Hasil untuk "artificial intelligence"

G. Lakemeyer, Bernhard Nebel

Marie-Laure Ryan

S.J.J. Smith

K. Korb, A. Nicholson

M. Minsky

Уалід Тузані, Хамід Еламрані, Мохаммед Лааруссі

The incorporation of artificial intelligence (AI) technologies into banking operations is attracting growing interest worldwide due to its ability to promote efficiency optimisation, risk reduction, and improved profitability. Major markets such as China and India have already demonstrated the positive impact of AI, particularly through intelligent credit scoring and conversational chatbots, thereby enhancing customer satisfaction. However, the Moroccan context remains understudied, despite the banking landscape gradually becoming more digitalised, and there being an urgent need for innovations focused on improving the customer experience in terms of personalisation, trust, and user-friendliness. To address this issue, we will conduct a systematic review of academic literature published between 2020 and 2025, adopting the PRISMA protocol. Focusing our research on Scopus, Web of Science, and IEEE Xplore enabled us to identify and select around 30 studies that met our strict criteria (application of AI in banking, financial performance, customer satisfaction measures, etc.). The selected articles were coded and analysed based on their empirical and methodological contributions. The results highlight the beneficial overall effect of AI on banking performance, including profitability, risk management, and operational efficiency, while emphasising the importance of trust and personalisation in improving the customer experience. In Morocco, the scarcity of empirical data makes targeted research necessary in order to assess the concrete impact of AI on local banks and support their digital transformation. In conclusion, this review suggests several areas for future research and offers a framework to inform AI adoption strategies in emerging banking institutions.

Wm To, T. Tryfonas, D. Farthing

An-Rong Wu, Sun-Yuan Hsieh, Hsin-Hung Chou et al.

Brain midline shift (MLS) indicates the severity of mass effect from intracranial lesions such as traumatic brain injury, stroke, brain tumor, or hematoma. Brain MLS can be used to determine whether patients require emergency surgery and to predict patients' prognosis. Since brain MLS is usually emergent, it must be diagnosed immediately. Therefore, this study presents a computer-aided deep-learning method for detecting MLS, aiming to predict mortality in a prognosis-predicting cohort using brain MLS and clinical in-formation. The brain midline is a 3-dimensional structure, but computed tomography (CT) slices are 2-dimensional which limits brain MLS detection. Here we propose a keypoint detection method to detect brain midline on each CT slice, acquiring brain MLS distance and area in each slice. Combined with clinical information, patient mortality can be predicted using the multilayer perceptron (MLP) model. The accuracy, precision, sensitivity, specificity, and F1-score for slice selection with the proposed model are 0.966, 0.952, 0.991, 0.932, and 0.971, respectively. Both MLS distance and volume were precisely predicted at slice-level and case-level with only the slightest error. The detected midlines were clearly separated into left and right brain with a dice coefficient of 0.98. The accuracy and AUC of the MLP model were both above 0.8. The model detected large brain MLS cases well in the prediction of outcomes in the prognosis-predicting cohort. The method performs well on slice selection and brain MLS detection, and predictions of MLS distance and volume combined with clinical information predicts the patient's prognosis well.

Zhang Bin, Feng Qian, Li Moxiao et al.

Abstract Predicting mining disaster risk levels is a critical component of intelligent mining systems. This study utilizes five common mining disaster datasets to predict various risk levels. By analyzing correlation coefficients and feature importance for each dataset, optimal evaluation indicators are identified. The Shapley Additive Explanations model is then applied to enhance interpretability. To address the presence of outliers and imbalanced data categories, the Mahalanobis Distance Discriminant Method and the Synthetic Minority Oversampling Technique algorithm based on Tomek Links are used for data preprocessing. Subsequently, Support Vector Machine, Random Forest, Extreme Gradient Boosting, one-dimensional Convolutional Neural Networks, and multi-Grained Cascade Forest algorithms are applied to the five mining disaster datasets. Comparative analysis reveals that the Deep Forest algorithm demonstrates superior performance and generalization in predicting stability levels of goaf, slope stability, rockburst intensity levels, pillar stability, and Hanging Wall stability, with prediction accuracies of 92.31%, 96.77%, 92.50%, 91.67%, and 95.00%, respectively. This research provides a systematic solution for mining disaster classification prediction, offering technical support and a scientific theoretical basis for intelligent mining development and mining safety operations.

Yasaman Khosravi, Nemat Rostami Mazouei, Hossein Badiei et al.

