Explore the vast range of titles available.

With the advancement in machine learning, researchers continue to devise and implement effective intelligent methods for fraud detection in the financial sector. Indeed, credit card fraud leads to billions of dollars in losses for merchants every year. In this paper, a multi-classifier framework is designed to address the challenges of credit card fraud detections. An ensemble model with multiple machine learning classification algorithms is designed, in which the Behavior-Knowledge Space (BKS) is leveraged to combine the predictions from multiple classifiers. To ascertain the effectiveness of the developed ensemble model, publicly available data sets as well as real financial records are employed for performance evaluations. Through statistical tests, the results positively indicate the effectiveness of the developed model as compared with the commonly used majority voting method for combination of predictions from multiple classifiers in tackling noisy data classification as well as credit card fraud detection problems.

Cybersecurity has become one of the focuses of organisations. The number of cyberattacks keeps increasing as Internet usage continues to grow. As new types of cyberattacks continue to emerge, researchers focus on developing machine learning (ML)-based intrusion detection systems (IDS) to detect zero-day attacks. They usually remove some or all attack samples from the training dataset and only include them in the testing dataset when evaluating the performance. This method may detect unknown attacks; however, it does not reflect the long-term performance of the IDS as it only shows the changes in the type of attacks. In this work, we focused on evaluating the long-term performance of ML-based IDS. To achieve this goal, we proposed evaluating the ML-based IDS using a dataset created later than the training dataset. The proposed method can better assess the long-term performance as the testing dataset reflects the changes in the attack type and network infrastructure changes over time. We have implemented six of the most popular ML models, including decision tree (DT), random forest (RF), support vector machine (SVM), naïve Bayes (NB), artificial neural network (ANN), and deep neural network (DNN). These models are trained and tested with a pair of datasets with symmetrical classes. Our experiments using the CIC-IDS2017 and the CSE-CIC-IDS2018 datasets show that SVM and ANN are most resistant to overfitting. Our experiments also indicate that DT and RF suffer the most from overfitting, although they perform well on the training dataset. On the other hand, our experiments using the LUFlow dataset have shown that all models can perform well when the difference between the training and testing datasets is small.

The present research study aims to produce unsaturated polyester composite made of denim fabric and unsaturated polyester by using three different fabrication methods, such as compression moulding, resin vacuum infusion and hand lay-up techniques. The average unsaturated polyester-to-composite weight ratio was strongly affecting the mechanical properties of the unsaturated polyester composite. The hand lay-up technique showed the highest unsaturated polyester-to-composite weight ratio at 70.63%, 55% in resin vacuum infusion technique, and 51.62% in compression moulding. The results showed that the higher the unsaturated polyester-to-composite weight ratio, the lower the mechanical properties. Nonetheless, the overall mechanical properties of denim fabric-reinforced unsaturated polyester composite were enhanced in comparison with the neat unsaturated polyester. In details, among all the composites samples, the compression moulded composite exhibited maximum tensile strength, modulus of elasticity, impact strength, flexural strength, and flexural modulus whereas resin vacuum infusion composite showed the highest elongation at break and average percentage in water absorption test. Moreover, the hand lay-up composites were found to have the lowest values in majority of mechanical properties, mainly due to their high unsaturated polyester weight ratio and poor interfacial interaction between denim and unsaturated polyester. The hand lay-up composite absorbed the least water molecules as compared to another two unsaturated polyester composites due to the highest unsaturated polyester-to-composite weight ratio as compared to two other fabrication methods. The morphological study has been investigated and the results in line with the mechanical properties of the unsaturated polyester composites.

The combination and aggregation of knowledge from multiple neural networks can be commonly seen in the form of mixtures of experts. However, such combinations are usually done using networks trained on the same tasks, with little mention of the combination of heterogeneous pre-trained networks, especially in the data-free regime. The problem of combining pre-trained models in the absence of relevant datasets is likely to become increasingly important, as machine learning continues to dominate the AI landscape, and the number of useful but specialized models explodes. This paper proposes multiple data-free methods for the combination of heterogeneous neural networks, ranging from the utilization of simple output logit statistics, to training specialized gating networks. The gating networks decide whether specific inputs belong to specific networks based on the nature of the expert activations generated. The experiments revealed that the gating networks, including the universal gating approach, constituted the most accurate approach, and therefore represent a pragmatic step towards applications with heterogeneous mixtures of experts in a data-free regime. The code for this project is hosted on github at https://github.com/cwkang1998/network-merging.

