Stream-based Approach Research Articles

A data stream-based approach for anomaly detection in surveillance videos

A data stream-based approach for anomaly detection in surveillance videos

Community-Engaged School District Design: A Stream-Based Approach

Community-Engaged School District Design: A Stream-Based Approach

STEAM to R-SLAMET Modification: An Integrative Thematic Play Based Learning with R-SLAMETS Content in Early Child-hood Education

STEAM to R-SLAMET Modification: An Integrative Thematic Play Based Learning with R-SLAMETS Content in Early Child-hood Education

Implementation of Project -Based Learning Model (PjBL) Using STREAM-Based Approach in Elementary Schools

Learning in 21st Century focuses on the development of students learning activities in order to increase their creativity through creative products. This study aims to describe the implementation of Project-based Learning model using STREAM-based approach (Science, Technology, Religion, Engineering, Art, and Mathematics) on students learning activities and the results of creative products made by students. The research method used was descriptive qualitative. The data source was chosen using purposive sampling technique. The research subjects are 3 elementary schools in Banjarmasin city, amounting to 80 students. Data collection techniques were done through observation and product assesment. The results showed that the implementation of the PjBL model using STREAM-based approach had an effect on increasing the result of students learning up to 92.1 (very good), and their activeness in learning activities made them become more creative in designing an interesting product. In this study there were 4 creative products created by the students namely drip irrigation, water cycle, water quality test, and religious-based picture story booklets. The implication of this research is a great opportunity for teachers in elementary schools to use PjBL model using STREAM-based approach in order to increase students’ interest, creativity, and social attitudes in learning.

A Load Identification Method Based on Active Deep Learning and Discrete Wavelet Transform

Non-Intrusive Load Monitoring (NILM) makes it possible for users and energy providers to track the fine-grained energy consumption information of residential and commercial buildings. The load identification methods in NILM usually require labeling many samples for training and evaluation, which is always expensive and time-consuming. In order to reduce the labeling cost, this paper proposed a load identification method based on Active Deep Learning (ADL). In this method, Discrete Wavelet Transform (DWT) was applied to extract high-dimensional appliance features from original current signals. Then a pool-based or stream-based active deep learning model was built to learn the features and select high-value samples that worthy of labeling. A mixed dataset based on three public datasets was formed to evaluate the proposed method and three sampling approaches of active learning. The results showed that the proposed method could significantly reduce labeling cost on large datasets, and the number of samples required is 33% lower than the state-of-the-art method when the F1 score is equal. Compared with pool-based sampling approaches, the stream-based approach's benefits are that the classifier improved and the query frequency decreased with continuous input of samples.

A traffic monitoring stream-based real-time vehicular offence detection approach

ABSTRACTTraffic offences are becoming increasingly serious as traffic volume increases rapidly in large cities, causing serious property damage and threatening public safety. Existing traffic monitoring systems lack the capability of detecting various types of offences in real-time. This paper proposes a novel monitoring stream-based vehicular offence detection algorithm, which discovers various types of offences from high-throughput traffic monitoring stream in real-time. An offence detecting and monitoring system is also designed and implemented. In order to achieve real-time detection, parallel computing techniques are utilized. An optimized data structure, a one producer-multiple consumer model and a re-hash strategy are proposed to reduce the synchronization cost incurred by multiple threads in the parallel implementation. Both real-world data and synthetic data are applied in the experiments. Experimental results demonstrate that the proposed algorithm is able to discover three types of offences from high-throughput traffic monitoring stream in real-time. Scalability is also observed. The experimental results indicate that the proposed system is sufficiently efficient to provide real-time offence detection for major metropolises.

Toward Stream-Based IP Flow Analysis

Analyzing IP flows is an essential part of traffic measurement for cyber security. Based on information from IP flows, it is possible to discover the majority of concurrent cyber threats in highspeed, large-scale networks. Some major prevailing challenges for IP flow analysis include, but are not limited to, analysis over a large volume of IP flows, scalability issues, and detecting cyber threats in real time. In this article, we discuss the transformation of present IP flow analysis into a stream-based approach to face current challenges in IP flow analysis. We examine the possible positive and negative impacts of the transformation and present examples of real-world applications, along with our recommendations. Our ongoing results show that stream-based IP flow analysis successfully meets the above-mentioned challenges and is suitable for achieving real-time network security analysis and situational awareness.

HippogriffDB

As data sets grow and conventional processor performance scaling slows, data analytics move towards heterogeneous architectures that incorporate hardware accelerators (notably GPUs) to continue scaling performance. However, existing GPU-based databases fail to deal with big data applications efficiently: their execution model suffers from scalability limitations on GPUs whose memory capacity is limited; existing systems fail to consider the discrepancy between fast GPUs and slow storage, which can counteract the benefit of GPU accelerators. In this paper, we propose HippogriffDB, an efficient, scalable GPU-accelerated OLAP system. It tackles the bandwidth discrepancy using compression and an optimized data transfer path. HippogriffDB stores tables in a compressed format and uses the GPU for decompression, trading GPU cycles for the improved I/O bandwidth. To improve the data transfer efficiency, HippogriffDB introduces a peer-to-peer, multi-threaded data transfer mechanism, directly transferring data from the SSD to the GPU. HippogriffDB adopts a query-over-block execution model that provides scalability using a stream-based approach. The model improves kernel efficiency with the operator fusion and double buffering mechanism. We have implemented HippogriffDB using an NVMe SSD, which talks directly to a commercial GPU. Results on two popular benchmarks demonstrate its scalability and efficiency. HippogriffDB outperforms existing GPU-based databases (YDB) and in-memory data analytics (MonetDB) by 1-2 orders of magnitude.

