Visual Analytics in Ship Performance and Navigation Information for Sensor Specific Fault Detection

2017 ◽  
Author(s):  
Lokukaluge P. Perera ◽  
Brage Mo
Keyword(s):  
Fault Detection ◽  
Visual Analytics ◽  
Domain Knowledge ◽  
Data Sets ◽  
Data Set ◽  
Navigation Data ◽  
Hidden Data ◽  
Ship Performance ◽  
Navigation Information ◽  
Erroneous Data

Ocean internet of things (IoT - onboard and onshore) collects big data sets of ship performance and navigation information under various data handling processes. That extract vessel performance and navigation information that are used for ship energy efficiency and emission control applications. However, the quality of ship performance and navigation data can play an important role in such applications, where sensor faults may introduce various erroneous data regions and that may degrade to the outcome. This study proposes visual analytics, where hidden data patterns, clusters, correlations and other useful information are visually from the respective data set extracted, to identify such erroneous data regions. The domain knowledge (i.e. ship performance and navigation conditions) has also been used to interpret such erroneous data regions and identify the respective sensors that relate to the same situations. Finally, a ship performance and navigation data set of a selected vessel is analyzed to identify erroneous data regions for three selected sensor fault situations (i.e. wind, log speed and draft sensors) under the proposed visual analytics. Hence, this approach can be categorized as a sensor specific fault detection methodology by considering the same results.

An Overview of Data Veracity Issues in Ship Performance and Navigation Monitoring

2018 ◽  
Author(s):  
Lokukaluge P. Perera ◽  
Brage Mo
Keyword(s):  
Data Analysis ◽  
Anomaly Detection ◽  
Domain Knowledge ◽  
Study Data ◽  
Data Sets ◽  
Data Handling ◽  
Navigation Data ◽  
Industrial Iot ◽  
Ship Performance

An overview of data veracity issues in ship performance and navigation monitoring in relation to data sets collected from a selected vessel is presented in this study. Data veracity relates to the quality of ship performance and navigation parameters obtained by onboard IoT (internet of things). Industrial IoT can introduce various anomalies into measured ship performance and navigation parameters and that can degrade the outcome of the respective data analysis. Therefore, the identification and isolation process of such data anomalies can play an important role in the outcome of ship performance and navigation monitoring. In general, these data anomalies can be divided into sensor and data acquisition (DAQ) faults and system abnormal events. A considerable amount of domain knowledge is required to detect and classify such data anomalies, therefore data anomaly detection layers are proposed in this study for the same purpose. These data anomaly detection layers are divided into several levels: preliminary and advanced levels. The outcome of a preliminary anomaly detection layer with respect to ship performance and navigation data sets of a selected vessel is presented with the respective data handling challenges as the main contribution of this study.

Visualization of Relative Wind Profiles in Relation to Actual Weather Conditions of Ship Routes

2017 ◽  
Author(s):  
Lokukaluge P. Perera ◽  
Brage Mo ◽  
Matthias P. Nowak
Keyword(s):  
Large Scale ◽  
Wind Profile ◽  
Weather Conditions ◽  
Data Sets ◽  
Navigation Data ◽  
External Data ◽  
Wind Profiles ◽  
Tools And Techniques ◽  
Ship Performance ◽  
Relative Wind

Ship performance and navigation data are collected by vessels that are equipped with various supervisory control and data acquisition systems (SCADA). Such information is collected as large-scale data sets, therefore various analysis tools and techniques are required to extract useful information from the same. The extracted information on ship performance and navigation conditions can be used to implement energy efficiency and emission control applications (i.e. weather routing type applications) on these vessels. Hence, this study proposes to develop data visualizing methods in order to extract ship performance and navigation information from the respective data sets in relation to weather conditions. The relative wind (i.e. apparent wind) profile (i.e. wind speed and direction) collected by onboard sensors and absolute weather conditions, which are extracted from external data sources by using position and time information a selected vessel (i.e. from the recorded ship routes), are considered. Hence, the relative wind profile of the vessel is compared with actual weather conditions to visualize ship performance and navigation parameters relationships, as the main contribution. It is believed that such relationships can be used to develop appropriate mathematical models to predict ship performance and navigation conditions under various weather conditions.

scholarly journals Marrying Medical Domain Knowledge With Deep Learning on Electronic Health Records: A Deep Visual Analytics Approach (Preprint)

