Sample Data Research Articles

Real-time phishing detection using deep learning methods by extensions

Phishing is an attack method that relies on a user’s insufficient vigilance and understanding of the internet. For example, an attacker creates an online transaction website and tricks users into logging into the fake website to steal their personal information, such as credit card numbers, email addresses, phone numbers, and physical addresses. This paper proposes implementing an extension to prevent phishing for internet users. In particular, this study develops a smart warning feature for the proposed extension using deep learning models. The proposed extension installed in the web browser protects users by checking for, warning about, and preventing untrusted connections. This study evaluated and compared the performance of machine learning models using a malicious uniform resource locator (URL) dataset containing 651,191 data samples. The results of the investigation confirm that the proposed extension using a convolutional neural network (CNN) achieved a high accuracy of 98.4%.

An investigation of machine learning algorithms and data augmentation techniques for diabetes diagnosis using class imbalanced BRFSS dataset

Diabetes is a prevalent chronic condition that poses significant challenges to early diagnosis and identifying at-risk individuals. Machine learning plays a crucial role in diabetes detection by leveraging its ability to process large volumes of data and identify complex patterns. However, imbalanced data, where the number of diabetic cases is substantially smaller than non-diabetic cases, complicates the identification of individuals with diabetes using machine learning algorithms. This study focuses on predicting whether a person is at risk of diabetes, considering the individual’s health and socio-economic conditions while mitigating the challenges posed by imbalanced data. We employ several data augmentation techniques, such as oversampling (Synthetic Minority Over Sampling for Nominal Data, i.e.SMOTE-N), undersampling (Edited Nearest Neighbor, i.e. ENN), and hybrid sampling techniques (SMOTE-Tomek and SMOTE-ENN) on training data before applying machine learning algorithms to minimize the impact of imbalanced data. Our study sheds light on the significance of carefully utilizing data augmentation techniques without any data leakage to enhance the effectiveness of machine learning algorithms. Moreover, it offers a complete machine learning structure for healthcare practitioners, from data obtaining to machine learning prediction, enabling them to make informed decisions.

Data mining for predictive analysis in gynecology: a focus on cervical health

Currently, data mining based on the application of detection of important patterns that allow making decisions according to cervical cancer is a problem that affects women from the age of 24 years and older. For this purpose, the Rapid Miner Studio tool was used for data analysis according to age. To perform this analysis, the knowledge discovery in databases (KDD) methodology was used according to the stages that this methodology follows, such as data selection, data preparation, data mining and evaluation and interpretation. On the other hand, the comparison of methodologies such as the standard intersectoral process for data mining (Crips-dm), KDD and sample, explore, modify, model, evaluate (Semma) is shown, which is separated by dimensions and in each dimension both methodologies are compared. In that sense, a graph was created comparing algorithmic models such as naive Bayes, decision tree, and rule induction. It is concluded that the most outstanding result was -1.424 located in cluster 4 in the attribute result date.

Temporal trends in use of antisecretory agents among patients administered clopidogrel-based dual antiplatelet therapy after percutaneous coronary intervention.

Antisecretory drugs are commonly prescribed with clopidogrel-based dual antiplatelet therapy (DAPT) to prevent gastrointestinal bleeding in high-risk patients after percutaneous coronary intervention (PCI). However, omeprazole and esomeprazole (inhibiting proton pump inhibitors [PPIs]) may increase cardiovascular event rates on co-administration with clopidogrel. This study aimed to examine trends in the use of antisecretory agents in patients administered clopidogrel-based DAPT and the concomitant use of clopidogrel and inhibiting PPIs. We used National Inpatient Sample data compiled by the Health Insurance Review & Assessment Service from 2009 to 2020. Further, we identified patients who were prescribed clopidogrel-based DAPT after PCI and investigated the concomitant use of antisecretory agents with clopidogrel. To verify the annual trend of drug utilization, we used the Cochran-Armitage trend test. From 2009 to 2020, the percentage of H2 receptor antagonist users decreased steadily (from 82.5% in 2009 to 25.3% in 2020); instead, the percentage of PPI users increased (from 23.7% in 2009 to 82.0% in 2020). The use of inhibiting PPI also increased (from 4.2% in 2009 to 30.7% in 2020). Potassium competitive acid blockers (P-CABs) were rarely used before 2019; however, in 2020, it accounted for 7.8% of the antisecretory users. Our study demonstrates that the use of inhibiting PPIs increased steadily in patients administered clopidogrel-based DAPT therapy. This is a major concern since the concomitant use of inhibiting PPIs with clopidogrel could increase the risk of cardiovascular events.