PurposeTo establish the 12-month treatment gap and its associated factors among adults with mental disorders in the Singapore resident population using data from the second Singapore Mental Health Study and to examine the changes since the last mental health survey conducted in 2010.Methods6126 respondents were administered selected modules of the Composite International Diagnostic Interview, to assess major depressive disorder (MDD), dysthymia, bipolar disorder, generalized anxiety disorder (GAD), obsessive compulsive disorder (OCD) and alcohol use disorder (AUD) (which included alcohol abuse and dependence). Past year treatment gap was defined as the absolute difference between the prevalence of a particular mental disorder in the past 12 months preceding the interview and those who had received treatment for that disorder.ResultsThe prevalence of overall 12-month treatment gap in this population was high (78.6%). A multiple logistic regression analysis revealed significantly higher odds of treatment gap among those diagnosed with OCD (compared to those with MDD) and in those with a comorbid chronic physical disorder; while those who had primary education and below and those who were unemployed were less likely to have a treatment gap as compared to those with post-secondary education and those employed, respectively.ConclusionsThe high treatment gap in the population is concerning and highlights the need to promote help-seeking and uptake of treatment. Given the unique demographic characteristics, i.e., those with higher education and employed were more likely not to seek treatment, targeted interventions in the educational and workplace settings should be implemented.

Background Multimorbidity is not uncommon and the associated impact it places on healthcare utilisation and societal costs is of increased concern. The aim of the current study was to estimate the economic burden of multimorbidity among older adults in Singapore by investigating its association with the healthcare and societal resource use and cost. Methods The Well-being of the Singapore Elderly (WiSE) study was a single phase, cross sectional survey among a nationally representative sample of Singapore residents ( N  = 2565) aged 60 years and above. Multimorbidity was defined in this study as having two or more chronic conditions, from a list of 10 conditions. Care was classified into healthcare which included direct medical care, intermediate and long-term care, indirect care, and social care, provided by paid caregivers and family members or friends. Costs were calculated from the societal perspective, including healthcare and social care costs, by multiplying each service unit with the relevant unit cost. Generalized linear models were used to investigate the relationship between total annual costs and various socio-demographic factors. Results The prevalence of multimorbidity was 51.5 %. Multimorbid respondents utilised more healthcare and social care resources than those with one or no chronic conditions. The total societal cost of multimorbidity equated to SGD$15,148 per person, annually, while for those with one or no chronic conditions the total annual societal costs per person were SGD$5,610 and SGD$2,806, respectively. Each additional chronic condition was associated with increased healthcare (SGD$2,265) and social care costs (SGD$3,177). Older age (i.e. 75–84 years old, and especially over 85 years), Indian ethnicity and being retired were significantly associated with higher total costs from the societal perspective, while older age (75 years and above) and ‘Other’ ethnicity were significantly associated with higher total healthcare costs. Conclusion Multimorbidity was associated with substantially higher healthcare utilisation and social care costs among older adults in Singapore. With the prevalence of multimorbidity increasing, especially as the population ages, we need healthcare systems that are evolving to address the emerging challenges associated with multimorbidity and the respective healthcare and societal costs.

Objective The objective of this study was to evaluate the efficacy and long-term outcomes of the use of Restorelle® Direct Fix (Coloplast, Humlebæk, Denmark) anterior mesh for transvaginal surgical management of anterior compartment prolapse. Methods A retrospective case series review was conducted for 123 patients who underwent surgery for Baden-Walker Grade three and four anterior compartment prolapse with the Restorelle Direct Fix anterior mesh between July 1, 2017 and September 30, 2018 in a single center. Follow-up was conducted at one, six, 12, 24, and 36 months after treatment. A standardized questionnaire and pelvic examination were conducted at each visit to assess operative complications and subjective and objective cure rates. Results Sixty patients were included in the analysis with a three-year follow-up rate of 70.0%. At three years post-operatively, subjective and objective cure rates were 97.7% and 95.3% respectively. Seven (11.7%) patients complained of de novo stress urinary incontinence, four (6.7%) complained of de novo urge urinary incontinence and one (1.7%) complained of symptomatic recurrence. Significantly, six (10.0%) patients had transvaginal mesh exposure over the three-year follow-up, mostly presenting within the first year. One (2.4%) patient developed new asymptomatic mesh erosion at the 36-month visit and one patient required mesh loosening one month post-surgery. Conclusions Management of anterior compartment prolapse with transvaginal surgery using the Restorelle® Direct Fix anterior mesh was associated with good subjective and objective cure rates. However, significant rates of post-operative mesh exposure were noted within three years post-surgery, which hinders the recommendation of this device for augmentation of repair for anterior compartment prolapse.

