Search

Over the years the number of research publications per year is growing exponentially. Finding research papers of quality from the massive literature of relevant articles is a challenging and time-consuming task. The approaches in the latest literature address citation recommendation by utilizing large bibliographic information and use machine learning and deep learning methods for the task. These techniques clearly require a large amount of training data as well as machines with high processing power. To overcome these issues, we propose a novel method by modifying the popular hyperlink induced topic search (HITS), a web page ranking algorithm, as citation recommendation using hyperlink induced topic search (CR-HITS) that works on a directed and weighted heterogeneous bibliographic network containing diverse types of nodes and edges. We define effective scoring schemes for nodes and edges based on basic bibliographic information like citations of papers, number of publications of an author, etc. Given a few seed papers, the citation recommendation algorithm CR-HITS is run on small neighborhoods of the seed papers and hence the time taken by the execution is very small to yield the final recommendations. To the best of our knowledge, HITS has been used for the first time for the citation recommendation problem. We perform extensive experimentation on DBLP (version-11) and ACM (version-9) datasets and compare the results with many baseline methods in terms of MAP, MRR, and recall@N measures. The performance of the proposed algorithms is superior with respect to the MAP metric and matches the second best for the other two metrics. Since the top two algorithms use deep learning methods and use much larger bibliographic information including abstracts of the papers, we claim that our approach utilizes very low resources, yet yields recommendations that are very close to the top recommendations. [ABSTRACT FROM AUTHOR]

Classification is a widely used supervised learning technique that enables models to discover the relationship between a set of features and a specified label using available data. Its applications span various fields such as engineering, telecommunication, astronomy, and medicine. In this paper, we propose a novel rule-based classifier called RUCIB (RUle-based Classifier Inspired by BRADO), which draws inspiration from the socio-inspired swarm intelligence algorithm known as BRADO. RUCIB introduces two key aspects: the ability to accommodate multiple values for features within a rule and the capability to explore all data features simultaneously. To evaluate the performance of RUCIB, we conducted experiments using ten databases sourced from the UCI machine learning database repository. In terms of classification accuracy, we compared RUCIB to ten well-known classifiers. Our results demonstrate that, on average, RUCIB outperforms Naive Bayes, SVM, PART, Hoeffding Tree, C4.5, ID3, Random Forest, CORER, CN2, and RACER by 9.32%, 8.97%, 7.58%, 7.4%, 7.34%, 7.34%, 7.22%, 5.06%, 5.01%, and 1.92%, respectively. [ABSTRACT FROM AUTHOR]

Machine learning techniques aim to mimic the human ability to automatically learn how to perform tasks through training examples. They have proven capable of tasks such as prediction, learning and adaptation based on experience and can be used in virtually any scientific application, ranging from biomedical, robotic, to business decision applications, and others. However, the lack of domain knowledge for a particular application can make feature extraction ineffective or even unattainable. Furthermore, even in the presence of pre-processed datasets, the iterative process of optimizing Machine Learning parameters, which do not translate from one domain to another, maybe difficult for inexperienced practitioners. To address these issues, we present in this paper a Vectorized Automated ML Pre-processIng and post-pRocEssing framework, approximately named (VAMPIRE), which implements feature extraction algorithms capable of converting large time-series recordings into datasets. Also, it introduces a new concept, the Activation Engine, which is attached to the output of a Multi Layer Perceptron and extracts the optimal threshold to apply binary classification. Moreover, a tree-based algorithm is used to achieve multi-class classification using the Activation Engine. Furthermore, the internet of things gives rise to new applications such as remote sensing and communications, so consequently applying Machine Learning to improve operation accuracy, latency, and reliability is beneficial in such systems. Therefore, all classifications in this paper were performed on the edge in order to reach high accuracy with limited resources. Moreover, forecasts were applied on three unrelated biomedical datasets, and on two other pre-processed urban and activity detection datasets. Features were extracted when required, and training and testing were performed on the Raspberry Pi remotely, where high accuracy and inference speed were achieved in every experiment. Additionally, the board remained competitive in terms of power consumption when compared with a laptop which was optimized using a Graphical Processing Unit. [ABSTRACT FROM AUTHOR]

