Big Data and Cognitive Computing

Three-dimensional Occupancy Prediction provides a detailed representation of the surrounding environment, essential for autonomous driving. Long temporal image sequence fusion is a common technique used to improve the occupancy prediction performance. However, existing temporal fusion methods are inefficient due to three issues: repetitive feature extraction from temporal images, redundant fusion of temporal features, and suboptimal fusion of long-term historical features. To address these challenges, we propose the Recurrent Occupancy Prediction Network (RecurrentOcc). We introduce the Scene Memory Gate, a new temporal fusion module that condenses temporal scene features into a single historical feature map. This eliminates the need for repeated extraction and aggregation of multiple temporal images, reducing computational overhead. The Scene Memory Gate selectively retains valuable information from historical features and recurrently updates the historical feature map, enhancing temporal fusion performance. Additionally, we design a simple yet efficient encoder, significantly reducing the number of model parameters. Compared with other real-time methods, RecurrentOcc achieves state-of-the-art performance of 39.9 mIoU on the Occ3D-NuScenes dataset with the fewest parameters of 59.1 M and an inference speed of 23.4 FPS. Full article

Career planning agencies and other organizations can help workers if they are able to effectively identify related occupations that are relevant to the task at hand. Occupational knowledge bases such as O*NET and ESCO represent mature attempts to categorize occupations and describe them in detail so that they can be used to search for related occupations. Vector databases offer an opportunity to find related occupations based on large pre-trained word and sentence embeddings and their associated retrieval algorithms for similarity search. This paper reports a systematic empirical evaluation of the possibilities of using vector databases for related occupation retrieval using different document structures, embeddings, and retrieval configurations for two popular open source vector databases, and using the O*NET curated database. The objective was to understand the extent to which curated relations capture all the meaningful relations in a context of retrieval. The results show that, independent of the database used, distance metrics, sentence embeddings, and the selection of text fragments are all significant in the overall retrieval performance when comparing with curated relations, but they also retrieve other relevant occupations based on text similarity. Further, the precision is high for smaller cutoffs in the results list, which is especially important for settings in which vector database retrieval is set up as part of a Retrieval Augmented Generation (RAG) pattern. The inspection of highly ranked retrieved related occupations not explicit in the curated database reveals that text similarity captures the taxonomical grouping of some occupations in some cases, but also other cross-cuts different aspects that are distinct from the hierarchical organization of the database in most of the cases. This suggests that text retrieval should be combined with querying explicit relations in practical applications. Full article

Ensuring reliability, availability, and security in modern software systems hinges on early fault detection, yet predicting which parts of a codebase are most at risk remains a significant challenge. In this paper, we analyze 2.4 million commits drawn from 33 heterogeneous open-source projects, spanning healthcare, security tools, data processing, and more. By examining each repository per file and per commit, we derive process metrics (e.g., churn, file age, revision frequency) alongside size metrics and entropy-based indicators of how scattered changes are over time. We train and tune a gradient boosting model to classify bug-prone commits under realistic class-imbalance conditions, achieving robust predictive performance across diverse repositories. Moreover, a comprehensive feature-importance analysis shows that files with long lifespans (high age), frequent edits (revision count), and widely scattered changes (entropy metrics) are especially vulnerable to defects. These insights can help practitioners and researchers prioritize testing and tailor maintenance strategies, ultimately strengthening software dependability. Full article

The transfer of learning (TL) is the process of applying knowledge and skills learned in one context to a new and different context. Efficient use of memory is essential in achieving successful TL and good learning outcomes. This study uses a cognitive computing approach to identify and explore brain activity patterns related to memory efficiency in the context of learning a new programming language. This study hypothesizes that prior programming knowledge reduces cognitive load, leading to improved memory efficiency. Spatio-temporal brain data (STBD) were collected from a sample of participants (n = 26) using an electroencephalogram (EEG) device and analyzed by applying a spiking neural network (SNN) approach and the SNN-based NeuCube architecture. The findings revealed the neural patterns demonstrating the effect of prior knowledge on memory efficiency. They showed that programming learning outcomes were aligned with specific theta and alpha waveband spike activities concerning prior knowledge and cognitive load, indicating that cognitive load was a feasible metric for measuring memory efficiency. Building on these findings, this study proposes that the methodology developed for examining the relationship between prior knowledge and TL in the context of learning a programming language can be extended to other educational domains. Full article