Effects of the freeze–thaw process on sources and pathways of subsurface flow in an alpine hillslope on the northeastern Qinghai-Tibet Plateau

The impact of the freeze–thaw process on the active layer is reflected in the changed subsurface flow (SSF) process in cold alpine regions. Identifying sources and pathways of SSF in the freeze–thaw process is critical but difficult, and the related dominant factors and mechanisms are still unknown. In this paper, the effective identification and analysis of SSF are promoted based on field sampling data from the thawing (June) to freezing (September) period of 2022 in the Qinghai Lake basin on the northeastern Qinghai-Tibetan Plateau. By the proposed method with a high sampling frequency and refined sampling spatial scale, the sources and pathways of SSF are clearly identified. The results are as follows: (1) The soil temperature is considered the most fundamental factor affecting the SSF pathways, it influences water infiltration to the deep layer and the effect is extended to the saprolite and weathered bedrock layers. (2) Thawing promotes water to infiltrating into deep layer. 30 cm soil water contributes the most to SSF (2 %–86 %) in the thawing period, while the contribution difference of the water from the 30 cm, 60 cm, and 90 cm layers is small (ranging from 32 %–33 %, 24 %–26 %, and 32 %–35 %, respectively) in the thawed period. (3) Meanwhile, the soil water from different slope positions contribute differently to SSF, and the SSF from deep soil layer is transit in prolong paths and depths. It is caused by the out-of-sync water transit process in the hillslope. With continuing climate warming, we propose that the differences in the water sources of SSF across soil layers may decrease, while the differences in the transit processes of SSF across soil layers may increase.

Comparison of WASPAS and VIKOR methods to determine non-cash food assistance recipients

Non-cash food assistance or bantuan pangan non-tunai (BPNT) is a government program of the Republic of Indonesia by distributes food assistance in non-cash to beneficiary families. The process of distributing BPNT still needs to be done with the data and criteria set, because the existing BPNT distribution is considered not right on target. We need a method that can help provide an objective decision. One method that can be used in making decisions is the weighted aggregated sum product assessment (WASPAS) and Vlsekriterijumsko Koompromisno Rangiranje (VIKOR) methods. The results of the calculations from the two methods will then be chosen which is the best, by conducting sensitivity tests and accuracy tests. This study uses 100 sample data and 16 criteria. The sensitivity test results are 9.780678997% for the WASPAS method and -0.0759182% for the VIKOR method, while the results of the accuracy test show that both methods have the same level of accuracy, which is 80%. Based on the comparison of the sensitivity test and accuracy test of the two methods, the WASPAS method is considered more accurate in determining the recipients of the BPNT program because the WASPAS method has a higher sensitivity test value than the VIKOR method.

