Maintaining roadway infrastructure is essential for ensuring a safe, efficient, a nd sustainable transportation system. However, manual data collection for detecting road damage is time-consuming, labor-intensive, and poses safety risks. Recent advancements in artificial intelligence, particularly deep learning, offer a promising solution for automating this process using road images. This paper presents a comprehensive workflow for road damage detection using deep learning models, focusing on optimizations for inference speed while preserving detection accuracy. Specifically, to accommodate hardware limitations, large images are cropped, and lightweight models are utilized. Additionally, an external pothole dataset is incorporated to enhance the detection of this underrepresented damage class. The proposed approach employs multiple model architectures, including a custom YOLOv7 model with Coordinate Attention layers and a Tiny YOLOv7 model, which are trained and combined to maximize detection performance. The models are further reparameterized to optimize inference efficiency. Experimental results demonstrate that the ensemble of the custom YOLOv7 model with three Coordinate Attention layers and the default Tiny YOLOv7 model achieves an F1 score of 0.7027 with an inference speed of 0.0547 seconds per image. The complete pipeline, including data preprocessing, model training, and inference scripts, is publicly available on the project’s GitHub repository, enabling reproducibility and facilitating further research.

Fingerprint verification is widely used to verify individuals based on their unique fingerprint patterns. However, existing fingerprint identification systems encounter challenges while dealing with poor-quality images, smudged or marked fingerprints, and variations in finger positioning. These issues are common for latent fingerprints. This work proposes to address these issues through a novel machine-learning model employing advanced image enhancement techniques. The model aims to enhance fingerprint image quality and minimize the impact of damage using machine learning, ultimately reducing the error rate in identification systems. Specifically, this work proposes to incorporate deep learning image enhancement techniques into the low-quality and latent fingerprints before passing them through another deep learning model to perform verification tasks. This work also provides insights as different experiments are made while applying different approaches to different real-life datasets.

Soil organic carbon is an essential element of environmental quality assessment because carbon dioxide is the main greenhouse gas causing global warming and climate change. Trees and plants acquire a lot of carbon in their lifetimes, and soil helps keep this portion of carbon from releasing to the environment. Thus, understanding the current carbon storage in the soil and the relationship with tree richness is important. The soil organic carbon datasets and corresponding tree information are available. However, the current visualizations for these datasets are either static visualizations that report initial findings to the public or visualizations that are too complicated to be useful for a broad audience. Therefore, this work proposes an interactive visualization solution for analyzing soil organic carbon values distribution and their relationship with tree heights and diameters at breast heights, called RaCAViz. The design of this interactive visualization is based on the analytical evaluation of visual and interaction idioms to tackle typical analysis tasks on this type of dataset. However, this paper also provides a specific use case to illustrate its usefulness in bringing different perspectives on soil organic carbon datasets to a broad audience.

Maintaining the roadway infrastructure is one of the essential factors in enabling a safe, economic, and sustainable transportation system. Manual roadway damage data collection is laborious and unsafe for humans to perform. This area is poised to benefit from the rapid advance and diffusion of artificial intelligence technologies. Specifically, deep learning advancements enable the detection of road damages automatically from the collected road images. This work proposes to collect and label road damage data using Google Street View and use YOLOv7 (You Only Look Once version 7) together with coordinate attention and related accuracy fine-tuning techniques such as label smoothing and ensemble method to train deep learning models for automatic road damage detection and classification. The proposed approaches are applied to the Crowdsensing-based Road Damage Detection Challenge (CRDDC2022), IEEE BigData 2022. The results show that the data collection from Google Street View is efficient, and the proposed deep learning approach results in F1 scores of 81.7% on the road damage data collected from the United States using Google Street View and 74.1% on all test images of this dataset. With these results, we received rank 2 (silver prize) as a data contributor and rank 3 (bronze prize) as the predictive model in this competition among 54 leaders (private companies and academic institutions) in this area.

