, this paper utilizes an example to show these definitions, computation as well as the advantages.
Concept Drift is one of the most challenging issues in various applications such as fraud detection, spam filtering and sensor networks, causing the performance degradation of learning algorithms which are dedicated t...
详细信息
ISBN:
(数字)9781728190129
ISBN:
(纸本)9781728190136
Concept Drift is one of the most challenging issues in various applications such as fraud detection, spam filtering and sensor networks, causing the performance degradation of learning algorithms which are dedicated to processing static data such as Naive Bayes, due to their poor adaptability to changes. Several techniques and algorithms have been proposed to address the concept drift problem, e.g. the ensemble methods that have already achieved great success. However, most of the ensemble methods consider either accuracy or diversity of a trained model when adapting to new concepts, and thus they can hardly deal with different types of drifts. This paper proposes a novel ensemble method that considers both accuracy and diversity. We linearly combine accuracy and diversity with different coefficients, and propose an entropy-based policy to compute these coefficients. Experiments on 10 synthetic datasets and 4 public datasets demonstrate that our method outperforms the state-of-the-art ones on dealing with different concept drifts. Our method is also applied to the electronic transaction fraud detection and achieves an excellent performance.
暂无评论