Reduction of rejects by combining data from the casting process and automatic X-ray inspection

Automatic inspection of castings with X-rays (radiographic and computed tomography) is widespread for parts that are relevant for safety or have high quality requirements. Examples in the automotive sector are aluminum wheels, chassis parts and new parts within the electric power train. Those parts are automatically inspected, which means that both the image acquisition and the evaluation of the images is done fully automatically. Today, in most industrial implementations, the generated data with a size up to several gigabytes per part is summarized to a simple good or bad decision, according to specification. All other data is dismissed, although this information can be valuable to optimize production processes and thus minimize rejects. This contribution gives an overview about the results of the project Cast Control, which is a collaboration of Fraunhofer Development Center for X-ray Technology EZRT, Fraunhofer Center for Applied Research on Supply Chain Services SCS and industry partner RONAL GROUP. RONAL GROUP is a major aluminum wheel manufacturer, mainly for the OEM market. Within the project we combined serial production data from the low pressure die casting process from a foundry of the RONAL GROUP with the data generated in the automatic X- ray inspection. After collecting a large base of sample data, we were able train a neural network for the prediction of error metrics obtained by X-ray inspection. We apply a combination of layer-wise relevance propagation and dimensionality reduction to find correlations between data of the casting machines (process and sensor) and the characteristics of anomalies detected during X-ray inspection. With this information, it is possible to adjust the casting process in an early stage – even before rejects are produced. This enables the foundry to reduce their rejects rate, which saves costs and energy and results in a better competitiveness.

Influence of advertisement on consumers’ buying behavior: A case study of FMCG products

The primary objective of the current study is to examine how advertising impacts the consumer behavior of University students (N = 200). This study aims to delve into the influence of advertising on attitudes related to purchasing behavior. A questionnaire was developed to assess the effect of advertising, and the sample data were gathered from college/University students aged between 20 and 30. Specifically, eight fast-moving consumer goods (FMCG) brands, namely Lux, Pears, Dettol, Dove, Head & Shoulders, L'Oreal Paris, Dove and Tresemme, were considered in the analysis. The data collection process involved evaluating the correlations between these brands and their respective advertisements. The analysis encompassed the utilization of mean, standard deviation, frequency, and percentage, which were then presented in tabular form. The findings unveiled that advertisements play a pivotal role in steering preferences and choices, significantly influencing consumer buying behavior. The model employed in this study appeared to have a more pronounced impact on consumers compared to conventional modes of advertising. Additionally, the results indicated that consumers perceive advertisements as a credible source of information in contrast to other opinions, such as those from friends, neighbors, or reference groups. The study demonstrated that advertisements wield a substantial influence across diverse income groups, regardless of the economic bracket. Interestingly, the study found that the costliness of products and the frequency of advertisement repetition did not significantly alter consumer buying behavior. Among the brands considered, DOVE emerged as the most preferred, followed by PEARS. Notably, more consumers were affected by Dove and Tresemme's advertisement than by Head & Shoulders. Due to Head & Shoulder’s competitive pricing, users favor it over other shampoo brands. The flash of advertising on the big screen drew attention to it and had an impact on consumers.

Strengthening Financial Resilience: A Holistic Approach to Combatting Fraud

Financial fraud poses a persistent threat to individuals, businesses, and the broader economy, requiring a proactive and collaborative response. This paper navigates through the intricate landscape of financial fraud, addressing its various dimensions and offering a comprehensive strategy to fortify defenses. The overarching theme revolves around the critical nexus of innovation and entrepreneurship in the financial sector. Recognizing the significance of these elements in shaping the contemporary business environment, the paper elucidates the key challenges and opportunities they present. The literature review provides a thorough examination of existing studies on innovation and entrepreneurship, identifying gaps and areas for further research. Major theories, frameworks, and models relevant to the theme are explored, supplemented by insightful case studies showcasing successful endeavors in these domains. A meticulously crafted conceptual framework integrates essential concepts related to innovation and entrepreneurship. This framework not only clarifies key terms but also informs the subsequent analysis by establishing theoretical underpinnings. The research methodology section delineates the chosen approach, whether qualitative, quantitative, or a mix of both. It justifies the selected methods, outlines data collection techniques, sampling strategies, and acknowledges any inherent limitations. Moving into the heart of the paper, the analysis and findings section presents the results of research or analysis. Visuals such as charts or graphs augment the discussion, offering a visual representation of key findings. These findings are then dissected and connected back to the conceptual framework and the literature reviewed. In the subsequent discussion section, the results are interpreted within the broader context of innovation and entrepreneurship. Practical implications for businesses, policymakers, or researchers are dissected, addressing unexpected results and limitations encountered during the study. Possible avenues for future study are also discovered. The concluding section ties together the main points, revisits the initial objectives, and emphasizes the paper’s key contributions to the field. The conclusion doesn’t merely recapitulate; it offers practical insights and recommendations derived from the collective insights gained throughout the paper. In essence, this paper provides a roadmap for fortifying defenses against financial fraud, proposing recommendations for businesses, policymakers, and regulatory bodies. By embracing collaboration, technological innovation, and prioritizing awareness, the strategies outlined aim to contribute to a more secure financial landscape.

