Monitoring And Modeling Of Complex Processes Using Hierarchical Self-organizing Maps

Simula, Olli; Alhoniemi, Esa; Hollmén, Jaakko; Vesanto, Juha

doi:10.1109/iscas.1996.598480

Cited by 11 publications

(10 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It should be mentioned that there are other hierarchical self-organizing neural networks [21], [22], but the neural network structures and learning process of these networks are different from ours. In [21], there are self-organizing maps at two levels: the state map and the dynamics maps.…”

Section: B Hierarchical Self-organizing Neural Networkmentioning

confidence: 88%

“…In [21], there are self-organizing maps at two levels: the state map and the dynamics maps. The dynamics maps are associated with each node of the state map and used to predict the next state of the state map.…”

Section: B Hierarchical Self-organizing Neural Networkmentioning

confidence: 99%

“…Second, the best matching unit from the second map is chosen and its index is the output of the hierarchical network. The existing hierarchical self-organizing map methods have a mul tilayer structure and two training phases (we refer to [21], [22] for more details). In our hierarchical self-organizing structure, all neurons are partitioned into several internal nets; the internal nets and the external net are both in the same layer; and there is no two stage training phase, unlike existing hierarchical self-organizing maps.…”

Section: B Hierarchical Self-organizing Neural Networkmentioning

confidence: 99%

See 2 more Smart Citations

A Hierarchical Self-Organizing Approach for Learning the Patterns of Motion Trajectories

Xie

Tan

2004

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

Abstract-The understanding and description of object behav iors is a hot topic in computer vision. Trajectory analysis is one of the basic problems in behavior understanding, and the learning of trajectory patterns that can be used to detect anomalies and predict object trajectories is an interesting and important problem in tra jectory analysis. In this paper, we present a hierarchical self-orga nizing neural network model and its application to the learning of trajectory distribution patterns for event recognition. The distribu tion patterns of trajectories are learnt using a hierarchical self-or ganizing neural network. Using the learned patterns, we consider anomaly detection as well as object behavior prediction. Compared with the existing neural network structures that are used to learn patterns of trajectories, our network structure has smaller scale and faster learning speed, and is thus more effective. Experimental results using two different sets of data demonstrate the accuracy and speed of our hierarchical self-organizing neural network in learning the distribution patterns of object trajectories.Index Terms-Hierarchical self-organizing neural network, tra jectory analysis and learning, anomaly detection, behavior predic tion.

show abstract

Section: B Hierarchical Self-organizing Neural Networkmentioning

confidence: 88%

Section: B Hierarchical Self-organizing Neural Networkmentioning

confidence: 99%

Section: B Hierarchical Self-organizing Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

A Hierarchical Self-Organizing Approach for Learning the Patterns of Motion Trajectories

Xie

Tan

2004

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

show abstract

“…Visualization of trajectories on U-matrices have been used for monitoring chemical processes (Ultsch 1993), and have been applied to complex processes, such as the dynamic behavior of a computer systems with regard to utilization rates and traffic volume (Simula et al 1996), to industrial processes, such as a continuous pulp digester, steel production and pulp and paper mills , and to different subjects with distinct sleep apnea diseases (Guimaraes et al 2001).…”

Section: Temporal Sequence Processing Without Modifying the Basic Sommentioning

confidence: 99%

“…First, the weights of the lower-level SOM are used as input without any further processing, either taking into account the information of the previous known classes (Kemke, Wichert 1993) or without considering any information on the classes (Walter, Ritter 1996). Second, a transformation of the network results is possible, for instance: 1) calculating the distances between the units (Carpinteiro 2000); 2) concatenating subsequent vectors into a single vector, thus representing the history of state transitions (Simula et al 1996); or 3) taking into account the information about clusters formed at this level, and adjusting the weights towards the cluster center (Guimarães 2000). The third possibility lies in interposing other algorithms or methods, such as segment classifiers (Behme et al 1993).…”

Section: Modification Of the Network Topologymentioning

confidence: 99%

Classification and Clustering in Business Cycle Analysis.

Heilemann¹,

Weihs²

2007

Schriften Des Rheinisch-Westfälischen Instituts Für Wirtschaftsforschung

View full text Add to dashboard Cite

PrefaceThe analysis of cyclical macroeconomic phenomena is an important field of econometric research. In the recent past, research interests have de-emphasized quantitative forecasting exercises and have addressed the qualitative diagnosis of the relative stance of the economy regarding "upswing", "recession", or "boom" periods, i. e. the classification of the state of the economy into a limited number of discrete states. In this context the principal challenge is to reduce the multifaceted and sometimes abundant quantitative information about the business cycle to such qualitative statements in an efficient way. For more than six years this task was the focus of the project "Multivariate determination and analysis of business cycles" within the SFB 475 "Reduction of complexity in multivariate data structures", funded by the German Research Foundation (DFG).The necessity for complexity reduction is, of course, not unique to business cycle analysis but is studied in many fields and in a number of ways. This broad interest in the reduction of problem dimensionality and in the appropriate combination of data and of theory caused the RWI Essen and the Statistical Department of the University of Dortmund in January 2002 to hold a workshop at the RWI Essen where the findings of this and similar projects were presented and discussed. The present publication collects revised versions of the papers presented at this workshop. Although the workshop took place some five years ago, these papers mark an importment juncture in the development of business cycle research.The RWI Essen thanks the editors of the volume and the authors for their contributions and the other participants for lively discussions. Particular thanks have to go to Joachim Schmidt who supervised the production of the volume and to Anette Hermanowski and Claudia Lohkamp whose support of the production of the manuscripts and of the organisation of the workshop was indispensable.Essen,

show abstract

Fault Detection and Diagnosis in Industrial Systems

Chiang

Russell

Braatz

2001

Advanced Textbooks in Control and Signal Processing

1,136

811

View full text Add to dashboard Cite

An alternative group of methods has emerged which do not require the use of an explicit model. This is the key basic construct for the data-driven paradigm. Model-free and non-parametric methods for fault detection, process optimisation and control design are currently at a particularly exciting stage of development.This new advanced textbook by Chiang, Russell and Braatz primarily tackles the data-driven routes to Fault Detection and Diagnosis. It is an outgrowth of a prior Advances in Industrial Control monograph; Russell, Chiang and Braatz. Data-driven Techniques for Fault Detection and Diagnosis in ChemicalProcesses, 2000, ISBN 1-85233-258-1. The new textbook expands the material of the monograph and gives a fuller presentation of some of the alternative modelbased methods, the analytical methods, and of the knowledge-based techniques. vi Series Editors' ForewordThis allows the reader to compare and contrast the different approaches to the problem of fault detection and diagnosis. Thus the text is suitable for advanced courses for process, chemical and control engineers.

show abstract

Monitoring And Modeling Of Complex Processes Using Hierarchical Self-organizing Maps

Cited by 11 publications

References 6 publications

A Hierarchical Self-Organizing Approach for Learning the Patterns of Motion Trajectories

A Hierarchical Self-Organizing Approach for Learning the Patterns of Motion Trajectories

Classification and Clustering in Business Cycle Analysis.

Fault Detection and Diagnosis in Industrial Systems

Contact Info

Product

Resources

About