There are several public cloud service providers (CSPs) across the globe supplying a variety of application, platform, middleware, database, and infrastructure services. The brewing challenge before any cloud user is how to be sure about the trustworthiness of a service being offered by various CSPs. With the overwhelming usage of cloud services by individuals and organizations, this problem has acquired a lot of attention these days. Precisely arriving and articulating that this particular service from a specific cloud CSP is trustworthy is becoming complicated because of many moving variables in this whole phenomenon. There are varying parameters and indicators for decisively proving that a particular service is trustworthy or not. In order to meet up this crucial challenge and concern being widely expressed by cloud customers, researchers and cloud professionals across the world have unearthed a few viable mechanisms. There are fresh algorithms, techniques, and tools to enable cloud users towards easily and quickly selecting trustworthy services. This paper digs deeper and dwells at length about the intrinsic challenges being associated with the provenance of the trustworthiness factor. The paper also presents a machine learning approach to compute the trust factor of CSPs and it is found that the methodology gains advantage when compared to similar works. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel indoor localization system in a multi-indoor environment using cloud computing. Prior studies show that there are always concerns about how to avoid signal occlusion and interference in the single indoor environment. However, we find some general rules to support our system being immune to interference generated by occlusion in the multi-indoor environment. A convenient way is measured to deploy Bluetooth low energy devices, which mainly collect large information to assist localization. A neural network-based classification is proposed to improve localization accuracy, compared with several algorithms and their performance comparison is discussed. We also design a distributed data storage structure and establish a platform considering the storage load with Redis. Our real experimental validation shows that our system will meet the four aspects of performance requirements, which are higher accuracy, less power consumption, and increased levels of system magnitude and deployment efficiency. [ABSTRACT FROM AUTHOR]

Resource management is the process of task scheduling and resource provisioning to provide requirements of cloud users. Since cloud resources are often heterogeneous, task scheduling and resource provisioning are major challenges in this area. Various methods have been introduced to improve resource utilization and thus increase the efficiency of cloud computing. Existing methods can be divided into several categories, including mathematical and statistical methods, heuristic- and meta-heuristic-based methods, and machine-learning-based methods. Since the resource management problem is NP-complete, several optimization methods have been also exploited in this area. Coral reefs algorithm is an evolutionary method that has showed appropriate convergence and response time for some problems, and thus is used in this paper to combine with reinforcement learning to improve efficiency of resource management in cloud environments. The proposed method of this paper consists of two phases. The initial allocation of resources to ready-to-perform tasks is done using the coral reefs algorithm in the first phase. The tasks are considered as corals and the resources are considered reefs in this method. The second phase utilizes reinforcement learning to avoid falling into the local optima and to make optimal use of resources using a long-term approach. The proposed model of this paper, called MO-CRAML, introduces a new hybrid algorithm for improving utilization and load balancing of cloud resources using the combination of coral reefs optimization algorithm and reinforcement learning. The results of the experiments show that the proposed algorithm has better performance in cloud resource utilization and load balancing in comparison with some other important methods of the literature. [ABSTRACT FROM AUTHOR]

This paper concentrates on a parallel acceleration method of optimizing Gaussian hyper-parameters with the maximum likelihood estimation. In the process of optimizing the hyper-parameters, many calculations of the kernel matrix inversion will be operated. With an increase of the kernel matrix scale, the high computation burden will be generated. In order to improve the calculating efficiency, we introduce a decomposing and iterative (DI) algorithm. This algorithm divides the large-scale kernel matrix into four blocks and solves the matrix inversion with constant iterations. Due to the independency of the calculations of the sub-matrix blocks, it is quite suitable to put the sub-matrix blocks computation in graphics processing unit. Hence, the parallel decomposing and iterative (DIP) algorithm is introduced. The inverted pendulum and ball-plate system experiments are carried out to confirm the effectiveness of the DI and DIP algorithms. Based on the simulation results, the proposed DI and DIP algorithms shed light on real engineering application in the future. This paper also provides a practical and feasible approach to accelerate the optimization of hyper-parameters with maximum likelihood estimation. [ABSTRACT FROM AUTHOR]

With growing heterogeneity and complexity in applications, demand to design an energy-efficient and fast computing system in multi-core architecture has heightened. This paper presents a regression-based dynamic voltage frequency scaling model which studies and utilizes workload characteristics to obtain optimal voltage–frequency (v–f) settings. The proposed framework leverages the workload profile information together with power constraints to compute the best-suited voltage–frequency (v–f) settings to (a) maintain global power budget at chip-level, (b) maximize performance while enforcing power constraints at the per-core level. The presented algorithm works in conjunction with the workload characterizer and senses change in application requirements and apply the knowledge to select the next setting for the core. Our results when compared with two state-of-the-art algorithms MaxBIPS and TPEq achieve the average power reduction of 33% and 25% respectively across 32-core architecture for PARSEC benchmarks. [ABSTRACT FROM AUTHOR]