The productivity and income effects of adopting improved rice varieties by smallholder farmers in Indonesia

PurposeThis study identified the determinants of improved rice variety adoption and measured their impact on farm productivity and the income of smallholder farmers in Indonesia.Design/methodology/approachWe used a multistage sampling procedure and data from household surveys in four districts in the Banten region of Indonesia. An endogenous switching regression (ESR) model was used to estimate the impact of adoption, and the propensity score matching (PSM) non-parametric method tested the strength of the ESR findings.FindingsThe farm productivity of adoption increased by 11.45% and the income of smallholder farmers increased by 12.10% when compared to that of traditional methods.Research limitations/implicationsThe positive and significant effects of improved rice variety adoption indicated that research programs that develop improved rice varieties can optimize the productivity and income of smallholder farmers.Originality/valueAdopting improved rice varieties increases the productivity and income of smallholder farmers, and the level of education, rice farming experience, access to extension workers, off-farm work, and mobile phone ownership have positive and significant effects on the adoption of improved rice varieties.

The effect of informal competition on the performance of formal firms: a systematic review of the literature

PurposeThe purpose of this paper is to examine the empirical literature on the effects of informal competition on firm performance and provide directions for future research.Design/methodology/approachWe search the Scopus and Web of Science databases and adopt a specific protocol to perform a systematic review of the literature. We screen 536 documents and apply three inclusion criteria for filtering.FindingsWe identify the impact of informal competition on nine types of formal firms' performance. We assess the findings according to the explanatory and response variables and the data samples of the articles reviewed. We find that five areas of focus are under-researched due to analyses of country or region-specific data, or lack of consensus because of contradicting findings.Originality/valueTo the best of our knowledge, this is the first review of the empirical literature on the effects of informal competition on the performance of formal firms.

Artificial intelligence and radiomics in the diagnosis of intraosseous lesions of the gnathic bones: A systematic review.

The purpose of this systematic review (SR) is to gather evidence on the use of machine learning (ML) models in the diagnosis of intraosseous lesions in gnathic bones and to analyze the reliability, impact, and usefulness of such models. This SR was performed in accordance with the PRISMA 2022 guidelines and was registered in the PROSPERO database (CRD42022379298). The acronym PICOS was used to structure the inquiry-focused review question "Is Artificial Intelligence reliable for the diagnosis of intraosseous lesions in gnathic bones?" The literature search was conducted in various electronic databases, including PubMed, Embase, Scopus, Cochrane Library, Web of Science, Lilacs, IEEE Xplore, and Gray Literature (Google Scholar and ProQuest). Risk of bias assessment was performed using PROBAST, and the results were synthesized by considering the task and sampling strategy of the dataset. Twenty-six studies were included (21 146 radiographic images). Ameloblastomas, odontogenic keratocysts, dentigerous cysts, and periapical cysts were the most frequently investigated lesions. According to TRIPOD, most studies were classified as type 2 (randomly divided). The F1 score was presented in only 13 studies, which provided the metrics for 20 trials, with a mean of 0.71 (±0.25). There is no conclusive evidence to support the usefulness of ML-based models in the detection, segmentation, and classification of intraosseous lesions in gnathic bones for routine clinical application. The lack of detail about data sampling, the lack of a comprehensive set of metrics for training and validation, and the absence of external testing limit experiments and hinder proper evaluation of model performance.

