scholarly journals A novel RSW&TST framework of MCPs detection for abnormal pattern recognition on large-scale time series and pathological signals in epilepsy

PLoS ONE ◽  
2021 ◽  
Vol 16 (12) ◽  
pp. e0260110
Author(s):  
Jinpeng Qi ◽  
Ying Zhu ◽  
Fang Pu ◽  
Ping Zhang
Keyword(s):  
Time Series ◽  
Pattern Recognition ◽  
Large Scale ◽  
Computing Time ◽  
Optimal Path ◽  
Search Tree ◽  
Change Points ◽  
Binary Search Tree ◽  
Local Optima ◽  
Abnormal Pattern

To quickly and efficiently recognize abnormal patterns from large-scale time series and pathological signals in epilepsy, this paper presents here a preliminary RSW&TST framework for Multiple Change-Points (MCPs) detection based on the Random Slide Window (RSW) and Trigeminal Search Tree (TST) methods. To avoid the remaining local optima, the proposed framework applies a random strategy for selecting the size of each slide window from a predefined collection, in terms of data feature and experimental knowledge. For each data segment to be diagnosed in a current slide window, an optimal path towards a potential change point is detected by TST methods from the top root to leaf nodes with O(log3(N)). Then, the resulting MCPs vector is assembled by means of TST-based single CP detection on data segments within each of the slide windows. In our experiments, the RSW&TST framework was tested by using large-scale synthetic time series, and then its performance was evaluated by comparing it with existing binary search tree (BST), Kolmogorov-Smirnov (KS)-statistics, and T-test under the fixed slide window (FSW) approach, as well as the integrated method of wild binary segmentation and CUSUM test (WBS&CUSUM). The simulation results indicate that our RSW&TST is both more efficient and effective, with a higher hit rate, shorter computing time, and lower missed, error and redundancy rates. When the proposed RSW&TST framework is executed for MCPs detection on pathological ECG (electrocardiogram)/EEG (electroencephalogram) recordings of people in epileptic states, the abnormal patterns are roughly recognized in terms of the number and position of the resultant MCPs. Furthermore, the severity of epilepsy is roughly analyzed based on the strength and period of signal fluctuations among multiple change points in the stage of a sudden epileptic attack. The purpose of our RSW&TST framework is to provide an encouraging platform for abnormal pattern recognition through MCPs detection on large-scale time series quickly and efficiently.

Abnormal Pattern of Sensor Monitoring Data Analysis and Recognition

2012 ◽  
Vol 452-453 ◽  
pp. 863-867 ◽  
Author(s):  
Shi Song Zhu ◽  
Fu Jing Zhu ◽  
Wen Hui Man
Keyword(s):  
Time Series ◽  
Pattern Recognition ◽  
Shape Index ◽  
Recognition System ◽  
Monitoring Data ◽  
Pattern Recognition Problem ◽  
Typical Time ◽  
Abnormal Pattern ◽  
Sensor Monitoring ◽  
Index Table

In order to solve the abnormal pattern recognition problem of the sensor monitoring data automatically, a set of method on the time series similarity measurement is used in this paper. Abnormal time series patterns clustering analysis based on the DTW distance is proposed firstly, thus the typical time series patterns can be obtained. From which the important shape indexes can be extracted and filtered based on piecewise shape measure method, then the shape index table can be established. With which a pattern recognition system can be designed used to recognize these abnormal patterns on real-time. As a case, this method has been used in a high gas coal mine and the important promotion application value has been proved in the sensor monitoring field.

scholarly journals Guidance system based on Dijkstra-ant colony algorithm with binary search tree for indoor parking system

2021 ◽  
Vol 24 (2) ◽  
pp. 1173
Author(s):  
K. Ibrahim Ata ◽  
A. Che Soh ◽  
A. J. Ishak ◽  
H. Jaafar
Keyword(s):  
Shortest Path ◽  
Ant Colony Algorithm ◽  
Optimal Path ◽  
Search Tree ◽  
Binary Search ◽  
Ant Colony ◽  
Binary Search Tree ◽  
Dijkstra Algorithm ◽  
Dijkstra's Algorithm ◽  
Parking System

A common algorithm to solve the single-source shortest path (SSSP) is the Dijkstra algorithm. However, the traditional Dijkstra’s is not accurate and need more time to perform the path in order it should visit all the nodes in the graph. In this paper, the Dijkstra-ant colony algorithm (ACO) with binary search tree (BST) has been proposed. Dijkstra and ACO are integrated to produce the smart guidance algorithm for the indoor parking system. Dijkstra algorithm initials the paths to finding the shortest path while ACO optimizes the paths. BST has been used to store the paths that Dijkstra algorithm initialled. The proposed algorithm is aimed to control the shortest path as well as guide the driver towards the nearest vacant available space near the entrance. This solution depending on applying the optimization on an optimal path while the traditional ACO is optimizing the random path based on the greedy algorithm hence we get the most optimal path. Moreover, the reason behind using the BST is to make the generation of the path by Dijkstra’s algorithm more accurate and less time performance. The results show a range of 8.3% to 26.8% improvement in the proposed path compared to the traditional Dijkstra’s algorithm.

scholarly journals Finding Dense Components in Large-Scale Network Using Randomized Binary Search Tree

