Moh Chegini Ph.D.

Methods from supervised machine learning allow the classification of new data automatically and are tremendously helpful for data analysis. The quality of supervised maching learning depends not only on the type of algorithm used, but also on the quality of the labelled dataset used to train the classifier. Labelling instances in a training dataset is often done manually relying on selections and annotations by expert analysts, and is often a tedious and time-consuming process. Active learning… Mehr anzeigen

Methods from supervised machine learning allow the classification of new data automatically and are tremendously helpful for data analysis. The quality of supervised maching learning depends not only on the type of algorithm used, but also on the quality of the labelled dataset used to train the classifier. Labelling instances in a training dataset is often done manually relying on selections and annotations by expert analysts, and is often a tedious and time-consuming process. Active learning algorithms can automatically determine a subset of data instances for which labels would provide useful input to the learning process. Interactive visual labelling techniques are a promising alternative, providing effective visual overviews from which an analyst can simultaneously explore data records and select items to a label. By putting the analyst in the loop, higher accuracy can be achieved in the resulting classifier. While initial results of interactive visual labelling techniques are promising in the sense that user labelling can improve supervised learning, many aspects of these techniques are still largely unexplored. This paper presents a study conducted using the mVis tool to compare three interactive visualisations, similarity map, scatterplot matrix (SPLOM), and parallel coordinates, with each other and with active learning for the purpose of labelling a multivariate dataset. The results show that all three interactive visual labelling techniques surpass active learning algorithms in terms of classifier accuracy, and that users subjectively prefer the similarity map over SPLOM and parallel coordinates for labelling. Users also employ different labelling strategies depending on the visualisation used. Weniger anzeigen

Parallel coordinates is a well-known technique for visual analysis of high-dimensional data. Although it is effective for interactive discovery of patterns in subsets of dimensions and data records, it also has scalability issues for large datasets. In particular, the amount of visual information potentially being shown in a parallel coordinates plot grows combinatorially with the number of dimensions. Choosing the right ordering of axes is crucial, and poor design can lead to visual noise and… Mehr anzeigen

Parallel coordinates is a well-known technique for visual analysis of high-dimensional data. Although it is effective for interactive discovery of patterns in subsets of dimensions and data records, it also has scalability issues for large datasets. In particular, the amount of visual information potentially being shown in a parallel coordinates plot grows combinatorially with the number of dimensions. Choosing the right ordering of axes is crucial, and poor design can lead to visual noise and a cluttered plot. In this case, the user may overlook a significant pattern, or leave some dimensions unexplored. In this work, we demonstrate how eye-tracking can help an analyst efficiently and effectively reorder the axes in a parallel coordinates plot. Implicit input from an inexpensive eye-tracker assists the system in finding unexplored dimensions. Using this information, the system guides the user either visually or automatically to find further appropriate orderings of the axes. Weniger anzeigen

Many machine learning algorithms require a labelled training dataset. The task of labelling a multivariate dataset can be tedious, but can be supported by systems combining interactive visualisation and machine learning techniques into a single interface. mVis is such a system, providing a unified ecosystem to explore multivariate datasets and execute machine learning algorithms to build labelled datasets.
This paper describes a pre-study evaluation of the mVis system, comprising case… Mehr anzeigen

Many machine learning algorithms require a labelled training dataset. The task of labelling a multivariate dataset can be tedious, but can be supported by systems combining interactive visualisation and machine learning techniques into a single interface. mVis is such a system, providing a unified ecosystem to explore multivariate datasets and execute machine learning algorithms to build labelled datasets.
This paper describes a pre-study evaluation of the mVis system, comprising case studies in two different domains: collaborative intelligence and daily activities. In each case study, a volunteer researcher was asked to use mVis to explore, analyse, and label their own dataset in their own environment, while thinking out loud. The case studies provided valuable leanings in terms of the usability of the system, understanding how different analysts work, and identifying important missing features. Weniger anzeigen