Fake news has eroded trust in credible news sources, driving the need for tools to verify the accuracy of circulating information. Fact verification addresses this issue by classifying claims as Supports (S), Refutes (R), or Not Enough Info (NEI) based on evidence. Neural Semantic Matching Networks (NSMN) is an algorithm designed for this purpose, but its reliance on BiLSTM has shown limitations, particularly overfitting. This study aims to enhance NSMN for fact verification through a structured framework comprising encoding, alignment, matching, and output layers. The proposed approach employed Siamese MaLSTM in the matching layer and introduced the Manhattan Fact Relatedness Score (MFRS) in the output layer, culminating in a novel algorithm called Deep One-Directional Neural Semantic Siamese Network (DOD–NSSN). Performance evaluation compared DOD–NSSN with NSMN and transformer-based algorithms (BERT, RoBERTa, XLM, XL-Net). Results demonstrated that DOD–NSSN achieved 91.86% accuracy and consistently outperformed other models, achieving over 95% accuracy across diverse topics, including sports, government, politics, health, and industry. The findings highlight the DOD–NSSN model’s capability to generalize effectively across various domains, providing a robust tool for automated fact verification. Full article

Understanding the mindset of people who die by suicide remains a key research challenge. We map conceptual and emotional word–word co-occurrences in 139 genuine suicide notes and in reference word lists, an Emotional Recall Task, from 200 individuals grouped by high/low depression, anxiety, and stress levels on DASS-21. Positive words cover most of the suicide notes’ vocabulary; however, co-occurrences in suicide notes overlap mostly with those produced by individuals with low anxiety (Jaccard index of 0.42 for valence and 0.38 for arousal). We introduce a “words not said” method: It removes every word that corpus A shares with a comparison corpus B and then checks the emotions of “residual” words in $A-B$. With no leftover emotions, A and B are similar in expressing the same emotions. Simulations indicate this method can classify high/low levels of depression, anxiety and stress with 80% accuracy in a balanced task. After subtracting suicide note words, only the high-anxiety corpus displays no significant residual emotions. Our findings thus pin anxiety as a key latent feature of suicidal psychology and offer an interpretable language-based marker for suicide risk detection. Full article

This paper presents an analysis of the severity of alcohol use disorder (AUD) based on electroencephalogram (EEG) signals and alcohol drinking experiments by utilizing power spectral density (PSD) and the transitions that occur as individuals drink alcohol in increasing amounts. We use data from brain—computer interface (BCI) experiments using alcohol as a stimulus recorded from a group of seventeen alcohol-drinking male participants and the assessment scores of the alcohol use disorders identification test (AUDIT). This method investigates the mild, moderate, and severe symptoms of AUD using the three key domains of AUDIT, which are hazardous alcohol use, dependence symptoms, and severe alcohol use. We utilize the EEG spectral power of the theta, alpha, and beta frequency bands by observing the transitions from the initial to the final phase of alcohol consumption. Our results are compared for people with low-risk alcohol consumption, harmful or hazardous alcohol consumption, and lastly a likelihood of AUD based on the individual assessment scores of the AUDIT. We use Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) to cluster the results of the transitions in EEG signals and the overall brain activity of all the participants for the entire duration of the alcohol-drinking experiments. This study can be useful in creating an automatic AUD severity level detection tool for alcoholics to aid in early intervention and supplement evaluations by mental health professionals. Full article

Many research studies have designed intelligent traffic light scheduling algorithms. Some researchers rely on specialized sensors and hardware to gather real-time traffic data at signalized road intersections. Others benefit from artificial intelligence techniques and/or cloud computing technologies. The technology of vehicular networks has been widely used to gather the traffic characteristics of competing traffic flows at signalized road intersections. Intelligent traffic light controlling systems aim to fairly liberate competing traffic at signalized road intersections and eliminate traffic crises. These algorithms have been initially developed without focusing on the consequences of security threats or attacks. However, the accuracy of gathered traffic data at each road intersection affects its performance. Fake and corrupted packets highly affect the accuracy of the gathered traffic data. Thus, in this work, we aim to investigate the aspects of security and confidentiality of intelligent traffic light systems. The possible attacks on the confidentiality of intelligent traffic light systems are examined. Then, a confidential traffic light control system that protects the privacy of traveling vehicles and drivers is presented. The proposed algorithm mainly prevents unauthorized traceability and linkability attacks that threaten people’s lives and violate their privacy. Finally, the proposed algorithm is evaluated through extensive experiments to verify its correctness and benefits compared to traditional insecure intelligent traffic light systems. Full article

