Recent Advances in Computer Science and Communications - Volume 14, Issue 7, 2021

Aims: The factors including light, weather, dynamic objects, seasonal effects and structures bring great challenges for the autonomous driving algorithm in the real world. Autonomous vehicles can detect different object obstacles in complex scenes to ensure safe driving. Background: The ability to detect vehicles and pedestrians is critical to the safe driving of autonomous vehicles. Automated vehicle vision systems must handle extremely wide and challenging scenarios. Objective: The goal of the work is to design a robust detector to detect vehicles and pedestrians. The main contribution is that the Multi-level Feature Fusion Block (MFFB) and the Detector Cascade Block (DCB) are designed. The multi-level feature fusion and multi-step prediction are used which greatly improve the detection object precision. Methods: The paper proposes a vehicle and pedestrian object detector, which is an end-to-end deep convolutional neural network. The key parts of the paper are to design the Multi-level Feature Fusion Block (MFFB) and Detector Cascade Block (DCB). The former combines inherent multi-level features by combining contextual information with useful multi-level features that combine high resolution but low semantics and low resolution but high semantic features. The latter uses multistep prediction, cascades a series of detectors, and combines predictions of multiple feature maps to handle objects of different sizes. Results: The experiments on the RobotCar dataset and the KITTI dataset show that our algorithm can achieve high precision results through real-time detection. The algorithm achieves 84.61% mAP on the RobotCar dataset and is evaluated on the well-known KITTI benchmark dataset, achieving 81.54% mAP. In particular, the detection accuracy of a single-category vehicle reaches 90.02%. Conclusion: The experimental results show that the proposed algorithm has a good trade-off between detection accuracy and detection speed, which is beyond the current state-of-the-art Refine- Det algorithm. The 2D object detector is proposed in the paper, which can solve the problem of vehicle and pedestrian detection and improve the accuracy, robustness and generalization ability in autonomous driving.

Background: In the era of internet, government of each country is thinking about the role of Electronic governance to facilitate its citizens using Information and Communication Technology (ICT). To achieve this, every government activities/ procedures are transforming to electronic medium with the help of websites, mobile applications, government social media accounts and establishing common service center for rural part of country. But it is matter to assess the satisfaction level of electronic services launched by government. If the satisfaction level is high then adoption of government services would be high. Objective: This objective of this research is finding the suitable factors for adoption of government E-services and ranking them accordingly. Methods: Thus this paper suggests an intelligent fuzzy conjoint technique to rank the factors that affects adoption of E-services launched by government. A survey is done to find the factors which may affect adoption of such electronic governance initiative. Results: Result shows that 62.5% are satisfied with services, 25% are dissatisfied and 12.5% are neutral with respected to e-government services Government has established common service center with highest ranking. Next finding is that customized government portals and mobile applications should be multilingual support. Conclusion: The people are dissatisfied with the standardization of polices and IT laws and self guided services. Artificial Intelligent techniques are also required to implement for smart governance. Currently Citizens are neutral to accept intelligent techniques.

Background: Cloud platform is widely applied in the field of electric power. Virtual machine co-resident attack is one of the fundamental security threats to the existing power cloud platform. Objective: This paper addresses a mechanism so as to defend virtual machine co-resident attack on power cloud platform. Methods: Our defense mechanism wields the DBSCAN algorithm to classify the results through the random forest and adopted improved virtual machine deployment strategy which combines the advantages of random round robin strategy and maximum/minimum resource strategy so as to deploy virtual machines. Results: We made a simulation experiment on power cloud platform of State Grid and verified the effectiveness of proposed defense deployment strategy. Conclusion: After the virtual machine deployment strategy is enhanced, the coverage of the virtual machine is remarkably reduced, which proves that our defense mechanism achieves some effect of defending the virtual machine from virtual machine co-resident attack.

Background: Knowledge discovering rare itemsets mining, and association rules are significant in a transactional dataset. Be that as it may, rare association rules are now and again more intriguing than frequent association rules since rare rules indicate an unforeseen or obscure association. The utilization of rare itemsets mining with item selector is unavoidable and has turned into an emerging field of research; therefore, this subject has numerous challenges. Objective: To perform the revenue examination of the marketing sector by rare itemsets selector by threshold and time series-based prediction technique. Methods: This paper gives the revenue examination of the marketing sector by rare itemsets selector by threshold and time series-based prediction technique. A new algorithm is proposed for locating the rare itemsets by Association Rare itemset Rule Mining (ARIRM) to produce rules and then utility itemsets discovery by the threshold. When the rare patterns are analyzed, the ARIMA model is used to anticipate the revenue. Based on the investigation of rare showcasing data with rules of the mining space, this methodology uses a tree structure to learn the rare items. Results: The test results in the "K" transactions with high revenues discovered utilizing the proposed model contrasted with other existing procedures; this forecast procedure is helpful for upcoming transactions. Conclusion: Based on the investigation of rare showcasing data with rules of the mining space, this methodology uses a tree structure to learn the rare items.