Supervised machine learning techniques require labelled multivariate training datasets. Many approaches address the issue of unlabelled datasets by tightly coupling machine learning algorithms with interactive visualisations. Using appropriate techniques, analysts can play an active role in a highly interactive and iterative machine learning process to label the dataset and create meaningful partitions. While this principle has been implemented either for unsupervised, semi-supervised, or… Mehr anzeigen

Supervised machine learning techniques require labelled multivariate training datasets. Many approaches address the issue of unlabelled datasets by tightly coupling machine learning algorithms with interactive visualisations. Using appropriate techniques, analysts can play an active role in a highly interactive and iterative machine learning process to label the dataset and create meaningful partitions. While this principle has been implemented either for unsupervised, semi-supervised, or supervised machine learning tasks, the combination of all three methodologies remains challenging.

In this paper, a visual analytics approach is presented, combining a variety of machine learning capabilities with four linked visualisation views, all integrated within the mVis (multivariate Visualiser) system. The available palette of techniques allows an analyst to perform exploratory data analysis on a multivariate dataset and divide it into meaningful labelled partitions, from which a classifier can be built. In the workflow, the analyst can label interesting patterns or outliers in a semi-supervised process supported by active learning. Once a dataset has been interactively labelled, the analyst can continue the workflow with supervised machine learning to assess to what degree the subsequent classifier has effectively learned the concepts expressed in the labelled training dataset. Using a novel technique called automatic dimension selection, interactions the analyst had with dimensions of the multivariate dataset are used to steer the machine learning algorithms. Weniger anzeigen

Large displays are capable of visualising a large amount of data on multiple views including scatterplots and parallel coordinates and are often present in meeting rooms. They can be used to interact with a dataset and foster discussion among team members. Although some of these large screens have multi-touch capabilities, in many cases it is cumbersome to have to stand close to the display in order to interact with it. One of the solutions is to use a small handheld display to interact with… Mehr anzeigen

Large displays are capable of visualising a large amount of data on multiple views including scatterplots and parallel coordinates and are often present in meeting rooms. They can be used to interact with a dataset and foster discussion among team members. Although some of these large screens have multi-touch capabilities, in many cases it is cumbersome to have to stand close to the display in order to interact with it. One of the solutions is to use a small handheld display to interact with the large display. This paper discusses how traditional interactions such as selection, brushing, and linking can be performed using a secondary handheld device. As a proof of concept, a system including scatterplots and parallel coordinates views is implemented. The interactions are straightforward and are useful for any interactive visual analysis application on a large display with wireless connectivity. Weniger anzeigen

An interesting question one may ask about multivariate graphs is if and how structural characteristics are reflected in multivariate attribute similarities and vice versa. In this work, we propose an integrated visualization of structure S and attribute similarity A in a single overview matrix. The idea is to show S and A in the lower and upper triangular matrix, respectively. Dynamic local rearrangement of matrix cells allows the user to create a side-by-side view of S and A for a detailed… Mehr anzeigen

An interesting question one may ask about multivariate graphs is if and how structural characteristics are reflected in multivariate attribute similarities and vice versa. In this work, we propose an integrated visualization of structure S and attribute similarity A in a single overview matrix. The idea is to show S and A in the lower and upper triangular matrix, respectively. Dynamic local rearrangement of matrix cells allows the user to create a side-by-side view of S and A for a detailed inspection of both aspects for a selected node. The approach is applied to a multivariate graph of soccer players. Weniger anzeigen

Analysts often use visualisation techniques like a scatterplot matrix (SPLOM) to explore multivariate datasets. The scatterplots of a SPLOM can help to identify and compare two‐dimensional global patterns. However, local patterns which might only exist within subsets of records are typically much harder to identify and may go unnoticed among larger sets of plots in a SPLOM. This paper explores the notion of local patterns and presents a novel approach to visually select, search for, and compare… Mehr anzeigen