Soil properties are vital to profiling and utilizing soil resources. Conventional approaches to measurements of soil properties often involve costly, environmental-unfriendly, and time-consuming laboratory procedures. Conversely, machine learning (ML) and deep learning (DL) are gaining traction in giving rapid, non-destructive, and cost-saving alternatives to predictions of soil properties. These ML/DL models are convenient and fast because they utilize spectral data, such as visible and near-infrared (Vis-NIR) spectra, that can be easily collected using proximal sensors for their training and prediction purposes. However, existing ML/DL approaches to this problem pose several limitations, such as having small sample sizes, needing to divide the sample data into local areas to increase accuracy, and having relatively low accuracy. Therefore, this work experiments various ML/DL methods that leverage Vis-NIR spectra collected from a rather large number of soil samples distributed all over the world to predict $p{H_{{H_2}O}}$ and pHKCl. We then propose a DL method, called RDNet, that outperforms the other existing approaches. We also utilize visualizations to verify if the proposed model learns legitimate information from the training data.

This paper details the methodologies and decisions making processes used while developing the attacking and defending models for the Graph Adversarial Attacks and Defense applied to a large citation graph. To handle the large graphs, our attack strategy is twofold: 1) randomly attack the structure first, 2) keep the structure unchanged, then continue the attack on the features using the gradient-based method. On the other hand, the defender is based on 1) filtering and normalizing the feature data, 2) applying the Graph Convolutional Network model, and 3) selecting the models with the highest accuracy and robustness based on our own attacking data. We applied these strategies in KDD Cup 2020 on Graph Adversarial Attacks and Defense dataset. The attacker can drop the accuracy of a surrogate 2-layer Graph Convolutional Network model from 60% to 30% on the test set. Our defending model has 68% accuracy on the validated data and has 89% of the target labels remained the same while adding fake nodes, generated by our attacking method, to the graph.

The road is vital for many aspects of life, and road maintenance is crucial for human safety. One of the critical tasks to allow timely repair of road damages is to quickly and efficiently detect and classify them. This work details the strategies and experiments evaluated for these tasks. Specifically, we evaluate Detectron2's implementation of Faster R-CNN using different base models and configurations. We also experiment with these approaches using the Global Road Damage Detection Challenge 2020, A Track in the IEEE Big Data 2020 Big Data Cup Challenge dataset. The results show that the X101-FPN base model for Faster R-CNN with Detectron2's default configurations is efficient and general enough to be transferable to different countries in this challenge. This approach results in F1 scores of 51.0% and 51.4% for the test1 and test2 sets of the challenge, respectively. Though the visualizations show good prediction results, the F1 scores are low. Therefore, we also evaluate the prediction results against the existing annotations and discover some discrepancies. Thus, we also suggest strategies to improve the labeling process for this dataset.

Detecting outliers is one of the fundamental tasks in visual analytics and valuable in many application domains, such as suspicious network cyberattack recognition. This paper introduces an approach to analyzing and visualizing high-dimensional time series, focusing on identifying multivariate observations that are significantly different from the others. We also propose a prototype, called MTSAD, to guide users when interactively exploring abnormalities in large time series. The prototype contains two views: the main window provides an overview of identified outliers overtime, the detail window investigates and explores the ranked temporal data entries based on their outlying contributions to the overall plots. The visual interface supports a full range of interactions, such as lensing, brushing and linking, ranking, and filtering. To validate the benefits and usefulness of our approach, we demonstrate MTSAD on real-world datasets of different numbers of attributes.

This paper describes a machine learning approach to the solar flare prediction competition, a track in IEEE Big Data 2019 Big Data Cup. The competition task is to predict whether or not there is a solar flare event basing on a given time series of solar magnetic field parameters. Our method involves exploring and constructing data-driven machine learning models for the classification task of two imbalanced class labels from time series. Specifically, the investigated models include boosting, logistic regression, multilayer perceptron neural network, and long short-term memory neural network. These models have been successfully deployed and combined in an ensemble framework with two tiers in our final proposed solution for this competition. Our proposed approach ranked at the second place in the competition (the first on the private board and the eleventh on the public board).

