Recent Patents on Computer Science - Volume 11, Issue 3, 2018

Background: Large-scale data has brought more challenges in the aspects of efficient storage and access requirements. Due merely to differences in the programming interface and database schema, the emerging new database cannot replace RDBMS completely. Therefore, in a longer period in the future, schema-free databases that will assist RDBMS to address the access bottleneck is a broad solution of big data access in industry and academia. Objective: Since schema-free data has the features of high performance and extendibility, it is generally used as the storage of data cache. But there are few effective solutions to keep high cache hit. The frequent access data is not always guaranteed in the cache. Method: This paper describes Patent Publication Number CN103631972A, titled "Method and System for column-aware data caching", issued by the State Intellectual Property Office of the P.R.C. on December 23, 2013. The caching process includes judging cache hit or miss, updating column access frequency, and change data capture. In order to increase the cache hit rate, the patent is related to cache replacement using column access frequency. There are three circumstances to update column access frequency and maintain cache replacement: transactional updates, non-transactional query, and cache listener. Transactional updates will synchronize the updates of the database to the cache system. Non-transactional query and cache listener will rectify the column access frequency using frequency counter. Results: There are four results. Firstly, column-aware data caching has the features of low query time and high throughput. Secondly, dynamic cache replacement using column access frequency improves the cache hit rate and guarantees eventual cache consistency. Thirdly, cache listener can clean the expired data to guarantee the hot data in the cache. Finally, this column-aware data caching system is transparent to the developers. Cache consistency in this paper is slightly different from the cache coherency issue in distributed environment. Conclusion: The idea and a disclosed embodiment of a patent (Patent CN103631972A, issued by the State Intellectual Property Office of the P.R.C.) are presented, which is based on the distribution of cache management system. In one disclosed embodiment, this method contains access judge, frequency counter, change data collector and data cache. The patent's applicability has been illustrated by efficiently solving automatic cache management.

Background: The rapid growth of data in the healthcare domain is a great challenge for traditional data management system for handling and processing such a huge volume of data. The massive growth of data is inevitable and there arises a quest for identifying an effective storage mechanism which can handle vast dynamic data. The advances in technology have paved way for a solution by means of cloud storage. In the current scenario, Cardio Vascular Disease is the major cause of human mortality across the world. This analysis is the hardcore need in today's medical research for prediction of Cardio Vascular Disease. Methods: Hence, in this paper, the Heart Disease dataset is taken for analysis. Various experiments have been carried out with the dataset to compare the performance of classification algorithms and Support Vector Machine is found to outperform other algorithms. Conclusion: Due to its limitation in handling big data, Parallel Support Vector Machine is adapted for big data analysis.

Background: In the space of a few years, Cloud computing has experienced remarkable growth. Indeed, its economic model based on demand use of hardware and software according to technical criteria such CPU utilization, memory, and bandwidth or package has strongly contributed to the liberalization of computing resources in the world. However, the development of Cloud computing requires the optimization of the performance of the different services offered by Cloud providers to ensure a high level of security, availability, and responsiveness. One major aspect of dealing with performance issues in Cloud computing is the load balancing. In fact, an efficient load balancing contributes to cost decrease and maximizes the availability of resources. Objective: In this paper, we aim to propose an enhanced load balancing strategy to improve the Cloud performance. This paper includes a comparative study of the previous research works conducted in the area of load balancing in the Cloud computing. Method: This research work introduces a global analysis of Cloud computing model that applies the Taguchi concept to highlight the parameters which have the greatest impact on the system performance. Finally, we propose a load balancing system to ensure an efficient response time with the lowest cost. Results: The proposed architecture has demonstrated an improvement in response time and processing time. The simulations carried out within CloudAnalyst platform showed an improvement of almost 11% of the response time recorded by the proposed ant colony algorithm compared to the response time achieved by the Round robin algorithm and improvement of the processing time by nearly 38%. Conclusion: We conclude this article by the potential application of the performance methodology applied and the proposed ant colony algorithm to improve the performance of the Cloud environment.

Background: The cloud storage service allows its users to easily store, manage and share their data using large variety of devices. However, upon outsourcing their data, the users require ways to ensure their data confidentiality without losing their granular and flexible data sharing capabilities. Method: We propose a scheme, based on cipher-text policy attribute based encryption that uses threshold- gate access structures as the main access structure. The proposed scheme is able to maintain data confidentiality while providing data owners with an efficient, flexible, scalable and easily manageable access control and efficient revocation. Results: The Experimental results show that the usage of threshold gate access structures results in a large increase in the efficiency of the encryption operation. Moreover, the proposed revocation process is able to achieve both forward and backward security while maintaining a low overhead on the data owner and the users. Conclusion: The proposed scheme uses threshold gate access structures which increases the flexibility of the access structure and reduces the size of linear secret sharing matrices and the computational overhead of the encryption scheme when having complex access structures. The scheme also introduces an efficient revocation operation that is able to achieve forward and backward security without incurring a large computation, storage or communication overhead.