In recent years, the incidence of penicillin-susceptible S. aureus (PSSA) bloodstream infection (BSI) has increased worldwide. However, the preferred antibiotic remains uncertain due to concerns of inducible resistance to benzylpenicillin. We compared outcomes associated with benzylpenicillin versus other antibiotics and investigated risk factors influencing treatment failure. Patients were grouped into benzylpenicillin and non-benzylpenicillin beta-lactam treatment groups (including anti-staphylococcal penicillins and cephalosporins). The primary outcome was overall treatment failure (30-day all-cause mortality and/or 90-day relapse). Of 335 patients, 74 (22.09%) received benzylpenicillin and 261 (77.91%) received a non-benzylpenicillin beta-lactam. While rates of overall treatment failure (13.51% vs. 17.24%; P  = 0.45) and occurrence of adverse drug events (6.76% vs. 7.66%; P  = 0.79) were comparable to non-benzylpenicillin beta-lactams, benzylpenicillin showed faster microbiological clearance [3.00 days (IQR, 2.00–4.00 days) vs. 4.00 days (IQR, 3.00–5.00 days); P  = 0.03] and fewer persistent infections (22.97% vs. 36.02%; P  = 0.04), suggesting potential to improve patient outcomes. We also found that unknown source (aOR 4.63, 95% CI 1.47–14.64; P  < 0.01) was associated with treatment failure, while review by Infectious Disease (ID) specialists (aOR 0.30, 95% CI 0.12–0.73; P  = 0.01) was protective, stressing the importance of early ID referral and thorough source identification. This study highlights benzylpenicillin as an effective treatment for PSSA BSI.

In this study, cabbage [Brassica rapa L. subsp. chinensis (L.) Hanelt cv. Xinza No 1], mung bean [Vigna radiata (L.) R. Wilczek var. radiata cv. VC‐3762], and wheat (Triticum aestivum L. cv. Altas 66) were grown in Pb‐contaminated soils. Application of ethylenediaminetetraacetic acid (EDTA) (3.0 mmol of EDTA/kg soil) to the soil significantly increased the concentrations of Pb in the shoots and roots of all the plants. Lead concentrations in the cabbage shoots reached 5010 and 4620 mg/kg dry matter on Days 7 and 14 after EDTA application, respectively. EDTA was the best in solubilizing soil‐bound Pb and enhancing Pb accumulation in the cabbage shoots among various chelates (EDTA, diethylenetriaminepentaacetic acid [DTPA], hydroxyethylenediaminetriacetic acid [HEDTA], nitrilotriacetic acid [NTA], and citric acid). Results of the sequential chemical extraction of soil samples showed that the Pb concentrations in the carbonate–specifically adsorbed and Fe–Mn oxide phases were significantly decreased after EDTA treatment. The results indicated that EDTA solubilized Pb mainly from these two phases in the soil. The relative efficiency of EDTA enhancing Pb accumulation in shoots (defined as the ratio of shoot Pb concentration to EDTA concentration applied) was highest when 1.5 or 3.0 mmol EDTA/kg soil was used. Application of EDTA in three separate doses was most effective in enhancing the accumulation of Pb in cabbage shoots and decreased mobility of Pb in soil compared with one‐ and two‐dose application methods. This approach could help to minimize the amount of chelate applied in the field and to reduce the potential risk of soluble Pb movement into ground water.

Background: Understanding the lower level of happiness among older adults with cognitive impairment has been a largely neglected issue. This study (1) reports on the level of happiness among older adults in Singapore and (2) examines the potential mediating roles of depression, disability, social contact frequency, and loneliness in the relationship between cognitive scores and happiness. Methods: Data for this study were extracted from the Well-being of the Singapore Elderly (WiSE) study: a cross-sectional; comprehensive single-phase survey conducted among Singapore citizens and permanent residents that were aged 60 years and above (n = 2565). The Geriatric Mental State examination (GMS) was administered to the participants. Questions pertaining to socio-demographic characteristics; happiness; loneliness; social contact; depression; and, disability were utilized in this study. Logistic regression analyses and mediation analyses were used to explore the correlates of happiness and potential mediating factors. Results: Overall, 96.2% of older adults in Singapore reported feeling either fairly happy or very happy. In the regression analysis, individuals of Malay descent, those who were married/cohabiting, or had higher education levels were more likely to report feeling happy. After controlling for socio-demographic factors, higher cognitive scores were associated with higher odds of reporting happiness. We found that the positive association between cognition and happiness was fully mediated by disability, depression, loneliness, and frequency of contact with friends. Conclusion: The majority of the older adult population reported feeling fairly or very happy. While cognitive impairment has shown limited reversibility in past studies, unhappiness among older adults with cognitive impairment might be potentially mitigated through interventions addressing accompanying issues of social isolation, disability, and depression