With the expansion of technology and artificial intelligence, professions such as auditing have undergone significant transformations. In today’s business environment, audit quality is recognized as a key factor in providing assurance to stakeholders. Artificial intelligence plays an effective role in enhancing this quality by increasing accuracy, speed, and improving data analysis. In Iran, the effective use of AI in auditing requires a proper understanding of the professional environment and the individual characteristics of auditors.The aim of this study is to propose a model for improving audit quality based on artificial intelligence and the personal attributes of auditors. This research is applied in nature, employs a descriptive-correlational method, uses quantitative data, and is conducted cross-sectionally. The statistical population includes auditors who are members of the Iranian Association of Certified Public Accountants. Data were collected through a questionnaire and analyzed using SPSS and Smart PLS software. Findings show that factors such as AI awareness and acceptance, digital skills, flexibility, creativity, and technological infrastructure significantly influence audit quality. These insights can pave the way for enhancing the quality of audit reports in Iran.

Dohyun Chun, Hae Woon Jung, Jongho Kang et al.

Objective We developed a precise, reliable artificial intelligence (AI) model for predicting the future height of children and adolescents based on anthropometric and body composition data. Materials and Methods We used an extensive longitudinal dataset from a large-scale Korean cohort study, which included 588,546 measurements from 96,485 children and adolescents aged 7–18. We developed a prediction model using the light gradient boosting method and integrated anthropometric and body composition metrics along with their standard deviation scores (SDSs) and velocity parameters. Model performance was assessed through root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). We employed Shapley additive explanations (SHAP) for model interpretability. Results The model accurately predicted future heights. For males, the average RMSE, MAE, and MAPE were 2.51 cm, 1.74 cm, and 1.14%, respectively, with female prediction results showing comparable accuracy (2.28 cm, 1.68 cm, and 1.13%, respectively). Shapley additive explanations analysis revealed that the SDS of height, height velocity, and soft lean mass velocity were key predictors of future height. The model created personalized growth curves through estimation of individual-specific height trajectories, comparison with actual measurements, and identification of key variables using local SHAP values. Conclusion Our model produces accurate and personalized growth curves, incorporating explainable AI techniques for enhanced clinical understanding. This method advances pediatric growth assessment and provides robust clinical decision support. Despite limitations including the absence of handwrist radiography comparison and Korean population specificity, our approach demonstrates significant potential for early identification of growth disorders and optimization of growth outcomes.

Šutović Milojica M.

Education represents a crucial developmental resource for any society, encompassing both its potential and prosperity. It is also a determinant of its success, an agent of socialization, and a means for learning values. It constitutes a key element of class structure and the stratification pyramid, and serves as an instrument for their maintenance. In order to maintain societal and state order, education plays a key role in institutionalizing value orientations, cementing cultural cohesion, and differentiating social roles, and frames occupations as vocations, responsibilities, obligations, and duties. Consequently, the most effective means of undermining a nation and its state is through the destruction of its educational and upbringing systems. This is achieved through mechanisms of control, surveillance, and repression, directing people towards advertisements, media, and social networks, marketing, and paternalism, all devoid of spirit, imagination, or spirituality. In this context, education within the globalized periphery has become an instrument of colonization and a tool of pedagogy aimed at subjugation and enslavement. It relegates individuals to metaphorical straitjackets, masked as neutral transitions, while enabling the rise of a volatile, gangster-style political capitalism with no alternatives. A system that values means over ends, utility over goods, earnings and profit over intellectual and human values, and relies increasingly on robots and artificial intelligence. In the near future, these technological advancements will displace many diploma-holders from the labor market. This is a new code and permit for inequality, the triumph of poverty, and the establishment of global totalitarianism, in which the classical form will become a mere swan song. This issue necessitates a reimagined discourse and a redefined understanding both theoretical and epistemological, as well as practical and educational. That is how all progress begins.

Aysenur Arpaci, Asel Usdat Ozturk, Ismail Okur et al.

