Machine learning algorithms play a pivotal role in a wide range of Artificial Intelligence (AI) applications. Explaining the results and behavior of a machine learning model, however, remains a challenge. In this paper, we present a new approach to the explanation of machine learning models using a large language model (LLM). In this work, we seek natural language descriptions of the behavioral patterns of a machine learning model by a combination of prompting and model sampling. A subspace sampling technique is developed to generate ML model outputs using partial features in a user defined space. A projective visualization method is employed to guide the sampling process, including user-directed interactive sampling and feature-based sampling, so that an optimal amount of information can be provided to the LLM to ensure accurate and concise natural language explanations. Two public datasets, a student performance dataset and a weather dataset, were used to test our approach under various conditions.

Advances in imaging technology and fluorescent calcium indicators have increased the popularity of the application of miniature microscopes (miniscope) in freely-moving rodents when they are performing sophisticated behaviors. The recently established analysis tools, such as Minian, allow for the efficient extraction of neuronal calcium transients from raw videos. However, there is still a gap in front of neuroscientists to draw any conclusions based on the neuronal calcium signals detected during behavioral tests. The poor signal-to-noise ratio and unreliable neuron registration in miniscope images bring significant challenges to the data analysis tasks. This paper describes a visual analytics approach to the exploration of potential neuronal clustering based on the calcium signals and the behavioral tasks. This new interactive solution can be applied for the analysis and exploration of the correlations between calcium imaging data and addiction behavioral data using a combination of visualization, user interaction, visual clustering, and event-related feature selection techniques. The associated interactive software tool is designed to help neuroscientists develop neurobiological hypotheses underlying both physiological and pathological behaviors, which could be a milestone in the field of neuroscience to understanding behavioral coding in the brain at the neuronal level.

Facial landmark placement is a key step in many biomedical and biometrics applications. This paper presents a computational method that efficiently performs automatic 3D facial landmark placement based on training images containing manually placed anthropological facial landmarks. After 3D face registration by an iterative closest point (ICP) technique, a visual analytics approach is taken to generate local geometric patterns for individual landmark points. These individualized local geometric patterns are derived interactively by a user's initial visual pattern detection. They are used to guide the refinement process for landmark points projected from a template face to achieve accurate landmark placement. Compared to traditional methods, this technique is simple, robust, and does not require a large number of training samples (e.g. in machine learning based methods) or complex 3D image analysis procedures. This technique and the associated software tool are being used in a 3D biometrics project that aims to identify links between human facial phenotypes and their genetic association.

Machine learning algorithms and traditional data mining process usually require a large volume of data to train the algorithm-specific models, with little or no user feedback during the model building process. Such a "big data" based automatic learning strategy is sometimes unrealistic for applications where data collection or processing is very expensive or difficult, such as in clinical trials. Furthermore, expert knowledge can be very valuable in the model building process in some fields such as biomedical sciences. In this paper, we propose a new visual analytics approach to interactive machine learning and visual data mining. In this approach, multi-dimensional data visualization techniques are employed to facilitate user interactions with the machine learning and mining process. This allows dynamic user feedback in different forms, such as data selection, data labeling, and data correction, to enhance the efficiency of model building. In particular, this approach can significantly reduce the amount of data required for training an accurate model, and therefore can be highly impactful for applications where large amount of data is hard to obtain. The proposed approach is tested on two application problems: the handwriting recognition (classification) problem and the human cognitive score prediction (regression) problem. Both experiments show that visualization supported interactive machine learning and data mining can achieve the same accuracy as an automatic process can with much smaller training data sets.

In spatiotemporal data visualization, integrating the time dimension with the spatial dimensions is a challenging problem. In this paper, we propose a new time representation method by mapping time onto a time curve in a color space. Since no spatial dimension is needed for the time axis, this approach is more effective in integrating time space and spatial dimensions. Several designs of the time curves in a 3D color space will be discussed. We apply this approach to the visualization application for a large taxi GPS dataset. The visualization is applied directly over an interactive map to depict the time patterns of a large set of driving paths on city roads. Spatial optimization techniques are also implemented to process large volumes of GPS data. This approach provides a new alternative for many spatiotemporal data visualization application.