2020 ◽  
Author(s):  
Rui Li ◽  
Changchang Yin ◽  
Samuel Yang ◽  
Buyue Qian ◽  
Ping Zhang
Keyword(s):  
Heart Failure ◽  
Deep Learning ◽  
Risk Prediction ◽  
Visual Analytics ◽  
Domain Knowledge ◽  
Knowledge Graph ◽  
Data Set ◽  
Medical Domain ◽  
Clinical Risks

BACKGROUND Deep learning models have attracted significant interest from health care researchers during the last few decades. There have been many studies that apply deep learning to medical applications and achieve promising results. However, there are three limitations to the existing models: (1) most clinicians are unable to interpret the results from the existing models, (2) existing models cannot incorporate complicated medical domain knowledge (eg, a disease causes another disease), and (3) most existing models lack visual exploration and interaction. Both the electronic health record (EHR) data set and the deep model results are complex and abstract, which impedes clinicians from exploring and communicating with the model directly. OBJECTIVE The objective of this study is to develop an interpretable and accurate risk prediction model as well as an interactive clinical prediction system to support EHR data exploration, knowledge graph demonstration, and model interpretation. METHODS A domain-knowledge–guided recurrent neural network (DG-RNN) model is proposed to predict clinical risks. The model takes medical event sequences as input and incorporates medical domain knowledge by attending to a subgraph of the whole medical knowledge graph. A global pooling operation and a fully connected layer are used to output the clinical outcomes. The middle results and the parameters of the fully connected layer are helpful in identifying which medical events cause clinical risks. DG-Viz is also designed to support EHR data exploration, knowledge graph demonstration, and model interpretation. RESULTS We conducted both risk prediction experiments and a case study on a real-world data set. A total of 554 patients with heart failure and 1662 control patients without heart failure were selected from the data set. The experimental results show that the proposed DG-RNN outperforms the state-of-the-art approaches by approximately 1.5%. The case study demonstrates how our medical physician collaborator can effectively explore the data and interpret the prediction results using DG-Viz. CONCLUSIONS In this study, we present DG-Viz, an interactive clinical prediction system, which brings together the power of deep learning (ie, a DG-RNN–based model) and visual analytics to predict clinical risks and visually interpret the EHR prediction results. Experimental results and a case study on heart failure risk prediction tasks demonstrate the effectiveness and usefulness of the DG-Viz system. This study will pave the way for interactive, interpretable, and accurate clinical risk predictions.

scholarly journals DataMeadow: A Visual Canvas for Analysis of Large-Scale Multivariate Data

2008 ◽  
Vol 7 (1) ◽  
pp. 18-33 ◽  
Author(s):  
Niklas Elmqvist ◽  
John Stasko ◽  
Philippas Tsigas
Keyword(s):  
Visual Analytics ◽  
Large Scale ◽  
Multidimensional Data ◽  
Data Sets ◽  
Data Set ◽  
Data Dependencies ◽  
Expert Review ◽  
History Of ◽  
Multidimensional Data Sets ◽  
High Degree

Supporting visual analytics of multiple large-scale multidimensional data sets requires a high degree of interactivity and user control beyond the conventional challenges of visualizing such data sets. We present the DataMeadow, a visual canvas providing rich interaction for constructing visual queries using graphical set representations called DataRoses. A DataRose is essentially a starplot of selected columns in a data set displayed as multivariate visualizations with dynamic query sliders integrated into each axis. The purpose of the DataMeadow is to allow users to create advanced visual queries by iteratively selecting and filtering into the multidimensional data. Furthermore, the canvas provides a clear history of the analysis that can be annotated to facilitate dissemination of analytical results to stakeholders. A powerful direct manipulation interface allows for selection, filtering, and creation of sets, subsets, and data dependencies. We have evaluated our system using a qualitative expert review involving two visualization researchers. Results from this review are favorable for the new method.

Data Compression of Ship Performance and Navigation Information Under Deep Learning

2016 ◽  
Author(s):  
Lokukaluge P. Perera ◽  
Brage Mo
Keyword(s):  
Energy Efficiency ◽  
Deep Learning ◽  
Communication Networks ◽  
System Architecture ◽  
Complex Data ◽  
Data Handling ◽  
Data Set ◽  
Ship Performance ◽  
Compressed Data ◽  
Navigation Information

