Making Machine-Learning Applications for Time-Series Sensor Data Graphical and Interactive (original) (raw)
Related papers
Visualizing Sensor Data: Towards an Experiment and Validation Platform
Lecture Notes in Computer Science, 2012
In the last decade, technological improvement on sensors increasingly motivates the use of sensor data in numerous application domains such as environmental, health, transportation, etc. Progressively, with the advances on user terminals, there is a strong trend towards interactive sensor data visualization. As viewing raw sensor data stored in multiple databases does not specially fulfill user requirements, data visualization raises challenges about supporting users to easily use and handle sensor data. In this paper, we address this particular subject with an extensible visualization and interaction platform. Within this platform, we provide developers the facility to experiment and validate multiple visualizations for sensor data, specially based sensor data properties and users' requirements. We illustrate our platform with a medical study case focused on ECG data visualization.
A Framework for Visual Analytics of Spatio-Temporal Sensor Observations from Data Streams
ISPRS International Journal of Geo-Information
Sensor networks generate substantial amounts of frequently updated, highly dynamic data that are transmitted as packets in a data stream. The high frequency and continuous unbound nature of data streams leads to challenges when deriving knowledge from the underlying observations. This paper presents (1) a state of the art review into visual analytics of geospatial, spatio-temporal streaming data, and (2) proposes a framework based on the identified gaps from the review. The framework consists of (1) the data model that characterizes the sensor observation data, (2) the user model, which addresses the user queries and manages domain knowledge, (3) the design model, which handles the patterns that can be uncovered from the data and corresponding visualizations, and (4) the visualization model, which handles the rendering of the data. The conclusion from the visualization model is that streaming sensor observations require tools that can handle multivariate, multiscale, and time series...
Visual Analytics and Human Involvement in Machine Learning
ArXiv, 2020
The rapidly developing AI systems and applications still require human involvement in practically all parts of the analytics process. Human decisions are largely based on visualizations, providing data scientists details of data properties and the results of analytical procedures. Different visualizations are used in the different steps of the Machine Learning (ML) process. The decision which visualization to use depends on factors, such as the data domain, the data model and the step in the ML process. In this chapter, we describe the seven steps in the ML process and review different visualization techniques that are relevant for the different steps for different types of data, models and purposes.
Summarisation & Visualisation of Large Volumes of Time-Series Sensor Data
2010
With the increasing ubiquity of sensor data, presenting this data in a meaningful way to users is a challenge that must be addressed before we can easily deploy real-world sensor network interfaces in the home or workplace. In this paper, we will present one solution to the visualisation of large quantities of sensor data that is easy to understand and yet provides meaningful and intuitive information to a user, even when examining many weeks or months of historical data. We will illustrate this visulalisation technique with two real-world deployments of sensing the person and sensing the home.
Opening up the Black Box of Sensor Processing Algorithms through New Visualizations
Informatics
Vehicles and platforms with multiple sensors connect people in multiple roles with different responsibilities to scenes of interest. For many of these human-sensor systems there are a variety of algorithms that transform, select, and filter the sensor data prior to human intervention. Emergency response, precision agriculture, and intelligence, surveillance and reconnaissance (ISR) are examples of these human-computation-sensor systems. The authors examined a case of the latter to understand how people in various roles utilize the algorithms output to identify meaningful properties in data streams given uncertainty. The investigations revealed: (a) that increasingly complex interactions occur across agents in the human-computation-sensor system; and (b) analysts struggling to interpret the output of "black box" algorithms given uncertainty and change in the scenes of interest. The paper presents a new interactive visualization concept designed to "open up the black box" of sensor processing algorithms to support human analysts as they look for meaning in feeds from sensors.
MaVis: Machine Learning Aided Multi-Model Framework for Time Series Visual Analytics
The ultimate goal of any visual analytic task is to make sense of the data and gain insights. Unfortunately, the continuously growing scale of the data nowadays challenges the traditional data analytics in the " big-data " era. Particularly, the human cognitive capabilities are constant whereas the data scale is not. Furthermore, most existing work focus on how to extract interesting information and present that to the user while not emphasizing on how to provide options to the analysts if the extracted information is not interesting. In this paper, we propose a visual analytic tool called MaVis that integrates multiple machine learning models with a plug-and-play style to describe the input data. It allows the analysts to choose the way they prefer to summarize the data. The MaVis framework provides multiple linked analytic spaces for interpretation at different levels. The low level data space handles data binning strategy while the high level model space handles model summarizations (i.e. clusters or trends). MaVis also supports model analytics that visualize the summarized patterns and compare and contrast them. This framework is shown to provide several novel methods of investigating co-movement patterns of timeseries dataset which is a common interest of medical sciences, finance, business and engineering alike. Lastly we demonstrate the usefulness of our framework via case study and user study using a stock price dataset.
Machine learning has become one of the foremost techniques used for extracting knowledge from large amounts of data. The programming expertise required to implement machine learning algorithms has led to the rise of software products that simplify the process. Many of these systems however, have sacrificed simplicity as they evolved and included more features. In this study, a machine learning software with a simple graphical user interface was developed with a special focus on enhancing usability. The system made use of basic graphical interface elements such as buttons and textboxes. Comparison of the system with other similar open-source tools revealed that the developed system showed an improvement in usability over the other tools.
Machine learning meets visualization – Experiences and lessons learned
it - Information Technology
In this article, we discuss how Visualization (VIS) with Machine Learning (ML) could mutually benefit from each other. We do so through the lens of our own experience working at this intersection for the last decade. Particularly we focus on describing how VIS supports explaining ML models and aids ML-based Dimensionality Reduction techniques in solving tasks such as parameter space analysis. In the other direction, we discuss approaches showing how ML helps improve VIS, such as applying ML-based automation to improve visualization design. Based on the examples and our own perspective, we describe a number of open research challenges that we frequently encountered in our endeavors to combine ML and VIS.
Biologists studying animals in their natural environment are increasingly using sensors such as accelerometers in animal-attached 'smart' tags because it is widely acknowledged that this approach can enhance the understanding of ecological and behavioural processes. The potential of such tags is tempered by the difficulty of extracting animal behaviour from the sensors which is currently primarily dependent on the manual inspection of multiple time-series graphs. This is time-consuming and error-prone for the domain expert and is now the limiting factor for realising the value of tags in this area. We introduce TimeClassifier, a visual analytic system for the classification of time-series data for movement ecologists. We deploy our system with biologists and report two real-world case studies of its use.
Human-centered machine learning through interactive visualization
2016
The goal of visual analytics (VA) systems is to solve complex problems by integrating automated data analysis methods, such as machine learning (ML) algorithms, with interactive visualizations. We propose a conceptual framework that models human interactions with ML components in the VA process, and makes the crucial interplay between automated algorithms and interactive visualizations more concrete. The framework is illustrated through several examples. We derive three open research challenges at the intersection of ML and visualization research that will lead to more effective data analysis.