This paper examines the distribution of eigenvalues for a $2\times 2$ random confusion matrix used in machine learning evaluation. We also analyze the distributions of the matrix’s trace and the difference between the traces of random confusion matrices. Furthermore, we demonstrate how these distributions can be applied to calculate the superiority probability of machine learning models. By way of example, we use the superiority probability to compare the accuracy of four disease outcomes machine learning prediction tasks.

The article focuses on Rayleigh wave propagation in a homogeneous isotropic semi-conductor thermoelastic medium rotating with fixed angular frequency with temperature-dependent properties. The elastic constants depend upon the temperature function. The effects of temperature dependency parameter, time derivative, and fractional order are illustrated. By the theory of thermo-elasticity, waves result in the generation of thermal signals that propagate through the medium. A heat conduction model of three-phase lag (3PL) along with fractional order time derivative is used to analyze the thermal signals. The secular equations of Rayleigh waves are derived mathematically at the stress-free, carrier density and thermally insulated boundaries. Some specific properties like velocity, attenuation coefficient, specific heat loss and penetration depth for Rayleigh waves have been evaluated and presented graphically. The secular equations are computed numerically and depicted graphically using Matlab.

Background: Although the frequency of surgical weight loss interventions has increased in Saudi Arabia, literature describing the clinical outcomes of bariatric surgery in Saudi Arabia is limited. This study aimed to assess whether weight loss intervention improves hemoglobin A1c (HbA1c) in obese patients and to identify its associated factors. Patients and methods: A retrospective study was carried out on 318 patients with obesity class 1 or higher (body mass index [BMI] ≥ 30 kg/m²) who underwent laparoscopic sleeve gastrectomy or Roux-en-Y gastric bypass at King Abdulaziz Medical City in Riyadh, Saudi Arabia, between January 1, 2001 and March 31, 2017. Preoperatively and 12 months postoperatively, characteristics of patients were collected. BMI reduction was calculated, based on which patients were divided into three groups (0–9, 10–14, and >14 kg/m²). Results: The postoperative HbA1c was 5.83±0.9, while the baseline level was 6.74±2.1 (P=0.001). Fifty-eight of the 318 patients had diabetes. We observed significantly higher HbA1c in diabetic than in non-diabetic patients preoperatively, whereas an insignificantly ­different HbA1c was observed postoperatively. Among those who had minimal reduction in BMI (0–9 kg/m²), we observed significantly higher HbA1c in diabetic than in non-diabetic patients, whereas among those who had large reduction in BMI (10–14 kg/m²) and (>14 kg/m²), we observed insignificant differences in HbA1c in diabetic than in non-diabetic patients. Conclusion: Being a diabetic patient was related to a significant reduction in HbA1c levels postoperatively. The study suggests that the reduction in HbA1c levels could be modified by BMI, wherein greater reduction in BMI leads to greater reduction in HbA1c levels.

Random forest (RF) is a widely used data prediction and variable selection technique. However, the variable selection aspect of RF can become unreliable when there are more irrelevant variables than relevant ones. In response, we introduced the Bayesian random forest (BRF) method, specifically designed for high-dimensional datasets with a sparse covariate structure. Our research demonstrates that BRF possesses the oracle property, which means it achieves strong selection consistency without compromising the efficiency or bias.

In this paper, we derive the distribution of eigenvalues for a 2×2 random confusion matrix stemming from a machine learning (ML) evaluation problem. Additionally, we present the distributions of both the matrix’s trace and the difference between the two traces of random confusion matrices. We also illustrate the application of these derived distributions in calculating the superiority probability of four baseline ML models.