Abstract Background The field of artificial intelligence (AI) has experienced considerable growth in recent years, with the advent of technologies that are transforming a range of industries, including healthcare and dentistry. Large language models (LLMs) and natural language processing (NLP) are pivotal to this transformation. This study aimed to assess the efficacy of AI-supported chatbots in responding to questions frequently asked by patients to their doctors regarding oral health. Methods Frequently asked questions in the oral health section of the World Dental Federation FDI website were asked about Google-Gemini Trends and ChatGPT-4 chatbots on July 9, 2024. Responses from ChatGPT and Gemini, as well as those from the FDI webpage, were recorded. The accuracy of the responses given by ChatGPT-4 and Gemini to the four specified questions, the detection of similarities and differences, and the comprehensive examination of ChatGPT-4 and Gemini’s capabilities were analyzed and reported by the researchers. Furthermore, the content of the texts was evaluated in terms of their similarity with respect to the following criteria: “Main Idea,” “Quality Analysis,” “Common Ideas,” and “Inconsistent Ideas.” Results It was observed that both ChatGPT-4 and Gemini exhibited performance comparable to that of the FDI responses in terms of completeness and clarity. Compared with Gemini, ChatGPT-4 provided responses that were more similar to the FDI responses in terms of relevance. Furthermore, ChatGPT-4 provided responses that were more accurate than those of Gemini in terms of the “Accuracy” criterion. Conclusions This study demonstrated that, according to the assessment conducted by FDI, the ChatGPT-4 and Gemini applications contain contemporary and comprehensible information in response to general inquiries concerning oral health. These applications are regarded as a prevalent and dependable source of information for individuals seeking to access such data.

Bechir Raggad, Elie Bouri

Abstract This study investigates the return and volatility connectedness between artificial intelligence (AI) stock ETF and each segment of the energy markets, namely clean energy, dirty energy, and WTI oil. Using a quantile-on-quantile connectedness approach on daily data from 14 September 2016 to 29 January 2024, the results reveal the following. Firstly, the degree of connectedness for the Clean-AI pair is more pronounced than that of the other pairs (AI-Dirty and AI-WTI), and Clean is mainly a receiver of return connectedness from AI stock ETF. Clean, Dirty, and WTI shift in roles to be primary transmitters of volatility shocks. Secondly, return and volatility shocks propagate more strongly at the tails of the conditional distribution than the middle of the distribution, and a dynamic analysis indicates that the average quantile-based total connectedness changes with time and strengthens during the COVID-19 outbreak. Thirdly, a portfolio and risk analysis with tail risk measures confirms the importance of considering a dynamic approach to tail-risk minimization.

Seongsik Park, Harksoo Kim

The syntactic information of a dependency tree is an essential feature in relation extraction studies. Traditional dependency-based relation extraction methods can be categorized into hard pruning methods, which aim to remove unnecessary information, and soft pruning methods, which aim to utilize all lexical information. However, hard pruning has the potential to overlook important lexical information, while soft pruning can weaken the syntactic information between entities. As a result, recent studies in relation extraction have been shifting from dependency-based methods to pre-trained language model (LM) based methods. Nonetheless, LM-based methods increasingly demand larger language models and additional data. This trend leads to higher resource consumption, longer training times, and increased computational costs, yet often results in only marginal performance improvements. To address this problem, we propose a relation extraction model based on an entity-centric dependency tree: a dependency tree that is reconstructed by considering entities as root nodes. Using the entity-centric dependency tree, the proposed method can capture the syntactic information of an input sentence without losing lexical information. Additionally, we propose a novel model that utilizes entity-centric dependency trees in conjunction with language models, enabling efficient relation extraction without the need for additional data or larger models. In experiments with representative sentence-level relation extraction datasets such as TACRED, Re-TACRED, and SemEval 2010 Task 8, the proposed method achieves F1-scores of 74.9%, 91.2%, and 90.5%, respectively, which are state-of-the-art performances.

Olszewski Adam

The main purpose of this paper, which takes the form of an essay, is an attempt to answer the question of the limits of artificial intelligence (AI). In the introductory section, we present the key milestones in AI development, both historical and future projections, in which two terms – Artificial Human (AH) and Artificial ‘god’ (AG) – play a special role. In the second section, we clarify the question of the limits of AI by indicating the hypothetical goal of AI development. The third section develops the argument proposed by C. F. Weizsäcker, originally formulated for cybernetics. The conclusion of this argument is optimistic about limitations to the possibilities of cybernetic simulations. We apply this argument to AI and subject it to a critique which ultimately undermines the legitimacy of its conclusion. We base the critique on two well-known results: the theorem of the unsolvability of the halting problem and Gödel’s first incompleteness theorem, and we formulate two objections interpreted without adopting Church’s thesis. In the crucial fourth section, we present a third objection in the form of a hypothesis for which we argue that AI (AH), understood as a subject, will always be solipsistic.

