Forest fires present a significant challenge, which requires advanced detection systems to mitigate their impact on ecosystems and human settlements. The evolving landscape of remote sensing and machine learning offers new avenues for enhancing the speed and accuracy of fire detection methods. This paper introduces ForestFlameNet, an innovative deep learning model that incorporates separable convolutions and attention mechanisms tailored to efficiently analyze satellite and aerial imagery to detect early signs of wildfires. ForestFlameNet stands out by integrating a lightweight architecture that maintains high accuracy while being computationally efficient while offering comparable performance with state of the art models. With a model size of only $\mathbf{0. 7 5 4 M}$ parameters, ForestFlameNet ensures efficiency without compromising performance, making it highly suitable for deployment on resource-limited devices. The Local Interpretable Model-Agnostic Explanations (LIME) bolstered the model’s efficacy, illustrating its predictive decisions, and increasing transparency and trust in its outputs. t-Distributed Stochastic Neighbor Embedding (t-SNE) and Gradient-weighted Class Activation Mapping (Grad-CAM) also help to understand how the model recognizes features by showing how different parts of the input data affect the outcome of the detection.

Scientific datasets are increasingly cited in peer-reviewed journal publications, facilitating easy access to research utilizing those datasets. Datasets undergo a life cycle where older versions of datasets are replaced by newer versions often due to improvements in data resolution, algorithms, and other factors. Unlike peer reviewed documents registered with a single Digital Unique Identifier (DOI), datasets can be updated over time and the newer version of the datasets are registered with a new DOI which is not necessarily linked to the previous version of the dataset. It is challenging when publications citing a dataset need to be traced over the entire life cycle of that dataset. We provide an innovative approach to link the dataset versions and publications using a knowledge graph (KG). KG can help to trace the dataset cited in publications over the entire dataset life cycle and shed light into dataset usage in various applied research areas. We fine-tuned the pretrained NASA IMPACT INDUS Large Language Model (LLM) on a set of labeled publications abstracts. Our results showed that 87% of the publications were classified into one of twenty applied research areas, while the remaining 13% were classified into non-applied research areas. By linking datasets to applied research areas through the KG and employing Global Change Master Directory (GCMD), a well-established controlled vocabulary of scientific keywords describing Earth science datasets, we contribute to a transparent and advanced search and discovery mechanism for datasets across the Earth data ecosystem. The integrated KG and LLM approach is now incorporated and operational in dataset publication management at one of NASA’s Earth science data archival centers.

With the proliferation of computationally powerful edge devices, edge computing has been widely adopted for wide-ranging computational tasks. Among these, edge artificial intelligence (AI) has become a new trend, allowing local devices to work cooperatively and build deep learning models. Federated learning is one of the representative frameworks in distributed machine learning paradigms. However, there are several major concerns with existing federated learning paradigms. Existing distributed frameworks rely on a central server to coordinate the computing process, where such a central node may raise security concerns. Federated learning also relies on several assumptions/requirements, e.g., the independent and identically distributed (i.i.d.) data and model homogeneity. Since more and more edge devices are able to train lightweight models with local data, such models are normally heterogeneous. To tackle these challenges, in this article, we develop a blockchain-empowered federated learning framework that enables learning in a fully decentralized manner while taking the model heterogeneity and data heterogeneity into account. In particular, a federated learning framework with a heterogeneous calibration process, i.e., Model and Feature Calibration (FL-MFC), is developed to enable collaboration among heterogeneous models. We further design a two-level mining process using blockchain to enable the secure decentralized learning process. Experimental results show that our proposed system achieves effective learning performance under a fully heterogeneous environment.

Tropical cyclones can produce devastating effects on humans, animals, and the environment. Globally, it has been a recurring problem across different continents. This has necessi-tated conducting research on different aspects relating to Tropical cyclones. To contribute to this research domain, in this study, three Deep Learning (DL) models were developed to predict Tropical Cyclone (TC) intensity. The study used the Hursat and Bestrack datasets from the United States National Oceanic and Atmospheric Administration and employed Convolutional Neural Networks (CNN), Longest Short-term Memory (LSTM), and a combination of CNN and LSTM (CNN-LSTM) to predict TC intensity. Results obtained from the study show that the LSTM model achieved the best results although the difference between the three models was not wide. Contributions from this study can aid in reducing damages to life and properties associated with ropical Cyclones by improving the prediction of TC intensity.