Background: Epilepsy is one among the neurological disorders, which occurs due to temporary and irregular electrical disturbance in the brain. EEG signals are recorded from the patients with seizure. Gaussian filter is one of the pre-processing approaches, which is mainly applied to remove the unwanted signal and improve the signal quality. Harris operator is utilized for key point localization to remove the bad key points from the pre-processed signal. Here, the Scale Invariant Feature Transform (SIFT) is employed with the Local Binary Pattern (LBP) to identify the key points of the signal and these signal values are considered as the feature vector values, which are applied on Multiclass least square Support Vector Machine (MLS-SVM) with four different coding schemes: Error correcting code output code of MLS-SVM (MLS-SVM_ECOC), minimum output codes of MLS-SVM (MLS-SVM_MOC), one versus one of MLS-SVM (MLS-SVM_1v1), one versus all of ML-SVM (MLS-SVM_1vA) using kernel function to categorize epileptic seizure in Multiclass EEG signals by comparing the test features with trained features. The attained results demonstrated that the proposed technique outperformed others existing methods. Methods: The noise is present in the real-time EEG signals. The noise is removed by applying Gaussian filter to the input signal. Key point is localized to generate the features by using the algorithm of SIFT. Then the local features are obtained by the method of LBP. Finally, the signals are categorized into a seizure and seizure-free signal by comparing the optimized features with trained features in SVM classifier. In this work, a novel classification algorithm i.e., Multiclass least square Support Vector Machine (MLS-SVM) is introduced with four coding schemes: MLS-SVM_ECOC, MLS-SVM_MOC, MLS-SVM_1v1, and MLS-SVM_1vA for multiclass classification of EEG signal. To solve the multiclass categorization in EEG signal, we reformulate the extracted features from the EEG signal to the set of binary classification by utilizing four different output coding schemes. The proposed technique is assessed using the performance metrics like accuracy and compared with other previously existing methods. Results: The performance of the proposed technique is evaluated in terms of accuracy. In the suggested classifiers, we select 5-fold cross-validation. At every time, one fold is kept for testing set and remaining four folds are utilized for training set. The result shows that γ= 1 and σ2=10 is the best optimal parameter combination that produces the highest average classification accuracy 99.22% for MLS-SVM_1v1 as compared to the rest of the three proposed output scheme codings. Conclusion: Epilepsy is the neurological disorder where the electrical movement of the brain is changed because of the person's activity changes from the up-normal level. The symptoms of the seizure are unconsciousness and health issues. It may disturb the complete body or parts of the brain. During evaluation, distinct experimental measures are used to analyze the performance of the classifier. The accuracy of 99.29 % was reported by using the proposed method. From the results, it is validated that the anticipated MLS-SVM_1v1 offers the better results.

Background: A decision-making method based on prospect theory and grey relational analysis is proposed for multi-attribute decision-making with interval-valued intuitionistic fuzzy number and incomplete attribute weight. Methods: According to the prospect theory and the improved exact score function, the prospective value function of interval-valued intuitionistic fuzzy numbers is defined, and the prospect decision matrix is constructed with zero as the decision reference point. The optimization model is established with the objective of maximizing the comprehensive prospective value and the subjective risk preference as the constraint. Results: The decision-maker's risk preference for alternatives is brought into the decision-making behavior, and the attribute weight is determined by grey relational analysis. The comprehensive prospective value of each alternative is calculated by combining the prospect matrix and attribute weight, and the best solution is obtained by ranking. The result of example analysis shows that the decision-making method proposed in this study is suitable for fuzzy decision-making environment and has high application value. Conclusion: The example is given to illustrate the feasibility and effectiveness of the proposed method.

Background: Wireless sensor networks have a wide range of applications such as home automation systems, industrial process monitoring, health care monitoring, and others. It contains a number of sensor nodes that collect data from the environments and send it to a base station. Packet scheduling algorithms are responsible of designing the structure of ready queues and selection of possible packets. They can be classified according to some factors such as data types, data priorities, data deliveries, and number of ready queues. Previous works have used few performance metrics to consider the efficiency of these algorithms. Objective: This paper performs a comparative study uses seven performance metrics to analyze the performance of three new packet scheduling algorithms in wireless sensor networks with two wellknown previous scheduling schemes. The metrics are throughput, packet delivery ratio, average waiting time, average delay, energy consumption, packet loss ratio, and network lifetime. Method: This paper uses a simulator network, which is programmed using Java language and is running on an Intel® Core™ i7 machine. The sensor nodes are deployment randomly and sequentially. Results: The simulation results illustrate that the scheduling algorithms are nearly the same in both deployments in packet delivery ratio and throughput metrics. Sequential deployment is preferred average waiting time, average delay, energy consumption, and packet loss ratio metrics while random deployment is preferred in the network lifetime metric. Conclusion: This paper illustrates the packet scheduling algorithms in wireless sensor networks that are the most suitable for the applications that consider a certain metric.

Background: Bearing is a key component of rotating machinery, and its operating condition directly affects the performance of the whole machine. Methods: Based on the investigation of related papers and patents, a fault diagnosis model using the self-adaptive cuckoo search algorithm combined with BP neural network (SaCSBP) is proposed for the effective identification of bearing fault location and loss degree. With respect to the Selfadaptive Cuckoo Search (SaCS) algorithm, a dimension by dimension improvement strategy is introduced to enhance the local search capability, and the control parameters are then set according to the solution quality. Results: The proposed SaCS is compared with several other algorithms on 14 benchmark functions, and the experimental results demonstrate that SaCS exhibits a better or comparable performance. Moreover, SaCSBP obtains the highest fault recognition accuracy. Conclusion: The proposed method has strong fault tolerance and can accurately identify different types and severities of bearing faults.