Ladislav Zjavka

Microgrid autonomous networks need an effective plan and control of power supply, energy storage, and retransmission. Prediction and monitoring of power quality (PQ) along with efficient utilization of Renewable Energy (RE) is unavoidable to optimize the system performance without abnormalities. Alterations and irregularities in PQ must remain within the prescribed norm ranges and characteristics to allow fault-tolerant operation of the detached system in various modes of attached equipment. The PQ data for all possible combinations of grid-attached household appliances and different inside/outside conditions cannot be measured completely or described exactly by physical equations. PQ predictions on a daily basis using Artificial Intelligence (AI) models are needed because atmospheric fluctuations and anomalies in local weather with uncertainties in system states primarily influence the induced power and operation of real off-grids. A novel soft-computing method using Differential Learning, which allows modelling of complex dynamics of weather-dependent systems, is presented and compared with the recent standard deep and probabilistic machine learning. The AI models were evolved using weather data and the binary status of attached equipment in the test predetermined daily training periods. Daily statistical models process 24-h forecast data and definition load series of trained input variables to calculate the target PQ parameters at the same times. Optimal utilization, efficiency, and failure-free operation of smart grids can be planned according to the suggested operable power consumption scenarios based on their PQ verification on a day-horizon. Executable load sequences can be automatically combined and scheduled in the system to be adapted to user needs, considering the RE production potential, charge state, and optimal PQ characteristics over the next 24 h. A parametric C++ application software with applied PQ and weather data is free available to allow reproducibility of the results.

Minhyeok Lee

The evolving field of generative artificial intelligence (GenAI), particularly generative deep learning, is revolutionizing a host of scientific and technological sectors. One of the pivotal innovations within this domain is the emergence of generative adversarial networks (GANs). These unique models have shown remarkable capabilities in crafting synthetic data, closely emulating real-world distributions. Notably, their application to gene expression data systems is a fascinating and rapidly growing focus area. Restrictions related to ethical and logistical issues often limit the size, diversity, and data-gathering speed of gene expression data. Herein lies the potential of GANs, as they are capable of producing synthetic gene expression data, offering a potential solution to these limitations. This review provides a thorough analysis of the most recent advancements at this innovative crossroads of GANs and gene expression data, specifically during the period from 2019 to 2023. In the context of the fast-paced progress in deep learning technologies, accurate and inclusive reviews of current practices are critical to guiding subsequent research efforts, sharing knowledge, and catalyzing continual growth in the discipline. This review, through highlighting recent studies and seminal works, serves as a key resource for academics and professionals alike, aiding their journey through the compelling confluence of GANs and gene expression data systems.

Meda Andrijauskiene, Dimosthenis Ioannidis, Daiva Dumciuviene et al.

Though much attention is dedicated to the development of its research and innovation policy, the European Union constantly struggles to match the level of the strongest innovators in the world. Therefore, there is a necessity to analyze the individual efforts and conditions of the 27 member states that might determine their final innovative performance. The results of a scientific literature review showed that there is a growing interest in the usage of artificial intelligence when seeking to improve decision-making processes. Data envelopment analysis, as a branch of computational intelligence methods, has proved to be a reliable tool for innovation efficiency evaluation. Therefore, this paper aimed to apply DEA for the assessment of the European Union’s innovation efficiency from 2000 to 2020, when innovation was measured by patent, trademark, and design applications. The findings showed that the general EU innovation efficiency situation has improved over time, meaning that each programming period was more successful than the previous one. On the other hand, visible disparities were found across the member states, showing that Luxembourg is an absolute innovation efficiency leader, while Greece and Portugal achieved the lowest average efficiency scores. Both the application of the DEA method and the gathered results may act as viable guidelines on how to improve R&I policies and select future investment directions.

Nian TAO, Sheng ZHANG, Tai FU

With the continuous development and maturity of new-generation information technologies such as 5G, artificial intelligence, and the Internet of things, the construction of smart communities has entered the second half, with greater space and higher requirements.Firstly, the development status of community and business needs was analyzed.Secondly, the design ideas based on “pan-sensing, data collection, and intelligent application” was proposed for the pain points of users, such as government, property management and residents.Thirdly, an overall plan for 5G+AIoT smart community was built, focusing on wisdom community 5G application scenarios, the industrial and social benefits of smart community construction were discussed.Finally, the future development trend of smart communities was looked forward.