Cloud computing provides utility computing in which clients pay the cost according to their demands and service use. There are some challenges to this technology. One of these issues in data centers is virtual machine (VM) placement so that mapping of these VMs to hosts is executed for a variety of objectives such as load balancing, reducing energy consumption, increasing resource utilization, shortening response time, etc. In this paper, a strategy is presented based on machine learning for VM replacement which aims to balance the load in host machines (HM). In this proposed strategy, the learning agent, in each learning episode by selecting an action from among the permissible actions and executing it on the environment receives a reward according to the desirability of the solution obtained by doing that action in the environment. Receiving a reward from the environment and updating the action value table enable the learner agent to learn in the following episodes that in each environment state, selecting and executing which action is better in the environment and this leads to further enhancement. Our proposed algorithm has, on average, improved the inter-HM load balance in terms of processor, memory, and bandwidth by 25%, 34%, and 32%, respectively, prior to the implementation of the algorithm. Our strategy was compared from diffrent aspects in three scenarios to the MOVMrB strategy. Finally, it was concluded that our proposed algorithm can be more effective in load balancing by having much less runtime and turning off more HMs. [ABSTRACT FROM AUTHOR]

In a large-scale data center, it is vital to precisely recognize the termination statuses of applications at an early stage. In recent years, many machine learning techniques have been applied to this issue, which is beneficial for optimizing the scheduling policy and improving the efficiency of resource utilization. However, if the application's dynamic information is insufficient at the early stage, the generalization performance of the machine learning model will be lessened, and the prediction accuracy could be low. To overcome this problem, a novel failure prediction method that is based on the association relationships between similar jobs is proposed in this paper to jointly predict task's termination statuses at an earlier stage. The similar jobs whose tasks have similar changing modes of consumed resources, an inherent structural correlation may exist, and the correlation information is significant for improving the prediction model's generalization performance. First, a job clustering algorithm is proposed for identifying the jobs with higher similarity from jobs that have various numbers of tasks. Second, based on the job clustering results, the robust multi-task learning algorithm is introduced to effectively utilize the domain information among jobs (i.e. interactional relationship among jobs on the termination statuses of task). Experiments are conducted on a Google cluster workload traces dataset. The results show that the proposed method can realize higher prediction accuracy, lower misjudgment rate, and higher predictive stability than several state-of-the-art methods at 1/3 the running time of the tasks. [ABSTRACT FROM AUTHOR]

Wireless sensor networks have been used in a wide variety of applications. Recently, networks consisting of directional sensors have gained prominence. An important challenge facing directional sensor networks (DSNs) is maximizing the network lifetime while covering all the targets in an area. One effective method for saving the sensors' energy and extending the network lifetime is to partition the DSN into several covers, each of which can cover all targets, and then to activate these covers successively. This paper first proposes a fully distributed algorithm based on irregular cellular learning automata to find a near-optimal solution for selecting each sensor's appropriate working direction. Then, to find a near-optimal solution that can cover all targets with the minimum number of active sensors, a centralized approximation algorithm is proposed based on distributed learning automata. This algorithm takes advantage of learning automata (LA) to determine the sensors that must be activated at each stage. As the presented algorithm proceeds, the activation process is focused on the sensor nodes that constitute the cover set with the minimum number of active sensors. Through simulations, we indicate that the scheduling algorithm based on LA has better performance than the greedy algorithm-based scheme in terms of maximizing network lifetime. [ABSTRACT FROM AUTHOR]

In some real-world problems solved by machine learning it is compulsory for the solution provided to be comprehensible so that the correct decision can be made. It is in this context that this paper compares bagging (one of the most widely used multiple classifier systems) with the consolidated trees construction (CTC) algorithm, when the learning problem to be solved requires the classification made to be provided with an explanation. Bearing in mind the comprehensibility shortcomings of bagging, the Domingos’ proposal, called combining multiple models, has been used to address this problem. The two algorithms have been compared from three main points of view: accuracy, quality of the explanation the classification is provided with, and computational cost. The results obtained show that it is beneficial to use CTC in situations where an explanation is required, because: CTC has a greater discriminating capacity than the explanation extraction algorithm added to bagging; the explanation provided is of a greater quality; it is simpler and more reliable; and CTC is computationally more efficient. [ABSTRACT FROM AUTHOR]