2020 ◽  
Vol 30.8 (147) ◽  
pp. 65-69
Author(s):  
Anh Phuc Trinh ◽  
◽  
Dang Hai Pham ◽  
Thi Thuy Dung Phan ◽  
Keyword(s):  
Large Scale ◽  
Complexity Analysis ◽  
Undirected Graph ◽  
Search Tree ◽  
Binary Search ◽  
Binary Search Tree ◽  
Large Scale Network ◽  
The Social ◽  
Vertex Set ◽  
Scale Network

Given a simple undirected graph G=(V, E), the density of a subgraph on vertex set S is defined as a ratio between the number of edges |E(S)| and the number of vertices |S|, where E(S) is the set of edges induced by vertices in S. Finding the maximum density subgraph has become an intense study in recent years, especially in the social network era. Being based on a greedy algorithm that connects with a suitable graph data structure, we have reduced its time complexity by using a randomized binary search tree, also called treap. We make the complexity analysis in both time and memory requirements, including computational experiments in large scale real networks.

scholarly journals Nonparametric Anomaly Detection on Time Series of Graphs

2019 ◽  
Author(s):  
Dorcas Ofori-Boateng ◽  
Yulia R. Gel ◽  
Ivor Cribben
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Change Point ◽  
Large Scale ◽  
Dynamic Networks ◽  
Change Point Detection ◽  
New Method ◽  
Global Network ◽  
Change Points ◽  
Finite Sample

AbstractIdentifying change points and/or anomalies in dynamic network structures has become increasingly popular across various domains, from neuroscience to telecommunication to finance. One of the particular objectives of the anomaly detection task from the neuroscience perspective is the reconstruction of the dynamic manner of brain region interactions. However, most statistical methods for detecting anomalies have the following unrealistic limitation for brain studies and beyond: that is, network snapshots at different time points are assumed to be independent. To circumvent this limitation, we propose a distribution-free framework for anomaly detection in dynamic networks. First, we present each network snapshot of the data as a linear object and find its respective univariate characterization via local and global network topological summaries. Second, we adopt a change point detection method for (weakly) dependent time series based on efficient scores, and enhance the finite sample properties of change point method by approximating the asymptotic distribution of the test statistic using the sieve bootstrap. We apply our method to simulated and to real data, particularly, two functional magnetic resonance imaging (fMRI) data sets and the Enron communication graph. We find that our new method delivers impressively accurate and realistic results in terms of identifying locations of true change points compared to the results reported by competing approaches. The new method promises to offer a deeper insight into the large-scale characterizations and functional dynamics of the brain and, more generally, into intrinsic structure of complex dynamic networks.

Cooperative Co-Evolution and MapReduce

2021 ◽  
Vol 12 (1) ◽  
pp. 29-62
Author(s):  
A. N. M. Bazlur Rashid ◽  
Tonmoy Choudhury
Keyword(s):  
Evolutionary Algorithms ◽  
Large Scale ◽  
Fault Tolerant ◽  
Computing Time ◽  
Search Space ◽  
Population Diversity ◽  
Real Word ◽  
Scale Problem ◽  
Local Optima ◽  
Mapreduce Model

Real-word large-scale optimisation problems often result in local optima due to their large search space and complex objective function. Hence, traditional evolutionary algorithms (EAs) are not suitable for these problems. Distributed EA, such as a cooperative co-evolutionary algorithm (CCEA), can solve these problems efficiently. It can decompose a large-scale problem into smaller sub-problems and evolve them independently. Further, the CCEA population diversity avoids local optima. Besides, MapReduce, an open-source platform, provides a ready-to-use distributed, scalable, and fault-tolerant infrastructure to parallelise the developed algorithm using the map and reduce features. The CCEA can be distributed and executed in parallel using the MapReduce model to solve large-scale optimisations in less computing time. The effectiveness of CCEA, together with the MapReduce, has been proven in the literature for large-scale optimisations. This article presents the cooperative co-evolution, MapReduce model, and associated techniques suitable for large-scale optimisation problems.

Requirements Prioritization Techniques for Global Software Engineering

2017 ◽  
Vol 6 (1) ◽  
Author(s):  
Samina Saghir ◽  
Tasleem Mustafa
Keyword(s):  
Software Engineering ◽  
Requirements Engineering ◽  
Search Tree ◽  
Software Requirements ◽  
Binary Search Tree ◽  
Time Zone ◽  
Analytic Hierarchy ◽  
Requirements Prioritization ◽  
Global Software Engineering ◽  
Hierarchy Process

<p>Increase in globalization of the industry of software requires an exploration of requirements engineering (RE) in software development institutes at multiple locations. Requirements engineering task is very complicated when it is performed at single site, but it becomes too much complex when stakeholder groups define well-designed requirements under language, time zone and cultural limits. Requirements prioritization (RP) is considered as an imperative part of software requirements engineering in which requirements are ranked to develop best-quality software. In this research, a comparative study of the requirements prioritization techniques was done to overcome the challenges initiated by the corporal distribution of stakeholders within the organization at multiple locations. The objective of this study was to make a comparison between five techniques for prioritizing software requirements and to discuss the results for global software engineering. The selected techniques were Analytic Hierarchy Process (AHP), Cumulative Voting (CV), Value Oriented Prioritization (VOP), Binary Search Tree (BST), and Numerical Assignment Technique (NAT). At the end of the research a framework for Global Software Engineering (GSE) was proposed to prioritize the requirements for stakeholders at distributed locations.<strong></strong></p>

Sign in / Sign up

Export Citation Format

Share Document