The surge in credit fraud incidents poses a critical threat to financial systems, driving the need for robust and adaptive fraud detection solutions. While various predictive models have been developed, existing approaches often struggle with two persistent challenges: extreme class imbalance and delays in detecting fraudulent activity. In this study, we propose an unsupervised Adversarial Autoencoder (AAE) framework designed to tackle these challenges simultaneously. The results highlight the potential of our approach as a scalable, interpretable, and adaptive solution for real-world credit fraud detection systems. Full article

The rapid development of unmanned aerial vehicles (UAVs) has had a significant impact on the growth of the economic, industrial, and social welfare of society. The possibility of reaching places that are difficult and dangerous for humans to access with minimal use of third-party resources increases the efficiency and quality of maintenance of construction structures, agriculture, and exploration, which are carried out with the help of drones with a predetermined trajectory. The widespread use of UAVs has caused problems with the control of the drones’ correctness following a given route, which leads to emergencies and accidents. Therefore, UAV monitoring with video cameras is of great importance. In this paper, we propose a Yolov12 architecture with positive–negative pulse-based optimization algorithms to solve the problem of drone detection on video data. Self-attention-based mechanisms in transformer neural networks (NNs) improved the quality of drone detection on video. The developed algorithms for training NN architectures improved the accuracy of drone detection by achieving the global extremum of the loss function in fewer epochs using positive–negative pulse-based optimization algorithms. The proposed approach improved object detection accuracy by 2.8 percentage points compared to known state-of-the-art analogs. Full article

Neural-network-based models have made considerable progress in many computer vision areas over recent years. However, many works have exposed their vulnerability to malicious input data manipulation—that is, to adversarial attacks. Although many recent works have thoroughly examined the adversarial robustness of classifiers, the robustness of Image Quality Assessment (IQA) methods remains understudied. This paper addresses this gap by proposing FM-GOAT (Frequency-Masked Gradient Orthogonalization Attack), a novel white box adversarial method tailored for no-reference IQA models. Using a novel gradient orthogonalization technique, FM-GOAT uniquely optimizes adversarial perturbations against multiple perceptual constraints to minimize visibility, moving beyond traditional $l_p$-norm bounds. We evaluate FM-GOAT on seven state-of-the-art NR-IQA models across three image and video datasets, revealing significant vulnerability to the proposed attack. Furthermore, we examine the applicability of adversarial purification methods to the IQA task, as well as their efficiency in mitigating white box adversarial attacks. By studying the activations from models’ intermediate layers, we explore their behavioral patterns in adversarial scenarios and discover valuable insights that may lead to better adversarial detection. Full article

Electronic payments are becoming increasingly popular in Taiwan; however, there is a lack of studies examining the factors affecting user satisfaction with electronic payments in Taiwan. This study focuses on Android phone users to identify key factors influencing their experiences based on user reviews of electronic payment mobile applications. It analyzes which factors contribute to positive satisfaction and which lead to negative experiences. The study employed BERTopic for topic modeling, which flexibly accommodates multiple languages, enabling effective examination of reviews written in Chinese. Additionally, we utilized the semantic understanding capabilities of large-scale language models to preliminarily name the generated topics with the help of ChatGPT Plus. These preliminary names were then manually refined to determine the final topic titles. The findings reveal that for Android phone users, electronic payment services that enhance user convenience and offer discounts tend to foster positive satisfaction. Conversely, the instability of electronic payment applications results in many user complaints. These research results can provide valuable insights for specialized electronic payment institutions in Taiwan to enhance their services. Full article

The Draw-A-Person Intellectual Ability test for children, adolescents, and adults is a widely used tool in psychology for assessing intellectual ability. This test relies on human drawings for initial raw scoring, with the subsequent conversion of data into IQ ranges through manual procedures. However, this manual scoring and IQ assessment process can be time-consuming, particularly for busy psychologists dealing with a high caseload of children and adolescents. Presently, DAP-IQ screening continues to be a manual endeavor conducted by psychologists. The primary objective of our research is to streamline the IQ screening process for psychologists by leveraging deep learning algorithms. In this study, we utilized the DAP-IQ manual to derive IQ measurements and categorized the entire dataset into seven distinct classes: Very Superior, Superior, High Average, Average, Below Average, Significantly Impaired, and Mildly Impaired. The dataset for IQ screening was sourced from primary to high school students aged from 8 to 17, comprising over 1100 sketches, which were subsequently manually classified under the DAP-IQ manual. Subsequently, the manual classified dataset was converted into digital images. To develop the artificial intelligence-based models, various deep learning algorithms were employed, including Convolutional Neural Network (CNN) and state-of-the-art CNN (Transfer Learning) models such as Mobile-Net, Xception, InceptionResNetV2, and InceptionV3. The Mobile-Net model demonstrated remarkable performance, achieving a classification accuracy of 98.68%, surpassing the capabilities of existing methodologies. This research represents a significant step towards expediting and enhancing the IQ screening for psychologists working with diverse age groups. Full article

