Dustin Arendt

Grace Guo, Maria Glenski, ZhuanYi Shaw et al.

Natural experiments are observational studies where the assignment of treatment conditions to different populations occurs by chance "in the wild". Researchers from fields such as economics, healthcare, and the social sciences leverage natural experiments to conduct hypothesis testing and causal effect estimation for treatment and outcome variables that would otherwise be costly, infeasible, or unethical. In this paper, we introduce VAINE (Visualization and AI for Natural Experiments), a visual analytics tool for identifying and understanding natural experiments from observational data. We then demonstrate how VAINE can be used to validate causal relationships, estimate average treatment effects, and identify statistical phenomena such as Simpson's paradox through two usage scenarios.

Maria Glenski, Ellyn Ayton, Robin Cosbey et al.

Evaluating model robustness is critical when developing trustworthy models not only to gain deeper understanding of model behavior, strengths, and weaknesses, but also to develop future models that are generalizable and robust across expected environments a model may encounter in deployment. In this paper we present a framework for measuring model robustness for an important but difficult text classification task - deceptive news detection. We evaluate model robustness to out-of-domain data, modality-specific features, and languages other than English. Our investigation focuses on three type of models: LSTM models trained on multiple datasets(Cross-Domain), several fusion LSTM models trained with images and text and evaluated with three state-of-the-art embeddings, BERT ELMo, and GloVe (Cross-Modality), and character-level CNN models trained on multiple languages (Cross-Language). Our analyses reveal a significant drop in performance when testing neural models on out-of-domain data and non-English languages that may be mitigated using diverse training data. We find that with additional image content as input, ELMo embeddings yield significantly fewer errors compared to BERT orGLoVe. Most importantly, this work not only carefully analyzes deception model robustness but also provides a framework of these analyses that can be applied to new models or extended datasets in the future.

Maria Glenski, Ellyn Ayton, Robin Cosbey et al.

With the increasing use of machine-learning driven algorithmic judgements, it is critical to develop models that are robust to evolving or manipulated inputs. We propose an extensive analysis of model robustness against linguistic variation in the setting of deceptive news detection, an important task in the context of misinformation spread online. We consider two prediction tasks and compare three state-of-the-art embeddings to highlight consistent trends in model performance, high confidence misclassifications, and high impact failures. By measuring the effectiveness of adversarial defense strategies and evaluating model susceptibility to adversarial attacks using character- and word-perturbed text, we find that character or mixed ensemble models are the most effective defenses and that character perturbation-based attack tactics are more successful.

Brittany Davis, Maria Glenski, William Sealy et al.

Research into the explanation of machine learning models, i.e., explainable AI (XAI), has seen a commensurate exponential growth alongside deep artificial neural networks throughout the past decade. For historical reasons, explanation and trust have been intertwined. However, the focus on trust is too narrow, and has led the research community astray from tried and true empirical methods that produced more defensible scientific knowledge about people and explanations. To address this, we contribute a practical path forward for researchers in the XAI field. We recommend researchers focus on the utility of machine learning explanations instead of trust. We outline five broad use cases where explanations are useful and, for each, we describe pseudo-experiments that rely on objective empirical measurements and falsifiable hypotheses. We believe that this experimental rigor is necessary to contribute to scientific knowledge in the field of XAI.

Winston Wu, Dustin Arendt, Svitlana Volkova

We evaluate machine comprehension models' robustness to noise and adversarial attacks by performing novel perturbations at the character, word, and sentence level. We experiment with different amounts of perturbations to examine model confidence and misclassification rate, and contrast model performance in adversarial training with different embedding types on two benchmark datasets. We demonstrate improving model performance with ensembling. Finally, we analyze factors that effect model behavior under adversarial training and develop a model to predict model errors during adversarial attacks.

Dustin Arendt, Zhuanyi Huang, Prasha Shrestha et al.

Evaluation beyond aggregate performance metrics, e.g. F1-score, is crucial to both establish an appropriate level of trust in machine learning models and identify future model improvements. In this paper we demonstrate CrossCheck, an interactive visualization tool for rapid crossmodel comparison and reproducible error analysis. We describe the tool and discuss design and implementation details. We then present three use cases (named entity recognition, reading comprehension, and clickbait detection) that show the benefits of using the tool for model evaluation. CrossCheck allows data scientists to make informed decisions to choose between multiple models, identify when the models are correct and for which examples, investigate whether the models are making the same mistakes as humans, evaluate models' generalizability and highlight models' limitations, strengths and weaknesses. Furthermore, CrossCheck is implemented as a Jupyter widget, which allows rapid and convenient integration into data scientists' model development workflows.

Alireza Karduni, Isaac Cho, Ryan Wesslen et al.

We present Verifi2, a visual analytic system to support the investigation of misinformation on social media. On the one hand, social media platforms empower individuals and organizations by democratizing the sharing of information. On the other hand, even well-informed and experienced social media users are vulnerable to misinformation. To address the issue, various models and studies have emerged from multiple disciplines to detect and understand the effects of misinformation. However, there is still a lack of intuitive and accessible tools that help social media users distinguish misinformation from verified news. In this paper, we present Verifi2, a visual analytic system that uses state-of-the-art computational methods to highlight salient features from text, social network, and images. By exploring news on a source level through multiple coordinated views in Verifi2, users can interact with the complex dimensions that characterize misinformation and contrast how real and suspicious news outlets differ on these dimensions. To evaluate Verifi2, we conduct interviews with experts in digital media, journalism, education, psychology, and computing who study misinformation. Our interviews show promising potential for Verifi2 to serve as an educational tool on misinformation. Furthermore, our interview results highlight the complexity of the problem of combating misinformation and call for more work from the visualization community.

Maria Glenski, Ellyn Ayton, Dustin Arendt et al.

This paper presents the results and conclusions of our participation in the Clickbait Challenge 2017 on automatic clickbait detection in social media. We first describe linguistically-infused neural network models and identify informative representations to predict the level of clickbaiting present in Twitter posts. Our models allow to answer the question not only whether a post is a clickbait or not, but to what extent it is a clickbait post e.g., not at all, slightly, considerably, or heavily clickbaity using a score ranging from 0 to 1. We evaluate the predictive power of models trained on varied text and image representations extracted from tweets. Our best performing model that relies on the tweet text and linguistic markers of biased language extracted from the tweet and the corresponding page yields mean squared error (MSE) of 0.04, mean absolute error (MAE) of 0.16 and R2 of 0.43 on the held-out test data. For the binary classification setup (clickbait vs. non-clickbait), our model achieved F1 score of 0.69. We have not found that image representations combined with text yield significant performance improvement yet. Nevertheless, this work is the first to present preliminary analysis of objects extracted using Google Tensorflow object detection API from images in clickbait vs. non-clickbait Twitter posts. Finally, we outline several steps to improve model performance as a part of the future work.

Lyndsey Franklin, Kyungsik Han, Zhuanyi Huang et al.

We present the results of a user study with novice NMR analysts (N=19) involving a gamified simulation of the NMR analysis process. Participants solved randomly generated spectrum puzzles for up to three hours. We used eye tracking, event logging, and observations to record symptoms of cognitive depletion while participants worked. Analysis of results indicate that we can detect both signs of learning and signs of cognitive depletion in participants over the course of the three hours. Participants' break strategies did not predict or reflect game scores, but certain symptoms appear predictive of breaks.