scholarly journals Heterogeneous Defect Prediction Based on Transfer Learning to Handle Extreme Imbalance

2020 ◽  
Vol 10 (1) ◽  
pp. 396 ◽  
Author(s):  
Kaiyuan Jiang ◽  
Yutong Zhang ◽  
Haibin Wu ◽  
Aili Wang ◽  
Yuji Iwahori
Keyword(s):  
Transfer Learning ◽  
Traffic Control ◽  
High Speed ◽  
Negative Impact ◽  
Prediction Method ◽  
Defect Prediction ◽  
Software Systems ◽  
Critical Systems ◽  
Public Domain Software ◽  
Heterogeneous Defect Prediction

Software systems are now ubiquitous and are used every day for automation purposes in personal and enterprise applications; they are also essential to many safety-critical and mission-critical systems, e.g., air traffic control systems, autonomous cars, and Supervisory Control And Data Acquisition (SCADA) systems. With the availability of massive storage capabilities, high speed Internet, and the advent of Internet of Things devices, modern software systems are growing in both size and complexity. Maintaining a high quality of such complex systems while manually keeping the error rate at a minimum is a challenge. This paper proposed a heterogeneous defect prediction method considering class extreme imbalance problem in real software datasets. In the first stage, Sampling with the Majority method (SWIM) based on Mahalanobis Distance is used to balance the dataset to reduce the influence of minority samples in defect data. Due to the negative impact of uncorrelated features on the classification algorithm, the second stage uses ensemble learning and joint similarity measurement to select the most relevant and representative features between the source project and the target project. The third phase realizes the transfer learning from the source project to the target project in the Grassmann manifold space. Our experiments, conducted using nine projects of three public domain software defect libraries and compared with four existing advanced methods to verify the effectiveness of the proposed method in this paper. The experimental results indicate that the proposed method is more accurate in terms of Area under curve (AUC).

Knowledge Engineering and Embedded Systems Software: Educational Challenges

2007 ◽  
Vol 11 (7) ◽  
pp. 842-847
Author(s):  
Kosuke Imamura ◽  
Keyword(s):  
Embedded Systems ◽  
Intelligent Systems ◽  
Knowledge Engineering ◽  
Negative Impact ◽  
Knowledge Bases ◽  
Software Systems ◽  
Critical Systems ◽  
Software Failures ◽  
Educational Challenges ◽  
Daunting Task

There is need for engineering tools to build knowledge bases suitable for realtime perception and control. As embedded systems are applied to complex systems, development of software becomes a daunting task and software failures become more likely with possible negative impact on our economy. Realtime, embedded systems are also common in critical systems so that failures in these systems could affect human lives. To counter this situation, application of knowledge engineering is sought. However, undergraduate software engineering curriculum for embedded systems doesn’t satisfy the current needs. We recently redesigned our embedded systems course to focus on issues and problems in embedded software systems to prepare our students for high level embedded applications and possible development of realtime/embedded intelligent systems.

Workload Perception of Air Traffic Control Officers and Pilots During Continuous Descent Operations Approach Procedures

2019 ◽  
Vol 9 (1) ◽  
pp. 2-11
Author(s):  
Marina Efthymiou ◽  
Frank Fichert ◽  
Olaf Lantzsch
Keyword(s):  
Fuel Consumption ◽  
Traffic Control ◽  
Air Traffic Control ◽  
Negative Impact ◽  
Air Traffic ◽  
Reduce Fuel Consumption ◽  
Total Load ◽  
Open Path

Abstract. The paper examines the workload perceived by air traffic control officers (ATCOs) and pilots during continuous descent operations (CDOs), applying closed- and open-path procedures. CDOs reduce fuel consumption and noise emissions. Therefore, they are supported by airports as well as airlines. However, their use often depends on pilots asking for CDOs and controllers giving approval and directions. An adapted NASA Total Load Index (TLX) was used to measure the workload perception of ATCOs and pilots when applying CDOs at selected European airports. The main finding is that ATCOs’ workload increased when giving both closed- and open-path CDOs, which may have a negative impact on their willingness to apply CDOs. The main problem reported by pilots was insufficient distance-to-go information provided by ATCOs. The workload change is important when considering the use of CDOs.