Eye tracking data can help design effective user interfaces by showing how users visually process information. In this study, three neural network models were developed and employed to classify three types of information display methods by using eye gaze data that was collected in visual information processing behavior studies. Eye gaze data was first converted into a sequence and was fed into neural networks to predict the information display type. The results of the study show a comparison between three methods for the creation of eye tracking sequences and how they perform using three neural network models including CNN-LSTM, CNN-GRU, and 3D CNN. The results were positive with all models having an accuracy of higher than 88 percent.

Historically, eye tracking analysis has been a useful approach to identify areas of interest (AOIs) where users have specific regions of the user interface (UI) in which they are interested. Many algorithms have been proposed to analyze eye tracking data in order to make user interfaces more effective. The objective of this study is to use convolutional neural networks (CNNs) to classify eye tracking data. First, a CNN was used to classify two different web interfaces for browsing news data. Then in a second experiment, a CNN was used to classify the nationalities of users. In addition, techniques of data-preprocessing and feature-engineering were applied. The algorithm used in this research is convolutional neural network (CNN), which is famous in deep learning field. Keras framework running on top of TensorFlow was used to define and train our CNN model. The purpose of this research is to explore how feature-engineering can affect evaluation metrics about our model. The results of the study show a number of interesting patterns and generally that deep learning shows promise in the analysis of eye tracking data.

The analysis of climate data as a complex network has shown a great deal of promise over the last decade. Because of the massive size of the resulting network structure, much of the literature in this area has focused on the analysis of a single network snapshot generalized for a single period of time. Therefore, existing analysis methods typically focus on network properties instead of structural nature of the network to find interesting patterns. Data analysis and mining in complex climate networks at a higher temporal granularity, as a result, creates a challenging yet computationally prohibitive big data problem. In this paper, we extend the analysis of complex climate networks by studying their underlying community structures, an important property that is commonly exhibited in most complex networks. In particular, we first propose a new two-layer network model that captures community structures at both generalized and hierarchical levels where sub-communities are contained within the resulting generalized communities. Our two-layer network significantly reduces the complexity of modeling climate networks, and therefore reduces the computational complexity associated with analyzing big climate data. To mine meaningful climate patterns, we further extend this approach and suggest algorithms to find persistent communities that are stable over a long period of time. Our methods were tested on global air temperature data and we found signature communities that are verified by well known global temperature patterns. The results also produced some interesting discoveries related to known climate events such as the 1997/1997 El Niño and the eruption of Mt. Pinatubo in 1991.

Data from sensors and sensor networks are being collected at astronomical rates. This results in a massive dataset that is increasingly difficult to navigate to find interesting time periods where the spatial pattern of a process changes. The ability to navigate to such areas can lead to new knowledge about the factors that contribute to a spatio-temporal process. This paper proposes a method to automatically characterize sensor datasets based on a measure of spatial change over time resulting in a set of multi-granular spatio-temporal intervals. The resulting intervals can be used to focus knowledge discovery tasks at multiple temporal granularities within the dataset. Furthermore, the intervals enable a drill-down-style analysis where events of varying magnitudes can be identified within each granularity. Experiments were performed on a real-world dataset measuring NEXRAD precipitation accumulation. The results show that the multi-granular spatio-temporal intervals identify interesting time periods in the dataset as evidenced by naturally occurring events.

Environmental research and knowledge discovery both require extensive use of data stored in various sources and created in different ways for diverse purposes. We describe a new metadata approach to elicit semantic information from environmental data and implement semantic-based techniques to assist users in integrating, navigating, and mining multiple environmental data sources. Our system contains specifications of various environmental data sources and the relationships that are formed among them. User requests are augmented with semantically related data sources and automatically presented as a visual semantic network. In addition, we present a methodology for data navigation and pattern discovery using multi-resolution browsing and data mining. The data semantics are captured and utilized in terms of their patterns and trends at multiple levels of resolution. We present the efficacy of our methodology through experimental results.