Analysts often use visualisation techniques like a scatterplot matrix (SPLOM) to explore multivariate datasets. The scatterplots of a SPLOM can help to identify and compare two‐dimensional global patterns. However, local patterns which might only exist within subsets of records are typically much harder to identify and may go unnoticed among larger sets of plots in a SPLOM. This paper explores the notion of local patterns and presents a novel approach to visually select, search for, and compare local patterns in a multivariate dataset. Model‐based and shape‐based pattern descriptors are used to automatically compare local regions in scatterplots to assist in the discovery of similar local patterns. Mechanisms are provided to assess the level of similarity between local patterns and to rank similar patterns effectively. Moreover, a relevance feedback module is used to suggest potentially relevant local patterns to the user. The approach has been implemented in an interactive tool and demonstrated with two real‐world datasets and use cases. It supports the discovery of potentially useful information such as clusters, functional dependencies between variables, and statistical relationships in subsets of data records and dimensions. Weniger anzeigen

The latest generation of large vertically-mounted multitouch displays brings new opportunities for solving visual analytics tasks. Due to their size, it is possible to visualise and collaboratively interact with high dimensional datasets and multiple views (e.g., scatterplots, scatterplot matrices and parallel coordinates). However, using only multi-touch for input can be overly restrictive. Other modalities need to be considered to utilise the power of these screens fully. By adding natural… Mehr anzeigen

The latest generation of large vertically-mounted multitouch displays brings new opportunities for solving visual analytics tasks. Due to their size, it is possible to visualise and collaboratively interact with high dimensional datasets and multiple views (e.g., scatterplots, scatterplot matrices and parallel coordinates). However, using only multi-touch for input can be overly restrictive. Other modalities need to be considered to utilise the power of these screens fully. By adding natural language interaction, the user can directly interact with the visual analytics application from a distance. Incorporating eye-tracking can help narrow down what the user is looking at or is interested in. In this paper, some of the challenges of using multi-touch as input for the analysis of scatterplot spaces on large vertically-mounted multitouch displays are discussed and addressed by proposing the incorporation of other interaction modalities. Weniger anzeigen

Large vertically-mounted high-resolution multitouch displays are becoming increasingly available for interactive data visualisation. Such devices are well-suited to small-team collaborative visual analysis. In particular, the visual analysis of large high-dimensional datasets can benefit from high-resolution displays capable of showing multiple coordinated views. This paper identifies some of the advantages of using large, high-resolution displays for visual analytics in general, and introduces… Mehr anzeigen

Large vertically-mounted high-resolution multitouch displays are becoming increasingly available for interactive data visualisation. Such devices are well-suited to small-team collaborative visual analysis. In particular, the visual analysis of large high-dimensional datasets can benefit from high-resolution displays capable of showing multiple coordinated views. This paper identifies some of the advantages of using large, high-resolution displays for visual analytics in general, and introduces a set of interactions to explore high-dimensional datasets on large vertically-mounted high-resolution multi-touch displays using scatterplots. A set of touch interactions for collaborative visual analysis of scatterplots have been implemented and are presented. Finally, three perception-based levels of detail techniques are introduced for such displays as a concept for further implementation. Weniger anzeigen

Nowadays, handheld devices such as smartphones provide users with multi-touch input screens. Displaying interactive and touch-enabled 3D environments in such handheld devices has become popular in different applications like games or virtual reality. Technologies such as Web3D and WebGL have made the creation and display of 3D environments in mobile devices easier than ever. However, object manipulation techniques are not as well developed. For example, moving an object within the 3D… Mehr anzeigen

Nowadays, handheld devices such as smartphones provide users with multi-touch input screens. Displaying interactive and touch-enabled 3D environments in such handheld devices has become popular in different applications like games or virtual reality. Technologies such as Web3D and WebGL have made the creation and display of 3D environments in mobile devices easier than ever. However, object manipulation techniques are not as well developed. For example, moving an object within the 3D environment or other similar object-specific manipulations are neither intuitive nor easy to perform. Current manipulation techniques like Gizmo that are successful in systems that use mouse and keyboard are not designed for and do not work well for multi-touch handheld devices. In this paper, we present a novel technique to perform object manipulation in 6DOF in multi-touch screens. Our performance evaluations show that our technique compared to existing techniques such as Gizmo improves task completion time by 63% while increasing task precision by 52%. Weniger anzeigen