SliceTeller: A Data Slice-Driven Approach for Machine Learning Model Validation

An Empirical Evaluation of the GPT-4 Multimodal Language Model on Visualization Literacy Tasks

Large Language Models (LLMs) like GPT-4 which support multimodal input (i.e., prompts containing images in addition to text) have immense potential to advance visualization research. However, many questions exist about the visual capabilities of such models, including how well they can read and interpret visually represented data. In our work, we address this question by evaluating the GPT-4 multimodal LLM using a suite of task sets meant to assess the model's visualization literacy. The task sets are based on existing work in the visualization community addressing both automated chart question answering and human visualization literacy across multiple settings. Our assessment finds that GPT-4 can perform tasks such as recognizing trends and extreme values, and also demonstrates some understanding of visualization design best-practices. By contrast, GPT-4 struggles with simple value retrieval when not provided with the original dataset, lacks the ability to reliably distinguish between colors in charts, and occasionally suffers from hallucination and inconsistency. We conclude by reflecting on the model's strengths and weaknesses as well as the potential utility of models like GPT-4 for future visualization research. We also release all code, stimuli, and results for the task sets at the following link: https://doi.org/10.17605/OSF.IO/F39J6.

ProactiV: Studying Deep Learning Model Behavior Under Input Transformations

Deep learning (DL) models have shown performance benefits across many applications, from classification to image-to-image translation. However, low interpretability often leads to unexpected model behavior once deployed in the real world. Usually, this unexpected behavior is because the training data domain does not reflect the deployment data domain. Identifying a model's breaking points under input conditions and domain shifts, i.e., input transformations, is essential to improve models. Although visual analytics (VA) has shown promise in studying the behavior of model outputs under continually varying inputs, existing methods mainly focus on per-class or instance-level analysis. We aim to generalize beyond classification where classes do not exist and provide a global view of model behavior under co-occurring input transformations. We present a DL model-agnostic VA method (ProactiV) to help model developers proactively study output behavior under input transformations to identify and verify breaking points. ProactiV relies on a proposed input optimization method to determine the changes to a given transformed input to achieve the desired output. The data from this optimization process allows the study of global and local model behavior under input transformations at scale. Additionally, the optimization method provides insights into the input characteristics that result in desired outputs and helps recognize model biases. We highlight how ProactiV effectively supports studying model behavior with example classification and image-to-image translation tasks.

Visual Analytics for Efficient Image Exploration and User-Guided Image Captioning

Recent advancements in pre-trained language-image models have ushered in a new era of visual comprehension. Leveraging the power of these models, this article tackles two issues within the realm of visual analytics: (1) the efficient exploration of large-scale image datasets and identification of data biases within them; (2) the evaluation of image captions and steering of their generation process. On the one hand, by visually examining the captions generated from language-image models for an image dataset, we gain deeper insights into the visual contents, unearthing data biases that may be entrenched within the dataset. On the other hand, by depicting the association between visual features and textual captions, we expose the weaknesses of pre-trained language-image models in their captioning capability and propose an interactive interface to steer caption generation. The two parts have been coalesced into a coordinated visual analytics system, fostering the mutual enrichment of visual and textual contents. We validate the effectiveness of the system with domain practitioners through concrete case studies with large-scale image datasets.

My Model is Unfair, Do People Even Care? Visual Design Affects Trust and Perceived Bias in Machine Learning

Machine learning technology has become ubiquitous, but, unfortunately, often exhibits bias. As a consequence, disparate stakeholders need to interact with and make informed decisions about using machine learning models in everyday systems. Visualization technology can support stakeholders in understanding and evaluating trade-offs between, for example, accuracy and fairness of models. This paper aims to empirically answer "Can visualization design choices affect a stakeholder's perception of model bias, trust in a model, and willingness to adopt a model?" Through a series of controlled, crowd-sourced experiments with more than 1,500 participants, we identify a set of strategies people follow in deciding which models to trust. Our results show that men and women prioritize fairness and performance differently and that visual design choices significantly affect that prioritization. For example, women trust fairer models more often than men do, participants value fairness more when it is explained using text than as a bar chart, and being explicitly told a model is biased has a bigger impact than showing past biased performance. We test the generalizability of our results by comparing the effect of multiple textual and visual design choices and offer potential explanations of the cognitive mechanisms behind the difference in fairness perception and trust. Our research guides design considerations to support future work developing visualization systems for machine learning.

Applying machine learning techniques to predict the risk of lung metastases from rectal cancer: a real-world retrospective study

BACKGROUND

Metastasis in the lungs is common in patients with rectal cancer, and it can have severe consequences on their survival and quality of life. Therefore, it is essential to identify patients who may be at risk of developing lung metastasis from rectal cancer.

METHODS

In this study, we utilized eight machine-learning methods to create a model for predicting the risk of lung metastasis in patients with rectal cancer. Our cohort consisted of 27,180 rectal cancer patients selected from the Surveillance, Epidemiology and End Results (SEER) database between 2010 and 2017 for model development. Additionally, we validated our models using 1118 rectal cancer patients from a Chinese hospital to evaluate model performance and generalizability. We assessed our models' performance using various metrics, including the area under the curve (AUC), the area under the precision-recall curve (AUPR), the Matthews Correlation Coefficient (MCC), decision curve analysis (DCA), and calibration curves. Finally, we applied the best model to develop a web-based calculator for predicting the risk of lung metastasis in patients with rectal cancer.

RESULT

Our study employed tenfold cross-validation to assess the performance of eight machine-learning models for predicting the risk of lung metastasis in patients with rectal cancer. The AUC values ranged from 0.73 to 0.96 in the training set, with the extreme gradient boosting (XGB) model achieving the highest AUC value of 0.96. Moreover, the XGB model obtained the best AUPR and MCC in the training set, reaching 0.98 and 0.88, respectively. We found that the XGB model demonstrated the best predictive power, achieving an AUC of 0.87, an AUPR of 0.60, an accuracy of 0.92, and a sensitivity of 0.93 in the internal test set. Furthermore, the XGB model was evaluated in the external test set and achieved an AUC of 0.91, an AUPR of 0.63, an accuracy of 0.93, a sensitivity of 0.92, and a specificity of 0.93. The XGB model obtained the highest MCC in the internal test set and external validation set, with 0.61 and 0.68, respectively. Based on the DCA and calibration curve analysis, the XGB model had better clinical decision-making ability and predictive power than the other seven models. Lastly, we developed an online web calculator using the XGB model to assist doctors in making informed decisions and to facilitate the model's wider adoption (https://share.streamlit.io/woshiwz/rectal_cancer/main/lung.py).

CONCLUSION

In this study, we developed an XGB model based on clinicopathological information to predict the risk of lung metastasis in patients with rectal cancer, which may help physicians make clinical decisions.