Family firms and the collaborative advantage: unveiling innovation efficiency across partnership types

PurposeBased on the resource orchestration perspective, this paper aims to examine whether family firms are more efficient in their collaboration for innovation process than non-family firms, considering different types of collaboration for innovation depending on the kind of partner.Design/methodology/approachThis study empirically develops and tests the hypotheses based on a panel data sample of 14,937 firm-year observations from 1,867 Spanish manufacturing firms over the period 2007–2014, performing a Propensity Score Matching (Propensity score matching)-based analysis.FindingsResults reveal that family firms outperform non-family firms, despite less collaboration and innovation inputs, thereby extending the ongoing debate surrounding the innovation efficiency of family firms. Family firms obtained better results through vertical collaborations for innovation, both in terms of product and process innovations. For horizontal collaborations, family firms only outperform their non-family counterparts in process innovation. When collaborating with universities and other research centers, there are no significant differences in the innovation outcomes between the two groups.Originality/valueRecent literature points out that more research is needed to know when, how and under what circumstances family firms show superior innovative efficiency. This work empirically proves that family firms outperform non-family firms in collaboration for innovation. However, not all collaboration partners help family firms to reach this superior innovative efficiency. Family firms obtained better results just through vertical and horizontal collaborations.

Learning to Diagnose: Meta-Learning for Efficient Adaptation in Few-Shot AIOps Scenarios

With the advancement of technologies like 5G, cloud computing, and microservices, the complexity of network management systems and the variety of technical components have greatly increased. This rise in complexity has rendered traditional operations and maintenance methods inadequate for current monitoring and maintenance demands. Consequently, artificial intelligence for IT operations (AIOps), which harnesses AI and big data technologies, has emerged as a solution. AIOps plays a crucial role in enhancing service quality and customer satisfaction, boosting engineering productivity, and reducing operational costs. This article delves into the primary tasks involved in AIOps, such as anomaly detection, and log fault analysis and classification. A significant challenge identified in many AIOps tasks is the scarcity of fault sample data, indicating a natural alignment of these tasks with few-shot learning. Inspired by model-agnostic meta-learning (MAML), we propose a new anomaly detector, MAML-KAD, for application in various AIOps tasks. Observations confirm that meta-learning algorithms effectively enhance AIOps tasks, showcasing the wide-ranging application prospects of meta-learning algorithms in the field of AIOps. Moreover, we introduced an AIOps platform that embeds meta-learning within its diagnostic core and features streamlined log collection, caching, and alerting to automate the AIOps workflow.

Determining Extracorporeal Membrane Oxygenation Outcomes in Obese Patients with ARDS

Extracorporeal membrane oxygenation (ECMO) is crucial for managing acute respiratory distress syndrome (ARDS) refractory to conventional therapy, but its impact on obese ARDS patients remains inadequately understood due to relative contraindications per the Extracorporeal Life Support Organization (ELSO) guidelines. This study aimed to analyze the association between obesity and ECMO outcomes in ARDS patients. Using five years of National Inpatient Sample data (2016–2020), we identified ARDS patients through ICD-10 codes, classifying those with a BMI > 30 as obese. Univariate and multivariate analyses were performed to assess mortality, rates of left ventricular assist devices, and tracheostomy differences between obese and non-obese patients, with secondary outcomes examining the impact of obesity on length of stay (LOS) and total hospitalization charges (TOTCHG). Among 3,219 ARDS patients requiring ECMO, 8.2% were obese. While initial univariate analysis showed a non-significant 43% lower mortality odd in obese patients, adjusting for confounders revealed a statistically significant 46% lower odd of mortality compared to non-obese patients (adjusted odds ratio 0.54, 95% confidence interval 0.3-0.94, p 0.032). LOS and TOTCHG did not significantly differ between the two groups, nor did the rates of LVAD or tracheostomy. These findings suggest that obesity may confer a survival advantage in ARDS patients undergoing ECMO, potentially due to factors such as nutritional reserves and the obesity paradox observed in critical illnesses. Therefore, BMI alone should not preclude obese patients from ECMO treatment, emphasizing the need for further research to inform clinical decision-making in this population.