Redisign of the Air Traffic Control System-Moving People through Distributed Management

1978 ◽  
Vol 22 (1) ◽  
pp. 485-485
Author(s):  
John G. Kreifeldt
Keyword(s):  
Decision Making ◽  
Control System ◽  
Traffic Control ◽  
Air Traffic Control ◽  
Negative Impact ◽  
Air Traffic ◽  
Critical Function ◽  
Congressional Hearings ◽  
The Future ◽  
Traffic Control System

The present national Air Traffic Control system is a ground-centralized, man intensive system which through design allows relatively little meaningful pilot participation in decision making. The negative impact of this existing design can be measured in delays, dollars and lives. The FAA's design plans for the future ATC system will result in an even more intensive ground-centralized system with even further reduction of pilot decision making participation. In addition, controllers will also be removed from on-line decision making through anticipated automation of some or all of this critical function. Recent congressional hearings indicate that neither pilots nor controllers are happy or sanguine regarding the FAA's design for the future ATC system.

scholarly journals Reliable Learning with PDE-Based CNNs and DenseNets for Detecting COVID-19, Pneumonia, and Tuberculosis from Chest X-Ray Images

Mathematics ◽  
2021 ◽  
Vol 9 (4) ◽  
pp. 434
Author(s):  
Anca Nicoleta Marginean ◽  
Delia Doris Muntean ◽  
George Adrian Muntean ◽  
Adelina Priscu ◽  
Adrian Groza ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Negative Impact ◽  
X Ray ◽  
Domain Specific ◽  
Direct Training ◽  
Public Data ◽  
Normal Chest ◽  
Chest X Ray ◽  
The Stability ◽  
Definition Of

It has recently been shown that the interpretation by partial differential equations (PDEs) of a class of convolutional neural networks (CNNs) supports definition of architectures such as parabolic and hyperbolic networks. These networks have provable properties regarding the stability against the perturbations of the input features. Aiming for robustness, we tackle the problem of detecting changes in chest X-ray images that may be suggestive of COVID-19 with parabolic and hyperbolic CNNs and with domain-specific transfer learning. To this end, we compile public data on patients diagnosed with COVID-19, pneumonia, and tuberculosis, along with normal chest X-ray images. The negative impact of the small number of COVID-19 images is reduced by applying transfer learning in several ways. For the parabolic and hyperbolic networks, we pretrain the networks on normal and pneumonia images and further use the obtained weights as the initializers for the networks to discriminate between COVID-19, pneumonia, tuberculosis, and normal aspects. For DenseNets, we apply transfer learning twice. First, the ImageNet pretrained weights are used to train on the CheXpert dataset, which includes 14 common radiological observations (e.g., lung opacity, cardiomegaly, fracture, support devices). Then, the weights are used to initialize the network which detects COVID-19 and the three other classes. The resulting networks are compared in terms of how well they adapt to the small number of COVID-19 images. According to our quantitative and qualitative analysis, the resulting networks are more reliable compared to those obtained by direct training on the targeted dataset.

scholarly journals Testing embedded systems using test cases generated through combinatorial techniques

2018 ◽  
Vol 7 (2.7) ◽  
pp. 146
Author(s):  
Lakshmi Prasad Mudarakola ◽  
J K.R. Sastry ◽  
V Chandra Prakash
Keyword(s):  
Embedded Systems ◽  
Embedded System ◽  
Traffic Control ◽  
Nuclear Reactor ◽  
Test Cases ◽  
Critical Systems ◽  
Daily Lives ◽  
Intelligent Devices ◽  
Wireless Infrastructure ◽  
Combinatorial Methods

Thorough testing of embedded systems is required especially when the systems are related to monitoring and controlling the mission critical and safety critical systems. The embedded systems must be tested comprehensively which include testing hardware, software and both together. Embedded systems are highly intelligent devices that are infiltrating our daily lives such as the mobile in your pocket, and wireless infrastructure behind it, routers, home theatre system, the air traffic control station etc. Software now makes up 90% of the value of these devices. In this paper, authors present different methods to test an embedded system using test cases generated through combinatorial techniques. The experimental results for testing a TMCNRS (Temperature Monitoring and Controlling Nuclear Reactor System) using test cases generated from combinatorial methods are also shown.

Sign in / Sign up

Export Citation Format

Share Document