Analyzing public user posts and shared information on social media can assist us in measuring various population characteristics, patterns, movements, and as well as the public health conditions. In recent years, researchers have been analyzing social media (such as Facebook or Twitter feeds) to detect and predict various emerging events and market trends. Fewer attentions have been paid to the epidemic of the diseases. In this paper, we present a social media analytics tool, called HealthTvizer, for exploring health awareness using Twitter data through interactive and interconnected multiple views. We use topic modeling to pick the relevant and meaningful terms from more than 57 million tweets. We detect the disease name and related contents which are shared by the users of different geographical locations (mostly in the United States). We believe that the collected geolocations from the users' tweets can reveal the patterns of diseases for a given term which allows a researcher to detect, analyze, and explore information about the diseases and hence take necessary steps to improve public health awareness. We validate the effectiveness of HealthTvizer through an informal user study. The feedback from this study also motivates us on interesting future extensions of the tool.

The explosion of social media has paved a way for big data in which entrepreneurs use this data to find out potential customers, market demands, individual behavior, thereby to improve existing products, to create new products according to users' need, or to analyze and evaluate financial risks. The challenges of the heterogeneity and fragmentation of data make it difficult for analysts to fully exploit the benefit of deluge information. Available statistical software lacks customization and address unknown research questions. This paper proposes FinanViz, a visual analytics tool for analyzing financial news on social media. The principal aim of FinanViz is to observe the dynamic behavior of terms/words over time along with their proximity to other terms/words. The tool provides an intuitive, interactive exploration of the financial topics and what events are emerging in which we would argue that it will give hints for financial marketers in the decision making process.

Visualizing and analyzing geospatial and temporal observations are common tasks for many application domains. In this paper, we introduce UFO Tracker, a visual analytic tool for analyzing unidentified flying object sightings from the National UFO Reporting Center. The goal here is to give the user a higher level view of where different types of sightings occur, to investigate whether sightings are increasing or decreasing over time, to discover the connections between different events which might happen at different geographic areas, and to quickly identify typical incidents at a given period of time without reading the whole sightings through topic modelling. Multiple visualization and data mining techniques are combined to make sense the increasingly large UFO reports which get updated hourly. The usefulness of the application is evaluated through a case study where anon-expert in ufology can find some typical interesting sightings. Our application can also be able to detect some misleading events such as missile launch or fireworks on a specific day through keywords and topic extraction. One limitation of our application is the data which is not up-todate when new sightings are posted since the application pulled and processed data locally. Our initial application targets UFO sighting reports. However, we believe our approach has wider applications in other research domains, such as analyzing text corpus obtained from social media.

Cyber attacks cause great damage to our national security, ranging from individual internet user to biggest governmental/industrial organizations, such as Equifax (Data Breach 145.5 Million Accounts, reported in July 2017) or Uber (Data Breach 57 Million Records, reported in November 2017). The cyber assault has significantly increased in breadth and depth. This paper introduces CVExplorer, a novel interactive system for visualizing cybersecurity threats reported in the National Vulnerability Database. The proposed system aims to work as a reporting and alerting tool that can help enhance the security against cyber attacks can potentially reduce network vulnerabilities. The CVExplorer system containing multiple linked views allows users to visualize the relationships of various dimensions in the large number of vulnerability reports, such as types and levels of vulnerability, vendors, and products. The CVExplorer provides an intuitive interface and supports a range of interactive features, such as filtering and ordering by vulnerability severity ratings, allowing users to narrow down topics of interest quickly. To demonstrate the effectiveness of the proposed system, we demonstrate the CVExplorer on two case studies of Common Vulnerabilities and Exposures retrieved from the National Vulnerability Database.

The "Internet of Things" is changing the way companies operate and consumers behave. Therefore, it is essential to capture trends in "Internet of Things". This paper proposes IoTViz, a visual analytics tool for analyzing "Internet of Things" news on social media. The principal aim of IoTViz is to observe the dynamic behavior of topics along with their proximity to other dimensions such as user comments and ratings in multiple coordinated views. IoTViz provides an interactive exploration of the IoT topics and supports of a range of interactive features, such as linking and filtering, allowing users to narrow down events of interest quickly. It is interesting to filter and visualize IoT news regarding the individual organization, e.g., user opinions/ratings regarding a company or its products.