Direct interaction in digital interactive media and stock performance: Evidence from Panorama.

Media information plays an essential role in the stock market. Recent financial research has verified that media information could shock stock price by influencing investors' expectation. Now, a new type of interactive media, called Digital Interactive Media (DIM), is popular in Chinese stock market and becomes the main channel for investors to understand listed companies. Unlike general news media or investor forums, DIM enables direct interaction between listed companies and investors. In the modern society where digital economy is booming, media information would largely affect investors' decisions. Therefore, it is urgent to use natural language processing (NLP) technology to deconstruct the massive questions and answers (Q&A) interactive information in DIM and extract valuable factors that affect stock prices and stock performances to explore the influence mechanism of digital interactive information on stock performances. This paper firstly uses web crawling technology to obtain approximately 110000 Q&A text information from the digital interactive platform ('Panoramic Network') from 2015 to 2021. Then we use big data text analysis technology and emotional quantification technology to extract valuable influencing factors from the massive text. A Multiple Linear Regression (MLR) model was created to explore specific influence mechanism of digital interactive information on stock price performance. The empirical results show that the emotions implicit in investors' questions do not significantly impact stock performance. However, the emotions and attitudes of the answers by listed companies can significantly affect corresponding stock prices, which indirectly confirms the Proximate Cause Effect of behavioral finance. This effect is particularly evident in the stock prices on the current trading day and the next trading day. In the Robustness Test, this paper replaces dependent variable and adds relevant control variables, and the conclusion remains valid. In the Endogeneity Test, this paper selects sample data before the launch of Panorama Network in 2014 as a comparison, and uses a Difference-in-Difference (DID) model to prove the significant impact of the launch of Panorama Network on Chinese stock market. In the Heterogeneity Test, the paper classifies the market value, region, and industry of listed companies and regressed the sub samples, once again confirming the reliability of the empirical conclusions. The results of Robustness Test, Endogeneity Test, and Heterogeneity Test conducted in this paper all support empirical conclusions.

Earning Moderates the Influence of CAR, NPL, and BOPO on Company Value in Conventional Banks Listed on the IDX for the 2020-2022 Period

The goal of research on traditional commercial banks listed on the Indonesia Stock Exchange (IDX) is to use earnings as a moderator to ascertain the impact of the capital adequacy ratio (CAR), non-performing loan (NPL), and BOPO on the value of the company. Variables This study uses a quantitative approach and sample data collection through purposive sampling. The samples obtained for this study are 69 samples that meet the standards of the Indonesia Stock Exchange (IDX) and financial statements for the 2020-2022 period. Classical hypothesis testing and moderate regression analysis (MRA) methods were used to analyze the results of this study. The results showed that the Capital Adequacy Ratio (CAR) did not have a significant effect on the value of the company, and non-performing loans (NPL) did not have a significant effect on the value of the company and operating costs. (BOPO) does not have a significant influence on the value of the company. Capital Adequacy Ratio (CAR) can be changed and affect company value, profit cannot moderate the impact of non-performing loans (NPL) on company performance Product value cannot reduce Operating Costs and Operating Income (BOPO) to company value.