The accurate forecasting of key performance indicators (KPIs) is essential for enhancing the reliability and operational efficiency of engineering systems under increasingly complex security challenges. However, existing approaches often neglect the heterogeneous nature of multivariate time series data, particularly the consistency of measurements and the influence of external factors, which limits their effectiveness in real-world scenarios. In this work, a Consistency-aware Hybrid Static-Dynamic Multivariate forecasting Network (CHSDM-Net) is proposed, which first applies a consistency-aware, optimization-driven segmentation to ensure high internal consistency within each segment across multiple variables. Secondly, a hybrid forecasting model integrating a Static Representation Module and a Dynamic Temporal Disentanglement and Attention Module for static and dynamic data fusion is proposed. For the dynamic data, the trend and periodic components are disentangled and fed into Trend-wise Attention and Periodic-aware Attention blocks, respectively. Extensive experiments on both synthetic and real-world radar detection datasets demonstrated that CHSDM-Net achieved significant improvements compared with existing methods. Comprehensive ablation and sensitivity analyses further validated the effectiveness and robustness of each component. The proposed method offers a practical and generalizable solution for intelligent KPI forecasting and decision support in industrial engineering applications. Full article

Large-language-model (LLM) APIs demonstrate impressive reasoning capabilities, but their size, cost, and closed weights limit the deployment of knowledge-aware AI within biomedical research groups. At the other extreme, standard attention-based neural language models (SANLMs)—including encoder–decoder architectures such as Transformers, Gated Recurrent Units (GRUs), and Long Short-Term Memory (LSTM) networks—are computationally inexpensive. However, their capacity for semantic reasoning in noisy, open-vocabulary knowledge bases (KBs) remains unquantified. Therefore, we investigate whether compact SANLMs can (i) reason over hybrid OpenIE-derived KBs that integrate commonsense, general-purpose, and non-communicable-disease (NCD) literature; (ii) operate effectively on commodity GPUs; and (iii) exhibit semantic coherence as assessed through manual linguistic inspection. To this end, we constructed four training KBs by integrating ConceptNet (600k triples), a 39k-triple general-purpose OpenIE set, and an 18.6k-triple OpenNCDKB extracted from 1200 PubMed abstracts. Encoder–decoder GRU, LSTM, and Transformer models (1–2 blocks) were trained to predict the object phrase given the subject + predicate. Beyond token-level cross-entropy, we introduced the Meaning-based Selectional-Preference Test (MSPT): for each withheld triple, we masked the object, generated a candidate, and measured its surplus cosine similarity over a random baseline using word embeddings, with significance assessed via a one-sided t-test. Hyperparameter sensitivity (311 GRU/168 LSTM runs) was analyzed, and qualitative frame–role diagnostics completed the evaluation. Our results showed that all SANLMs learned effectively from the point of view of the cross entropy loss. In addition, our MSPT provided meaningful semantic insights: for the GRUs (256-dim, 2048-unit, 1-layer): mean similarity $\left({\mu }_{sts}\right)$ of 0.641 to the ground truth vs. 0.542 to the random baseline (gap 12.1%; $p<{10}^{-180}$). For the 1-block Transformer: ${\mu }_{sts}=0.551$ vs. $0.511$ (gap 4%; $p<{10}^{-25}$). While Transformers minimized loss and accuracy variance, GRUs captured finer selectional preferences. Both architectures trained within <24 GB GPU VRAM and produced linguistically acceptable, albeit over-generalized, biomedical assertions. Due to their observed performance, LSTM results were designated as baseline models for comparison. Therefore, properly tuned SANLMs can achieve statistically robust semantic reasoning over noisy, domain-specific KBs without reliance on massive LLMs. Their interpretability, minimal hardware footprint, and open weights promote equitable AI research, opening new avenues for automated NCD knowledge synthesis, surveillance, and decision support. Full article