Various emission control regulations enforce vessels to collect performance and navigation data and evaluate ship energy efficiency by implementing onboard sensors and data acquisition (DAQ) systems. These DAQ systems are designed to collect, store and communicate large amounts of performance and navigation information through complex data handling processes. It is suggested that this information should eventually be transferred to shore based data analysis centers for further processing and storage. However, the associated data transfer costs introduce additional challenges in this process and enforce to investigate cost effective data handling approaches in shipping. That mainly relates to the amount of data that are transferring through various communication networks (i.e. satellites & wireless networks) between vessels and shore based data centers. Hence, this study proposes to use a deep learning approach (i.e. autoencoder system architecture) to compress ship performance and navigation information, which can be transferred through the respective communication networks as a reduced data set. The compressed data set can be expanded in the respective data center requiring further analysis. Therefore, a data set of ship performance and navigation information is analyzed (i.e. compression and expansion) through an autoencoder system architecture in this study. The compressed data set represents a subset of ship performance and navigation information can also be used to evaluate energy efficiency type applications in shipping. Furthermore, the respective input and output data sets of the autoencoder are also compared as statistical distributions to evaluate the network performance.

Digitalization of Seagoing Vessels Under High Dimensional Data Driven Models

2017 ◽  
Author(s):  
Lokukaluge P. Perera ◽  
Brage Mo
Keyword(s):  
Domain Knowledge ◽  
Large Scale ◽  
Dimensional Space ◽  
Data Driven ◽  
High Dimensional ◽  
Data Handling ◽  
Shipping Industry ◽  
Data Set ◽  
Data Clusters ◽  
Ship Performance

Modern ships are supported by internet of things (IoT) to collect ship performance and navigation information. That should be utilized towards digitalization of the shipping industry. However, such information collection systems are always associated with large-scale data sets, so called Big Data, where various industrial challenges are encountered during the respective data handling processes. This study proposes a data handling framework with data driven models (i.e. digital models) to cope with the shipping industrial challenges as the main contribution, where conventional mathematical models may fail. The proposed data driven models are developed in a high dimensional space, where the respective ship performance and navigation parameters of a selected vessel are separated as several data clusters. Hence, this study identifies the distribution of the respective data clusters and the structure of each data cluster in relation to ship performance and navigation conditions. An appropriate structure into the data set of ship performance and navigation parameters is assigned by this method as the main contribution. However, the domain knowledge (i.e. vessel operational and navigation conditions) is also included in this situation to derive a meaningful data structure.

scholarly journals The Active Segmentation Platform for Microscopic Image Classification and Segmentation

2021 ◽  
Vol 11 (12) ◽  
pp. 1645
Author(s):  
Sumit K. Vohra ◽  
Dimiter Prodanov
Keyword(s):  
Machine Learning ◽  
Image Segmentation ◽  
Image Classification ◽  
Domain Knowledge ◽  
Feature Space ◽  
Ground Truth ◽  
Classification Problem ◽  
Data Sets ◽  
Learning Approaches ◽  
Data Set

Image segmentation still represents an active area of research since no universal solution can be identified. Traditional image segmentation algorithms are problem-specific and limited in scope. On the other hand, machine learning offers an alternative paradigm where predefined features are combined into different classifiers, providing pixel-level classification and segmentation. However, machine learning only can not address the question as to which features are appropriate for a certain classification problem. The article presents an automated image segmentation and classification platform, called Active Segmentation, which is based on ImageJ. The platform integrates expert domain knowledge, providing partial ground truth, with geometrical feature extraction based on multi-scale signal processing combined with machine learning. The approach in image segmentation is exemplified on the ISBI 2012 image segmentation challenge data set. As a second application we demonstrate whole image classification functionality based on the same principles. The approach is exemplified using the HeLa and HEp-2 data sets. Obtained results indicate that feature space enrichment properly balanced with feature selection functionality can achieve performance comparable to deep learning approaches. In summary, differential geometry can substantially improve the outcome of machine learning since it can enrich the underlying feature space with new geometrical invariant objects.

scholarly journals Marrying Medical Domain Knowledge With Deep Learning on Electronic Health Records: A Deep Visual Analytics Approach

10.2196/20645 ◽  
2020 ◽  
Vol 22 (9) ◽  
pp. e20645
Author(s):  
Rui Li ◽  
Changchang Yin ◽  
Samuel Yang ◽  
Buyue Qian ◽  
Ping Zhang
Keyword(s):  
Heart Failure ◽  
Deep Learning ◽  
Risk Prediction ◽  
Visual Analytics ◽  
Domain Knowledge ◽  
Knowledge Graph ◽  
Data Set ◽  
Medical Domain ◽  
Clinical Risks

