More Like this

1. Using Close Reading as a Method for Evaluating Visualizations

   https://doi.org/10.1109/BELIV51497.2020.00011  

   Bares, Annie; Zeller, Stephanie; Jackson, Cullen D.; Keefe, Daniel F.; Samsel, Francesca (October 2020, 2020 IEEE Workshop on Evaluation and Beyond - Methodological Approaches to Visualization (BELIV))

   null (Ed.)

   Visualization research and practice that incorporates the arts make claims to being more effective in connecting with users on a human level. However, these claims are difficult to measure quantitatively. In this paper, we present a follow-on study to use close reading, a humanities method from literary studies, to evaluate visualizations created using artistic processes [Bares 2020]. Close reading is a method in literary studies that we've previously explored as a method for evaluating visualizations. To use close reading as an evaluation method, we guide participants through a series of steps designed to prompt them to interpret the visualization's formal, informational, and contextual features. Here we elaborate on our motivations for using close reading as a method to evaluate visualizations, and enumerate the procedures we used in the study to evaluate a 2D visualization, including modifications made because of the COVID-19 pandemic. Key findings of this study include that close reading is an effective formative method to elicit information related to interpretation and critique; user subject position; and suspicion or skepticism. Information gained through close reading is valuable in the visualization design and iteration processes, both related to designing features and other formal elements more effectively, as well as in considering larger questions of context and framing. 

   more » « less
2. Stacked LSTM based deep recurrent neural network with kalman smoothing for blood glucose prediction

   https://doi.org/10.1186/s12911-021-01462-5  

   Rabby, Md Fazle; Tu, Yazhou; Hossen, Md Imran; Lee, Insup; Maida, Anthony S.; Hei, Xiali (December 2021, BMC Medical Informatics and Decision Making)

   Abstract Background Blood glucose (BG) management is crucial for type-1 diabetes patients resulting in the necessity of reliable artificial pancreas or insulin infusion systems. In recent years, deep learning techniques have been utilized for a more accurate BG level prediction system. However, continuous glucose monitoring (CGM) readings are susceptible to sensor errors. As a result, inaccurate CGM readings would affect BG prediction and make it unreliable, even if the most optimal machine learning model is used. Methods In this work, we propose a novel approach to predicting blood glucose level with a stacked Long short-term memory (LSTM) based deep recurrent neural network (RNN) model considering sensor fault. We use the Kalman smoothing technique for the correction of the inaccurate CGM readings due to sensor error. Results For the OhioT1DM (2018) dataset, containing eight weeks’ data from six different patients, we achieve an average RMSE of 6.45 and 17.24 mg/dl for 30 min and 60 min of prediction horizon (PH), respectively. Conclusions To the best of our knowledge, this is the leading average prediction accuracy for the ohioT1DM dataset. Different physiological information, e.g., Kalman smoothed CGM data, carbohydrates from the meal, bolus insulin, and cumulative step counts in a fixed time interval, are crafted to represent meaningful features used as input to the model. The goal of our approach is to lower the difference between the predicted CGM values and the fingerstick blood glucose readings—the ground truth. Our results indicate that the proposed approach is feasible for more reliable BG forecasting that might improve the performance of the artificial pancreas and insulin infusion system for T1D diabetes management. 

   more » « less
3. Reading Fascists Reading Shakespeare: Literary Populism in White Power Fiction

   https://doi.org/10.1215/08992363-11714141  

   Ahmann, Chloe; Proctor, Devin (January 2025, Public Culture)

   This article explores the peculiar reading habits that permeate White power fiction, with a focus on the repeated invocation of William Shakespeare and other Western literary giants by a group of writers here theorized as “literary populists.” Literary populists not only reject “elite” interpretations and reclaim the canon “for the masses.” They also marshal canonistic texts as masses, while enrolling them in autocratic schemes. In the process, they promote a strongman's reading praxis that limits the interpretive terrain in ways that echo archetypical fascist movements. Following a broad review of fascist literary practice, the article examines this phenomenon as it structures five works of speculative fiction by American ecofascist Harold Covington, showing how venerated books work as loudspeakers in his texts, where White self-;realization takes the form of literary mastery. It then moves beyond the texts to a reading of his online library as a mediating force that combines classic works of literature, philosophy, and political theory with the verboten work of radicals like Covington, in an attempt to authorize the latter. Through this analysis, the article shows how extremist visions of the world-;to-;come gain ground as part of a storied intellectual tradition, whose texts are plotted as proof of an impending revolution and whose status as banned knowledge is essential to its potency. At a time when fascism often looks like banning books, this work suggests that reading is fundamental to fomenting fascist visions of a world worth fighting for, and for seizing the minds of organic intellectuals who might just bring those futures into being. 

   more » « less
4. VisuaLint: Sketchy In Situ Annotations of Chart Construction Errors

   https://doi.org/10.1111/cgf.13975  

   Hopkins, Aspen K; Correll, Michael; Satyanarayan, Arvind. (January 2020, Computer graphics forum)

   Chart construction errors, such as truncated axes or inexpressive visual encodings, can hinder reading a visualization, or worse, imply misleading facts about the underlying data. These errors can be caught by critical readings of visualizations, but readers must have a high level of data and design literacy and must be paying close attention. To address this issue, we introduce VisuaLint: a technique for surfacing chart construction errors in situ. Inspired by the ubiquitous red wavy underline that indicates spelling mistakes, visualization elements that contain errors (e.g., axes and legends) are sketchily rendered and accompanied by a concise annotation. VisuaLint is unobtrusive — it does not interfere with reading a visualization — and its di- rect display establishes a close mapping between erroneous elements and the expression of error. We demonstrate five examples of VisualLint and present the results of a crowdsourced evaluation (N = 62) of its efficacy. These results contribute an empiri- cal baseline proficiency for recognizing chart construction errors, and indicate near-universal difficulty in error identification. We find that people more reliably identify chart construction errors after being shown examples of VisuaLint, and prefer more verbose explanations for unfamiliar or less obvious flaws. 

   more » « less
5. Physics-Guided Long Short-Term Memory Networks for Emission Prediction in Laser Powder Bed Fusion

   https://doi.org/10.1115/1.4063270  

   Lei, Rong; Guo, Y B; Guo, Weihong “Grace” (January 2024, Journal of Manufacturing Science and Engineering)

   Abstract Powder bed fusion (PBF) is an additive manufacturing process in which laser heat liquefies blown powder particles on top of a powder bed, and cooling solidifies the melted powder particles. During this process, the laser beam heat interacts with the powder causing thermal emission and affecting the melt pool. This paper aims to predict heat emission in PBF by harnessing the strengths of recurrent neural networks. Long short-term memory (LSTM) networks are developed to learn from sequential data (emission readings), while the learning is guided by process physics including laser power, laser speed, layer number, and scanning patterns. To reduce the computational efforts on model training, the LSTM models are integrated with a new approach for down-sampling the pyrometry raw data and extracting useful statistical features from raw data. The structure and hyperparameters of the LSTM model reflect several iterations of tuning based on the training on the pyrometer readings data. Results reveal useful knowledge on how raw pyrometer data should be processed to work the best with LSTM, how physics features are informative in predicting overheating, and the effectiveness of physics-guided LSTM in emission prediction. 

   more » « less