Background: Collusion resistance is preserving the value generated at individual nodes secret so that if any nodes share the information together or collude with each other, the secret values cannot be revealed.It is really a challenging task to securely resist the collusion of cloud server and users Collusion resistance also involves growing the number of people. Objective: Users can achieve an effective and economical approach for data sharing among group members in the cloud with the characters of low maintenance and little management cost. Methods: Hybrid Cache supported One way hash chain technique. Results: To test with the sparse caching HHCS and one-way hash chain model and the various sparse caching configurations are fully implemented by this work. Multiple situations and various scenarios are considered in the experiments and tests for the estimation of HHCS’ performance. The results which are measured in terms of storage units needed for the completion of one internet session and in terms of efficiency (which is the number of hash operations done in one session) are compared and then contrasted. Conclusion: Security improvement is proposed in this paper in group key management method, in order to solve the issue of collusion attack between whole members of the group for a secure group communication. It is shown that using lightweight is easy for the implementation of sparse caching method which can greatly improve the cryptographic one-way hash chain method’s performance which is used widely, for securing the transmission of data in cloud by using two levels of security scheme. A memory-times computation complexity metric has been introduced to help choose the best size of cache depending on storage requirements of the cloud application. For demonstrating various connection behaviors, various cache spacing methods have been evaluated. This paper presented and evaluated the potential of hybrid solution in which one-way hash chains which are divided are provided with caching capacities in order to store the values of session key and retrieve them when needed for authenticating a user session. Lower computational overhead is achieved by us as measured by the count of hash operations and computation time needed to give protection.

Aim: This proposed work is used to develop an improved and robust machine learning model for predicting Myocardial Infarction (MI) could have substantial clinical impact. Objectives: This paper explains how to build machine learning based computer-aided analysis system for an early and accurate prediction of Myocardial Infarction (MI) which utilizes framingham heart study dataset for validation and evaluation. This proposed computer-aided analysis model will support medical professionals to predict myocardial infarction proficiently. Methods: The proposed model utilize the mean imputation to remove the missing values from the data set, then applied principal component analysis to extract the optimal features from the data set to enhance the performance of the classifiers. After PCA, the reduced features are partitioned into training dataset and testing dataset where 70% of the training dataset are given as an input to the four well-liked classifiers as support vector machine, k-nearest neighbor, logistic regression and decision tree to train the classifiers and 30% of test dataset is used to evaluate an output of machine learning model using performance metrics as confusion matrix, classifier accuracy, precision, sensitivity, F1-score, AUC-ROC curve. Results: Output of the classifiers are evaluated using performance measures and we observed that logistic regression provides high accuracy than K-NN, SVM, decision tree classifiers and PCA performs sound as a good feature extraction method to enhance the performance of proposed model. From these analyses, we conclude that logistic regression having good mean accuracy level and standard deviation accuracy compared with the other three algorithms. AUC-ROC curve of the proposed classifiers is analyzed that logistic regression exhibits good AUC-ROC score, i.e around 70% compared to k-NN and decision tree algorithm. Conclusion: From the result analysis, we infer that this proposed machine learning model will act as an optimal decision making system to predict the acute myocardial infarction at an early stage than an existing machine learning based prediction models and it is capable to predict the presence of an acute myocardial Infarction with human using the heart disease risk factors, in order to decide when to start lifestyle modification and medical treatment to prevent the heart disease.

Background: Taxi sharing is an emerging transportation arrangement that has helped improve passengers’ travel efficiency and reduce costs. This study proposes an urban taxi sharing system. Methods: Considering each side congestion of the transport network, their corresponding reliability and failure probability are analyzed. Under the constraints of the number of passengers and their time windows, the analysis is performed on passengers whose optimal path is inclusive. Results: According to the optimal strategy, the different passengers can be arranged into the same taxi to realize the taxi sharing. Then the shared taxi route can be optimized. Conclusion: Due to the reasonable vehicle route planning and passenger combination, these can effectively alleviate the traffic congestion, save the driving time, reduce the taxi no-load rate, and save the driving distance. At last, a numerical example is used to demonstrate the proposed method.

Background: A broad range of nature inspired optimization techniques are proposed in the literature and applied for stock market prediction. They performed notably differently across the stock market datasets. Higher Order Neural Network (HONN) have capacity to expand the input demonstration space with fewer trainable weights and perform high learning capabilities. Methods: This article attempts to construct a Cooperative Optimization Algorithm (COA) framework as an alternative of employing solitary algorithm. The COA considers two optimizations, i.e. a genetic algorithm and an artificial chemical reaction optimization as constituent techniques. The COA framework executes each constituent algorithm with a fraction of the whole computation time budget. It encourages interaction between them, so that they can be benefited from each other. Here, optimal model parameters of two HONNs, i.e. Pi-Sigma Neural Network (PSNN) and Functional Link Artificial Neural Network (FLANN) are searched by COA, hence forming two COA-HONN hybrid models. Results and Conclusions: The models are evaluated on forecasting daily closing prices of five real stock datasets. The experimental results confirm that the COA approach enhances the prediction accuracy over individual algorithm. We conducted the Deibold-Mariano test to check the statistical significance of the proposed models, and it was found to be significant. Hence, the proposed approach can be used as a promising tool for stock market forecasting.