Background Deep learning models have attracted significant interest from health care researchers during the last few decades. There have been many studies that apply deep learning to medical applications and achieve promising results. However, there are three limitations to the existing models: (1) most clinicians are unable to interpret the results from the existing models, (2) existing models cannot incorporate complicated medical domain knowledge (eg, a disease causes another disease), and (3) most existing models lack visual exploration and interaction. Both the electronic health record (EHR) data set and the deep model results are complex and abstract, which impedes clinicians from exploring and communicating with the model directly. Objective The objective of this study is to develop an interpretable and accurate risk prediction model as well as an interactive clinical prediction system to support EHR data exploration, knowledge graph demonstration, and model interpretation. Methods A domain-knowledge–guided recurrent neural network (DG-RNN) model is proposed to predict clinical risks. The model takes medical event sequences as input and incorporates medical domain knowledge by attending to a subgraph of the whole medical knowledge graph. A global pooling operation and a fully connected layer are used to output the clinical outcomes. The middle results and the parameters of the fully connected layer are helpful in identifying which medical events cause clinical risks. DG-Viz is also designed to support EHR data exploration, knowledge graph demonstration, and model interpretation. Results We conducted both risk prediction experiments and a case study on a real-world data set. A total of 554 patients with heart failure and 1662 control patients without heart failure were selected from the data set. The experimental results show that the proposed DG-RNN outperforms the state-of-the-art approaches by approximately 1.5%. The case study demonstrates how our medical physician collaborator can effectively explore the data and interpret the prediction results using DG-Viz. Conclusions In this study, we present DG-Viz, an interactive clinical prediction system, which brings together the power of deep learning (ie, a DG-RNN–based model) and visual analytics to predict clinical risks and visually interpret the EHR prediction results. Experimental results and a case study on heart failure risk prediction tasks demonstrate the effectiveness and usefulness of the DG-Viz system. This study will pave the way for interactive, interpretable, and accurate clinical risk predictions.

Disk Diagram: An Interactive Visualization Technique of Fuzzy Set Operations for the Analysis of Fuzzy Data

2010 ◽  
Vol 9 (3) ◽  
pp. 220-232 ◽  
Author(s):  
Yeseul Park ◽  
Jinah Park
Keyword(s):  
Fuzzy Sets ◽  
Fuzzy Set ◽  
Visual Analytics ◽  
Interactive Visualization ◽  
Data Sets ◽  
Fuzzy Data ◽  
Visualization Technique ◽  
Data Set ◽  
Set Operations ◽  
Data Elements

Fuzzy set refers to the data set which does not have separate, distinct clusters, and they contain data elements whose membership degrees are between 0.0 and 1.0. Many fuzzy sets exist in the real world, and one of the important issues is to make a decision from the fuzzy sets using visual analytics tools by extracting information in the data set intuitively. To analyze the element data in fuzzy sets, the visualization of fuzzy sets needs to show an overview of the data with membership degree and the relationship among the sets. In this article, we suggest an interactive visualization technique of fuzzy set operations, called Disk Diagram, which offers distribution of fuzzy data and two scenarios to allow users to interpret inter-dependency among fuzzy sets. A Disk Diagram enables to depict complexity of fuzzy sets by showing the degree of resemblance between the sets with the layout of star coordinates. This article describes the use of a Disk Diagram with two different data sets such as fuzzy disease set and terror related words set. Lastly, we report the results of heuristic evaluation to show that our technique supports visual perception, usability, and knowledge discovery process in the areas of visual representation and interaction.

The social wasp Vespula germanica (Fabricius) (Hymenoptera: Vespidae) population dynamics in England over 39 years.

2018 ◽  
Vol 154 (2) ◽  
pp. 149-155
Author(s):  
Michael Archer
Keyword(s):  
Population Dynamics ◽  
Population Dynamic ◽  
Ecological Factors ◽  
Social Wasp ◽  
Data Sets ◽  
Data Set ◽  
Vespula Germanica ◽  
The Social ◽  
Minimum Number ◽  
Suction Traps

1. Yearly records of worker Vespula germanica (Fabricius) taken in suction traps at Silwood Park (28 years) and at Rothamsted Research (39 years) are examined. 2. Using the autocorrelation function (ACF), a significant negative 1-year lag followed by a lesser non-significant positive 2-year lag was found in all, or parts of, each data set, indicating an underlying population dynamic of a 2-year cycle with a damped waveform. 3. The minimum number of years before the 2-year cycle with damped waveform was shown varied between 17 and 26, or was not found in some data sets. 4. Ecological factors delaying or preventing the occurrence of the 2-year cycle are considered.

Sign in / Sign up

Export Citation Format

Share Document