Effective hierarchical clustering based on structural similarities in nearest neighbor graphs

Wu, Chunrong; Peng, Qinglan; Lee, Jia; Leibnitz, Kenji; Xia, Yunni

doi:10.1016/j.knosys.2021.107295

Cited by 40 publications

(10 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we apply the agglomerative hierarchical clustering technique to the steam flooding data set because this method is straightforward and has a stable structure, which allows for a fully customized design in the algorithm (e.g., number of clusters cut off, distance, and linkage) and prevents the “black-box” processing information from being stored as in other algorithms (like ANN). 43 , 44 The structure and outcomes of HCA can be presented in a dendrogram and scatterplot, which depicts the closeness among all projects, detects special cases, and reveals the hidden pattern in the data set. The framework of the implementation of HCA in this work is made up of six main steps: Perform data preprocessing.…”

Section: Methodologiesmentioning

confidence: 99%

Pattern Recognition for Steam Flooding Field Applications Based on Hierarchical Clustering and Principal Component Analysis

et al. 2022

View full text Add to dashboard Cite

Steam flooding is a complex process that has been considered as an effective enhanced oil recovery technique in both heavy oil and light oil reservoirs. Many studies have been conducted on different sets of steam flooding projects using the conventional data analysis methods, while the implementation of machine learning algorithms to find the hidden patterns is rarely found. In this study, a hierarchical clustering algorithm (HCA) coupled with principal component analysis is used to analyze the steam flooding projects worldwide. The goal of this research is to group similar steam flooding projects into the same cluster so that valuable operational design experiences and production performance from the analogue cases can be referenced for decision-making. Besides, hidden patterns embedded in steam flooding applications can be revealed based on data characteristics of each cluster for different reservoir/fluid conditions. In this research, principal component analysis is applied to project original data to a new feature space, which finds two principal components to represent the eight reservoir/fluid parameters (8D) but still retain about 90% of the variance. HCA is implemented with the optimized design of five clusters, Euclidean distance, and Ward’s linkage method. The results of the hierarchical clustering depict that each cluster detects a unique range of each property, and the analogue cases present that fields under similar reservoir/fluid conditions could share similar operational design and production performance.

show abstract

Section: Methodologiesmentioning

confidence: 99%

Pattern Recognition for Steam Flooding Field Applications Based on Hierarchical Clustering and Principal Component Analysis

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Hierarchical clustering allows for better performance in grouping heterogeneous and non-circular data sets than centered clustering, with increasing temporal complexity. Meanwhile, the bottom-up approach to hierarchical clustering methods often tends to be sensitive to datasets containing ambiguous cluster boundaries [43]. In order to solve the problem of hierarchical classification in traditional evaluation, a hierarchical clustering method has been developed for unknown grading standards.…”

Section: Hierarchical Clustering Methodsmentioning

confidence: 99%

Cell Production System Design: A Literature Review

Ghahremani-Nahr

Bathaee

Mazloumzadeh

et al. 2021

IJIMES

View full text Add to dashboard Cite

In a cell production system, a number of machines that differ in function are housed in the same cell. The task of these cells is to complete operations on similar parts that are in the same group. Determining the family of machine parts and cells is one of the major design problems of production cells. Cell production system design methods include clustering, graph theory, artificial intelligence, meta-heuristic, simulation, mathematical programming. This article discusses the operation of methods and research in the field of cell production system design. Methodology:To examine these methods, from 187 articles published in this field by authoritative scientific sources, based on the year of publication and the number of restrictions considered and close to reality, which are searched using the keywords of these restrictions and among them articles Various aspects of production and design problems, such as considering machine costs and cell size and process routing, have been selected simultaneously.Findings: Finally, the distribution diagram of the use of these methods and the limitations considered by their researchers, shows the use and efficiency of each of these methods. By examining them, more efficient and efficient design fields of this type of production system can be identified.Originality/Value: In this article, the literature on cell production system from 1972 to 2021 has been reviewed.

show abstract

“…Recent academic literature [24,38] has explored multiple clustering of intrusion detection systems. There are many different techniques we can use in this domain.…”

Section: Clustering Techniques For the Intrusion Detection Systemmentioning

confidence: 99%

A Survey on Intrusion Detection Systems for Fog and Cloud Computing

et al. 2022

View full text Add to dashboard Cite

The rapid advancement of internet technologies has dramatically increased the number of connected devices. This has created a huge attack surface that requires the deployment of effective and practical countermeasures to protect network infrastructures from the harm that cyber-attacks can cause. Hence, there is an absolute need to differentiate boundaries in personal information and cloud and fog computing globally and the adoption of specific information security policies and regulations. The goal of the security policy and framework for cloud and fog computing is to protect the end-users and their information, reduce task-based operations, aid in compliance, and create standards for expected user actions, all of which are based on the use of established rules for cloud computing. Moreover, intrusion detection systems are widely adopted solutions to monitor and analyze network traffic and detect anomalies that can help identify ongoing adversarial activities, trigger alerts, and automatically block traffic from hostile sources. This survey paper analyzes factors, including the application of technologies and techniques, which can enable the deployment of security policy on fog and cloud computing successfully. The paper focuses on a Software-as-a-Service (SaaS) and intrusion detection, which provides an effective and resilient system structure for users and organizations. Our survey aims to provide a framework for a cloud and fog computing security policy, while addressing the required security tools, policies, and services, particularly for cloud and fog environments for organizational adoption. While developing the essential linkage between requirements, legal aspects, analyzing techniques and systems to reduce intrusion detection, we recommend the strategies for cloud and fog computing security policies. The paper develops structured guidelines for ways in which organizations can adopt and audit the security of their systems as security is an essential component of their systems and presents an agile current state-of-the-art review of intrusion detection systems and their principles. Functionalities and techniques for developing these defense mechanisms are considered, along with concrete products utilized in operational systems. Finally, we discuss evaluation criteria and open-ended challenges in this area.

show abstract

Effective hierarchical clustering based on structural similarities in nearest neighbor graphs

Cited by 40 publications

References 34 publications

Pattern Recognition for Steam Flooding Field Applications Based on Hierarchical Clustering and Principal Component Analysis

Pattern Recognition for Steam Flooding Field Applications Based on Hierarchical Clustering and Principal Component Analysis

Cell Production System Design: A Literature Review

A Survey on Intrusion Detection Systems for Fog and Cloud Computing

Contact Info

Product

Resources

About