This paper proposes a comprehensive deep-learning framework, SentiStack, for Bitcoin price forecasting and trading strategy evaluation by integrating multimodal data sources, including market indicators, macroeconomic variables, and sentiment information extracted from financial news and social media. The model architecture is based on a Stacking-LSTM ensemble, which captures complex temporal dependencies and non-linear patterns in high-dimensional financial time series. To enhance predictive power, sentiment embeddings derived from full-text analysis using the DeepSeek language model are fused with traditional numerical features through early and late data fusion techniques. Empirical results demonstrate that the proposed model significantly outperforms baseline strategies, including Buy & Hold and Random Trading, in cumulative return and risk-adjusted performances. Feature ablation experiments further reveal the critical role of sentiment and macroeconomic inputs in improving forecasting accuracy. The sentiment-enhanced model also exhibits strong performance in identifying high-return market movements, suggesting its practical value for data-driven investment decision-making. Overall, this study highlights the importance of incorporating soft information, such as investor sentiment, alongside traditional quantitative features in financial forecasting models. Full article

With the demand for workflow processing driven by edge computing in the Internet of Things (IoT) and cloud computing growing at an exponential rate, task scheduling in heterogeneous distributed systems has become a key challenge to meet real-time constraints in resource-constrained environments. Existing studies now attempt to achieve the best balance in terms of time constraints, energy efficiency, and system reliability in Dynamic Voltage and Frequency Scaling environments. This study proposes a two-stage collaborative optimization strategy. With the help of an innovative algorithm design and theoretical analysis, the multi-objective optimization challenges mentioned above are systematically solved. First, based on a reliability-constrained model, we propose a topology-aware dynamic priority scheduling algorithm (EAWRS). This algorithm constructs a node priority function by incorporating in-degree/out-degree weighting factors and critical path analysis to enable multi-objective optimization. Second, to address the time-varying reliability characteristics introduced by DVFS, we propose a Fibonacci search-based dynamic frequency scaling algorithm (SEFFA). This algorithm effectively reduces energy consumption while ensuring task reliability, achieving sub-optimal processor energy adjustment. The collaborative mechanism of EAWRS and SEFFA has well solved the dynamic scheduling challenge based on DAG in heterogeneous multi-core processor systems in the Internet of Things environment. Experimental evaluations conducted at various scales show that, compared with the three most advanced scheduling algorithms, the proposed strategy reduces energy consumption by an average of 14.56% (up to 58.44% under high-reliability constraints) and shortens the makespan by 2.58–56.44% while strictly meeting reliability requirements. Full article

Real-time ash content control in dense medium coal separation is challenged by time lags between detection and density adjustment, along with nonlinear/noisy signals. This study proposes a hybrid model for clean coal ash content in dense medium separation by integrating empirical mode decomposition, long short-term memory networks, and sparrow search algorithm optimization. A key innovation lies in removing noise-containing intrinsic mode functions (IMFs) via EMD to ensure clean signal input to the LSTM model. Utilizing production data from a Shanxi coal plant, EMD decomposes ash content time series into intrinsic mode functions (IMFs) and residuals. High-frequency noise-containing IMFs are selectively removed, while LSTM predicts retained components. SSA optimizes LSTM parameters (learning rate, hidden layers, epochs) to minimize prediction errors. Results demonstrate the EMD-IMF1-LSTM-SSA model achieves superior accuracy (RMSE: 0.0099, MAE: 0.0052, MAPE: 0.047%) and trend consistency (NSD: 12), outperforming baseline models. The study also proposes the novel “Vector Value of the Radial Difference (VVRD)” metric, which effectively quantifies prediction trend accuracy. By resolving time-lag issues and mitigating noise interference, the model enables precise ash content prediction 16 min ahead, supporting automated density control, reduced energy waste, and eco-friendly coal processing. This research provides practical tools and new metrics for intelligent coal separation in the context of green mining. Full article

Steel plays a fundamental role in modern smart city development, where its surface structural integrity is decisive for operational safety and long-term sustainability. While deep learning approaches show promise, their effectiveness remains limited by inadequate receptive field adaptability, suboptimal feature fusion strategies, and insufficient sensitivity to small defects. To overcome these limitations, we propose FAD-Net, a deep learning framework specifically designed for surface defect detection in steel materials within urban infrastructure. The network incorporates three key innovations: The RFCAConv module, which leverages dynamic receptive field construction and coordinate attention mechanisms to enhance feature representation for defects with long-range spatial dependencies and low-contrast characteristics. The MSDFConv module, employing multi-scale dilated convolutions with optimized dilation rates to preserve fine details while expanding the receptive field. An Auxiliary Head that introduces hierarchical supervision to improve the detection of small-scale defects. Experiments on the GC10-DET dataset showed that FAD-Net achieved 5.0% higher mAP@0.5 than baseline models. Cross-dataset validation with NEU and RDD2022 further confirmed its robustness. These results demonstrate FAD-Net’s effectiveness for automated infrastructure health monitoring. Full article