A novel semi-supervised learning rolling bearing fault diagnosis method based on SNNGAN

In practical industrial environments, rotating machinery typically operates under normal conditions. As a result, the signals collected are primarily normal signals. This imbalance in the sample data diminishes the effectiveness of fault diagnosis. To address this issue, this paper produces a novel semi-supervised fault diagnosis approach based on a Siamese neural network combined with a generative adversarial network (SNNGAN) to enhance classification accuracy. Firstly, vibration signals collected are subjected to continuous wavelet transformation to obtain time–frequency representations, which are utilized for pre-training convolutional encoders in the generator and discriminator. Subsequently, a cosine similarity algorithm is employed to ensure the quality of generated samples. For generated data, set a similarity threshold. Those surpassing the threshold are assigned their corresponding labels and added to the original sample set. Otherwise, those falling below the threshold are transformed back into vibration vectors through an inverse transform and then serve as input to create new samples. Finally, fault diagnosis experiments are conducted on the newly balanced data set. In four imbalanced data experiments, the results demonstrate that SNNGAN outperforms other methods in average accuracy, G-mean, and F1 score, with average accuracy values of 0.919, 0.948, 0.927, and 0.953 for the respective datasets. Therefore, SNNGAN exhibits outstanding fault diagnosis performance under conditions of data imbalance.

Association of CYP2D6 and CYP2C19 metabolizer status with switching and discontinuing antidepressant drugs: an exploratory study

BackgroundTailoring antidepressant drugs (AD) to patients’ genetic drug-metabolism profile is promising. However, literature regarding associations of ADs’ treatment effect and/or side effects with drug metabolizing genes CYP2D6 and CYP2C19 has yielded inconsistent results. Therefore, our aim was to longitudinally investigate associations between CYP2D6 (poor, intermediate, and normal) and CYP2C19 (poor, intermediate, normal, and ultrarapid) metabolizer-status, and switching/discontinuing of ADs. Next, we investigated whether the number of perceived side effects differed between metabolizer statuses.MethodsData came from the multi-site naturalistic longitudinal cohort Netherlands Study of Depression and Anxiety (NESDA). We selected depression- and/or anxiety patients, who used AD at some point in the course of the 9 years follow-up period (n = 928). Medication use was followed to assess patterns of AD switching/discontinuation over time. CYP2D6 and CYP2C19 alleles were derived using genome-wide data of the NESDA samples and haplotype data from the PharmGKB database. Logistic regression analyses were conducted to investigate the association of metabolizer status with switching/discontinuing ADs. Mann–Whitney U-tests were conducted to compare the number of patient-perceived side effects between metabolizer statuses.ResultsNo significant associations were observed of CYP metabolizer status with switching/discontinuing ADs, nor with the number of perceived side effects.ConclusionsWe found no evidence for associations between CYP metabolizer statuses and switching/discontinuing AD, nor with side effects of ADs, suggesting that metabolizer status only plays a limited role in switching/discontinuing ADs. Additional studies with larger numbers of PM and UM patients are needed to further determine the potential added value of pharmacogenetics to guide pharmacotherapy.

Rapid heat source layout optimization in three‐dimensional integrated circuits using artificial neural network reduced‐order model in combination with Bayesian optimization

AbstractIn this study, an efficient optimization framework was developed to determine the parameters of through‐silicon vias and the layout of heat sources in three‐dimensional integrated circuits (3D ICs), employing an artificial neural network (ANN) reduced‐order model in conjunction with a Bayesian optimization (BO) algorithm. The proposed method effectively predicts the temperature distribution in 3D ICs and refines their thermal parameters, offering solutions to thermal management challenges. Latin hypercube sampling was utilized for data sampling, enhancing the previously established rapid thermal analysis method through parameterization of heat source locations. The temperature distribution data for varying hotspot locations in 3D ICs were fitted using an appropriately defined objective function, leading to the development of a reduced‐order ANN model that accelerates temperature prediction. The computational results demonstrate that the neural network model exhibits a deviation in predicted values of less than 2%, and the coefficient of determination R2 approximately 0.93, underscoring high predictive accuracy. Additionally, the optimization outcomes and the efficiency of the selected BO algorithm were thoroughly evaluated. Notably, the BO algorithm achieved the global optimum in just 4.07 s across 250 iterations, demonstrating an effective power distribution strategy for the 3D ICs model.