Visual Methods for Analyzing Probabilistic Classification Data

Alsallakh, Bilal; Hanbury, Allan; Hauser, Helwig; Miksch, Silvia; Rauber, Andreas

doi:10.1109/tvcg.2014.2346660

Cited by 101 publications

(96 citation statements)

References 36 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Users can explore the model results, and interactions support model refinement. Alsallakh et al propose the confusion wheel [AHH*14] and visualize true positive, false positive, false negative, and false positive classification results.…”

Section: Pva Pipelinementioning

confidence: 99%

“…Predictive visual analytics is currently being used to make complex models generated by black‐box AI techniques simpler to interpret thus allowing for more accuracy. PVA systems can provide tools that aid in supervised learning to generate interpretable classification models [HKBE12, AHH*14] or to allow the injection of domain knowledge during the construction of decision trees [VDEvW11]. However, PVA must find ways to scale to the increasingly large and complex models that are in use for emerging applications.…”

Section: Future Directions In Pvamentioning

confidence: 99%

See 1 more Smart Citation

The State‐of‐the‐Art in Predictive Visual Analytics

García

Hansen

et al. 2017

Computer Graphics Forum

View full text Add to dashboard Cite

Predictive analytics embraces an extensive range of techniques including statistical modeling, machine learning, and data mining and is applied in business intelligence, public health, disaster management and response, and many other fields. To date, visualization has been broadly used to support tasks in the predictive analytics pipeline. Primary uses have been in data cleaning, exploratory analysis, and diagnostics. For example, scatterplots and bar charts are used to illustrate class distributions and responses. More recently, extensive visual analytics systems for feature selection, incremental learning, and various prediction tasks have been proposed to support the growing use of complex models, agent‐specific optimization, and comprehensive model comparison and result exploration. Such work is being driven by advances in interactive machine learning and the desire of end‐users to understand and engage with the modeling process. In this state‐of‐the‐art report, we catalogue recent advances in the visualization community for supporting predictive analytics. First, we define the scope of predictive analytics discussed in this article and describe how visual analytics can support predictive analytics tasks in a predictive visual analytics (PVA) pipeline. We then survey the literature and categorize the research with respect to the proposed PVA pipeline. Systems and techniques are evaluated in terms of their supported interactions, and interactions specific to predictive analytics are discussed. We end this report with a discussion of challenges and opportunities for future research in predictive visual analytics.

show abstract

Section: Pva Pipelinementioning

confidence: 99%

Section: Future Directions In Pvamentioning

confidence: 99%

The State‐of‐the‐Art in Predictive Visual Analytics

García

Hansen

et al. 2017

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…Alsallakh et al [2] propose to use multiple box plots to visualize the feature distribution of training samples and the ability of a feature to separate data into different classes. FeatureInsight [6] is a system that combines human and machine intelligence to examine classification errors to identify and build potentially useful features.…”

Section: Related Workmentioning

confidence: 99%

A Visual Technique to Analyze Flow of Information in a Machine Learning System

Chaudhuri¹

2018

View full text Add to dashboard Cite

Machine learning (ML) algorithms and machine learning based software systems implicitly or explicitly involve complex flow of information between various entities such as training data, feature space, validation set and results. Understanding the statistical distribution of such information and how they flow from one entity to another influence the operation and correctness of such systems, especially in large-scale applications that perform classification or prediction in real time. In this paper, we propose a visual approach to understand and analyze flow of information during model training and serving phases. We build the visualizations using a technique called Sankey Diagram -conventionally used to understand data flow among sets -to address various use cases of in a machine learning system. We demonstrate how the proposed technique, tweaked and twisted to suit a classification problem, can play a critical role in better understanding of the training data, the features, and the classifier performance. We also discuss how this technique enables diagnostic analysis of model predictions and comparative analysis of predictions from multiple classifiers. The proposed concept is illustrated with the example of categorization of millions of products in the e-commerce domain -a multi-class hierarchical classification problem.

show abstract

“…In other words, we are interested in general techniques that work for any classifier rather than specific classifier types. One of the main challenges is to find a way of projecting instance points and decision boundaries meaningfully in a low-dimensional representation [Alsallakh et al 2014]. For example, Frank and Hall [2003] propose a general method by discretizing the attributes in the feature space, which allows one to obtain a separation of feature space in disjoint rectangular regions (Figure 3(a)).…”

Section: Classifier Visualizationmentioning

confidence: 99%

Adding Human Learning in Brain--Computer Interfaces (BCIs)

Kosmyna

Tarpin-Bernard

Rivet

2015

ACM Trans. Comput.-Hum. Interact.

View full text Add to dashboard Cite

In this article, we introduce CLBCI (Co-Learning for Brain-Computer Interfaces), a BCI architecture based on co-learning in which users can give explicit feedback to the system rather than just receiving feedback. CLBCI is based on minimum distance classification with Independent Component Analysis (ICA) and allows for shorter training times compared to classical BCIs, as well as faster learning in users and a good performance progression. We further propose a new scheme for real-time two-dimensional visualization of classification outcomes using Wachspress coordinate interpolation. It allows us to represent classification outcomes for n classes in simple regular polygons. Our objective is to devise a BCI system that constitutes a practical interaction modality that can be deployed rapidly and used on a regular basis. We apply our system to an event-based control task in the form of a simple shooter game in which we evaluate the learning effect induced by our architecture compared to a classical approach. We also evaluate how much user feedback and our visualization method contribute to the performance of the system.

show abstract

Visual Methods for Analyzing Probabilistic Classification Data

Cited by 101 publications

References 36 publications

The State‐of‐the‐Art in Predictive Visual Analytics

The State‐of‐the‐Art in Predictive Visual Analytics

A Visual Technique to Analyze Flow of Information in a Machine Learning System

Adding Human Learning in Brain--Computer Interfaces (BCIs)

Contact Info

Product

Resources

About