Search Results (7)

The preservation of bridge infrastructure has become increasingly critical as aging assets face accelerated deterioration due to climate change, environmental loading, and operational stressors. This issue is particularly pronounced in regions with limited maintenance budgets, where delayed interventions compound structural vulnerabilities. Although traditional bridge inspections generate detailed condition ratings, these are often viewed as isolated snapshots rather than part of a continuous structural health timeline, limiting their predictive value. To overcome this, recent studies have employed various Artificial Intelligence (AI) models. However, these models are often restricted by fixed input sizes and specific report formats, making them less adaptable to the variability of real-world data. Thus, this study introduces a Transformer architecture inspired by Natural Language Processing (NLP), treating condition ratings, and other features as tokens within temporally ordered inspection “sentences” spanning 1993–2024. Due to the self-attention mechanism, the model effectively captures long-range dependencies in patterns, enhancing forecasting accuracy. Empirical results demonstrate 96.88% accuracy for short-term prediction and 86.97% across seven years, surpassing the performance of comparable time-series models such as Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRUs). Ultimately, this approach enables a data-driven paradigm for structural health monitoring, enabling bridges to “speak” through inspection data and empowering engineers to “listen” with enhanced precision. Full article

With the advancement of information technology, artificial intelligence (AI) has demonstrated significant potential in clinical prediction, helping to improve the level of intelligent medical care. Current clinical practice primarily relies on patients’ time series data and clinical notes to predict health status and makes predictions by simply concatenating cross-modal features. However, they not only ignore the inherent correlation between cross-modal features, but also fail to analyze the collaborative representation of multi-granularity features from diverse perspectives. To address these challenges, we propose a deep dynamic memory-driven cross-modal feature representation network for clinical outcome prediction. Specifically, we use a Bi-directional Gated Recurrent Unit (BiGRU) network to capture dynamic features in time series data and a dual-view feature encoding model with sentence-aware and entity-aware capabilities to extract clinical text features from global semantic and local concept perspectives, respectively. Furthermore, we introduce a memory-driven cross-modal attention mechanism, which dynamically establishes deep correlations between clinical text and time series features through learnable memory matrices. In addition, we also introduce a memory-aware constrained layer normalization to alleviate the challenges of multi-modal feature heterogeneity. Besides, we use gating mechanisms and dynamic memory components to enable the model to learn feature information of different historical-current patterns, further improving the model’s performance. Lastly, we combine the integrated gradients for feature attribution analysis to enhance the model’s interpretability. Finally, we evaluate the model’s performance on the MIMIC-III dataset, and the experimental results demonstrate that the model outperforms current advanced baselines in clinical outcome prediction tasks. Notably, our model maintains high predictive accuracy and robustness even when faced with imbalanced data. It can also provide a new perspective for researchers in the field of AI medicine. Full article

As the population ages, Alzheimer’s disease (AD) and Parkinson’s disease (PD) are increasingly common neurodegenerative diseases among the elderly. Human voice signals contain various characteristics, and the voice recording signals with time-series properties include key information such as pitch, tremor, and breathing cycle. Therefore, this study aims to propose an algorithm to classify normal individuals, Alzheimer’s patients, and Parkinson’s patients using these voice signal characteristics. The study subjects consist of a total of 700 individuals, who provided data by uttering 40 predetermined sentences. To extract the main characteristics of the recorded voices, a Mel–spectrogram was used, and these features were analyzed using a Convolutional Neural Network (CNN). The analysis results showed that the classification based on DenseNet exhibited the best performance. This study suggests the potential for classification of cognitive impairment through voice signal analysis. Full article

Analyzing time-series sentencing data presents many challenges. The data have many dimensions and change with time. This makes it difficult to identify patterns and discuss their similarities over time. This work proposes a machine learning approach to associate patterns with clusters. This allows a representation of sentencing data regarding trajectories in the appropriate (time, cluster) space. We propose to use the Hamming distance of trajectories to measure the similarity of sentencing data across districts. For any offense, we can define the average Hamming distance that has a simple interpretation as the average period when sentencing patterns are different. We introduce simple statistical measures on trajectories to show similarities and changes in sentencing behavior over time. We illustrate our approach by analyzing sentencing data for narcotics and retail theft. Full article