An intelligent value stream-based approach to collaboration of food traceability cyber physical system by fog computing

Good advanced food traceability systems help to minimize unsafe or poor quality products in food supply chain through value-based process. From the emerging technologies forthcoming for industry automation, future advanced food traceability system must consider not only cyber physical system (CPS) and fog computing but also value-added business in food supply chain. Accordingly, this study presents a novel intelligent value stream-based food traceability cyber physical system approach integrated with enterprise architectures, EPCglobal and value stream mapping method by fog computing network for traceability collaborative efficiency. Furthermore, the proposed intelligent approach explores distributive and central traceable stream mechanism in assessing the most critical traceable events for tracking and tracing process. Successful case study, software system design and implementation demonstrated the performance of the proposed approach. Furthermore, experiment shows the better results obtained after the simulation execution for intelligent predictive algorithm.

Pedestrian-oriented Flow Characterization

Definitions of flow indicators are derived by adapting Edie's definitions through a stream-based approach and a data-driven spatio-temporal discretization framework. The stream-based approach accounts for the multidirectional nature of pedestrian flows. The data-driven discretization framework, based on three-dimensional Voronoi tessellations, accounts for the pedestrian heterogeneity in the case when only samples of points are available, instead of the analytical description of trajectories. The proposed approach enables the pedestrian flow characterization to be applicable to any space-time domain of interest.

A computational model of time for stiff hybrid systems applied to control synthesis

Computation has quickly become of paramount importance in the design of engineered systems, both to support their features as well as their design. Tool support for high-level modeling formalisms has endowed design specifications with executable semantics. Such specifications typically include not only discrete-time and discrete-event behavior, but also continuous-time behavior that is stiff from a numerical integration perspective. The resulting stiff hybrid dynamic systems necessitate variable-step solvers to simulate the continuous-time behavior as well as solver algorithms for the simulation of discrete-time and discrete-event behavior. The combined solvers rely on complex computer code which makes it difficult to directly solve design tasks with the executable specifications. To further leverage the executable specifications in design, this work aims to formalize the semantics of stiff hybrid dynamic systems at a declarative level by removing implementation detail and only retaining ‘what’ the computer code does and not ‘how’ it does it. A stream-based approach is adopted to formalize variable-step solver semantics and to establish a computational model of time that supports discrete-time and discrete-event behavior. The corresponding declarative formalization is amenable to computational methods and it is shown how model checking can automatically generate, or synthesize, a feedforward control strategy for a stiff hybrid dynamic system. Specifically, a stamper in a surface mount device is controlled to maintain a low acceleration of the stamped component for a prescribed minimum duration of time.

On the Structure of Time in Computational Semantics of a Variable-Step Solver for Hybrid Behavior Analysis

AbstractHybrid dynamic systems combine continuous and discrete behavior. Often, computational approaches approximate behavior of an analytic solution, for example, numerical integration to approximate differential equation behavior. The accuracy and computational efficiency of the integration usually depend on the complexity of the method and its implicated approximation errors, especially when repeated over iterations. This work formally defines the computational semantics of a solver in a denotational sense so as to analyze discrete- and continuous-time behavior of time-based block diagram models. A stream-based approach is used to analyze the numerical integration implemented by the solver. The resulting solver applies the principle of nonmonotonic time and so consecutive values may be computed in a temporally nonmonotonic manner. This allows shifting the evaluation points backward and forward in time. Stratification recovers a partially ordered structure in time. Solver dynamics are thus made explicit and can be studied in concert with behavior of discontinuous models parts.

Parallel LDPC Decoding on GPUs Using a Stream-Based Computing Approach

Low-Density Parity-Check (LDPC) codes are powerful error correcting codes adopted by recent communication standards. LDPC decoders are based on belief propagation algorithms, which make use of a Tanner graph and very intensive message-passing computation, and usually require hardware-based dedicated solutions. With the exponential increase of the computational power of commodity graphics processing units (GPUs), new opportunities have arisen to develop general purpose processing on GPUs. This paper proposes the use of GPUs for implementing flexible and programmable LDPC decoders. A new stream-based approach is proposed, based on compact data structures to represent the Tanner graph. It is shown that such a challenging application for stream-based computing, because of irregular memory access patterns, memory bandwidth and recursive flow control constraints, can be efficiently implemented on GPUs. The proposal was experimentally evaluated by programming LDPC decoders on GPUs using the Caravela platform, a generic interface tool for managing the kernels' execution regardless of the GPU manufacturer and operating system. Moreover, to relatively assess the obtained results, we have also implemented LDPC decoders on general purpose processors with Streaming Single Instruction Multiple Data (SIMlD) Extensions. Experimental results show that the solution proposed here efficiently decodes several codewords simultaneously, reducing the processing time by one order of magnitude.