A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Zainal, Anazida
- Design of Misbehavior Detection Scheme by Combining Lane Change and Braking Alerts
Authors
1 Universiti Teknologi Malaysia. 81310, Johor Bahru, MY
Source
Indian Journal of Science and Technology, Vol 9, No 46 (2016), Pagination:Abstract
Objective: To design and develop an enhanced Misbehaviour Detection Scheme (MDS) that addresses the problem of transmitting false information in Vehicular Ad hoc Network (VANET). Methods/Analysis: To achieve the purpose of this paper, data was collected through simulating a vehicle crash in different traffic scenarios. The data collected was then used to design a Misbehaviour Detection Scheme considering two inputs of Emergency Electronic Brake Light (EEBL) and Lane Change (LC). To confirm the veracity of transmitted Post-Crash Notification (PCN) alert, Bayes’ rule was used to combine the two alert evidences. Findings: In each of the experiments conducted, the scenario belief values (probability of individual events) were calculated and Bayes’ rule was used for combining the two evidences to obtain a better belief value. Simulation results show that increasing vehicle speed improves detection accuracy. Traffic scenarios having vehicles with low speed transmits fewer secondary alerts. Existing MDS uses single secondary alerts for verifying received PCN alerts. The proposed scheme combines evidences from more than one secondary alert to enhance the belief value of received PCN alert. Applications/Improvements: Combining multiple alert evidences shows that the proposed MDS makes significant enhancement to the existing scheme. Testing the proposed scheme with vehicles on high speeds shows 100% detection accuracy for transmitted PCN alerts.Keywords
Braking Alerts, Lane Change, Misbehaviour Detection, Post-Crash Notification, VANET.- A Classification Model using Neuro Fuzzy Classifier for Imbalanced Data
Authors
1 Department of Computer Science and A.I., University of Granada, Granada, ES
2 Department of Computer Science, University of Jaén, ES
Source
Data Mining and Knowledge Engineering, Vol 11, No 9 (2019), Pagination: 161-166Abstract
Most of the biological data often deals with class imbalance problem. This happens due to the heterogeneous data and also several categorical attributes. This induces the researchers to work in this area to handle the data imbalance problem. The main challenges faced in bioinformatics are the manner by which to unravel the logical issues as opposed to concentrating too vigorously on gathering and examining biological information. As a result of the unpredictability, there are various testing research issues in bioinformatics. For the most part, information examination related issues in bioinformatics can be separated into three classes as indicated by the sort of biological data: sequences, structures, and networks. Classification and clustering strategies of data mining plays a critical part to dissect biological data such as genomic/DNA microarray data classification and analysis. Learning from imbalanced datasets is a common problem found in many bioinformatics applications, such as gene prediction, splice site prediction, promoter prediction, protein classification and many more. In this work neuro fuzzy model is presented for the data imbalance classification problem.