Adductor spasmodic dysphonia (AdSD) disrupts laryngeal muscle control during speech and, therefore, affects the onset and offset of phonation. In this study, the goal is to use laryngeal high-speed videoendoscopy (HSV) to measure the glottal attack time (GAT) and glottal offset time (GOT) during connected speech for normophonic (vocally normal) and AdSD voices. A monochrome HSV system was used to record readings of six CAPE-V sentences and part of the “Rainbow Passage” from the participants. Three raters visually analyzed the HSV data using a playback software to measure the GAT and GOT. The results show that the GAT was greater in the AdSD group than in the normophonic group; however, the clinical significance of the amount of this difference needs to be studied further. More variability was observed in both GATs and GOTs of the disorder group. Additionally, the GAT and GOT time series were found to be nonstationary for the AdSD group while they were stationary for the normophonic voices. This study shows that the GAT and GOT measures can be potentially used as objective markers to characterize AdSD. The findings will potentially help in the development of standardized measures for voice evaluation and the accurate diagnosis of AdSD. Full article

This paper presents an attempt at establishing an association between crime levels and prison populations across European countries. We observe that the situation in Central and Eastern European countries differs distinctly from the rest of Europe. Building on this, we offer justification that is methodologically based on correlations and regressions of country incarceration rates on crime rates, with reference to governance indicators. Our cross-sectional analysis uses data on crime and prisoner rates by offence from Eurostat and SPACE for the year 2018. The paper’s empirical analysis is preceded by a discussion of the challenges faced when attempting to compare crime between countries in Europe. A review of research focused on relationships between incarceration and crime follows, with the emphasis on the deterrence effect and the prison paradox. Typically, this stream of research uses microdata covering a single country or limited to a smaller geographic area. International comparisons are rare, and are usually based on time series and trend analyses. The quantitative approach applied here is based on recognizing two clusters of countries: the Central and Eastern European (CEE) cluster and the Western European (WE) cluster. We show that the observation of higher prisoner rates and lower crime rates for CEE countries is confirmed with regression analysis. Our study encompasses four types of offences: assault, rape, robbery, and theft. The final section of the paper presents an attempt to incorporate Worldwide Governance Indicators into the analysis of the association between incarceration and crime rates. The results confirm that crime rates in WE countries are distinctly higher than in CEE countries, while incarceration rates in WE are significantly lower than in CEE countries. We think this is due to a higher percentage of crimes being reported and the greater accuracy of police statistics in WE countries. The prison population in each country is largely determined by its criminal and penal policies, which differ substantially between CEE and WE countries (e.g., in terms of frequency of imposing prison sentences and the length of imprisonment). These tendencies result in higher incarceration rates in CEE countries, despite lower crime rates when compared to WE countries. Full article

Arabic is one of the official languages recognized by the United Nations (UN) and is widely used in the middle east, and parts of Asia, Africa, and other countries. Social media activity currently dominates the textual communication on the Internet and potentially represents people’s views about specific issues. Opinion mining is an important task for understanding public opinion polarity towards an issue. Understanding public opinion leads to better decisions in many fields, such as public services and business. Language background plays a vital role in understanding opinion polarity. Variation is not only due to the vocabulary but also cultural background. The sentence is a time series signal; therefore, sequence gives a significant correlation to the meaning of the text. A recurrent neural network (RNN) is a variant of deep learning where the sequence is considered. Long short-term memory (LSTM) is an implementation of RNN with a particular gate to keep or ignore specific word signals during a sequence of inputs. Text is unstructured data, and it cannot be processed further by a machine unless an algorithm transforms the representation into a readable machine learning format as a vector of numerical values. Transformation algorithms range from the Term Frequency–Inverse Document Frequency (TF-IDF) transform to advanced word embedding. Word embedding methods include GloVe, word2vec, BERT, and fastText. This research experimented with those algorithms to perform vector transformation of the Arabic text dataset. This study implements and compares the GloVe and fastText word embedding algorithms and long short-term memory (LSTM) implemented in single-, double-, and triple-layer architectures. Finally, this research compares their accuracy for opinion mining on an Arabic dataset. It evaluates the proposed algorithm with the ASAD dataset of 55,000 annotated tweets in three classes. The dataset was augmented to achieve equal proportions of positive, negative, and neutral classes. According to the evaluation results, the triple-layer LSTM with fastText word embedding achieved the best testing accuracy, at 90.9%, surpassing all other experimental scenarios. Full article