Search

Using Innovative Cluster Mining Algorithms, the objective is to implement a cloud-based framework for the medical system in a semi-rural health administration dataset. This framework will be compared to the Real Time Data Stream Algorithm. Innovative Cluster Mining is a method that is based on machine learning and consists of a set of data points that are grouped together into clusters in such a way that all of the objects represent the same group. A number of subsets are created from the data using this innovative clustering technique. These subsets are made up of datasets that are related to one another and are referred to as clusters. The Components and Procedures: The dataset that is required for the Cloud-Based Framework that is used for the health administration system is obtained from the Kaggle website that is operated by Google. There are columns in the data set that contain the following information: patient name, age, gender, blood group, and medical reason. in order to conduct an analysis, design, and implementation of the infrastructures of the cloud framework and applications for cloud computing. In semi-rural regions, the data sets are imported, and innovative cluster mining algorithms and real-time data stream algorithms are put through their paces under testing conditions. With two different methods, there are a total of two groups. Each group has a sample size of eighty people. The correctness of the inputs that were supplied is determined by the outcomes, which are obtained in this manner. It is via the utilisation of the IBM SPSS application that the findings are obtained. Based on the findings, it has been demonstrated that the Innovative Cluster Mining Algorithms exhibit a higher level of accuracy than the Real Time Data Stream Algorithm, which has a rate of 75.71 percent, with a significance level of p = 0.01 (<0.05, 2-tailed). This indicates that the algorithm is more accurate than the value. For the purpose of determining the mean, standard deviation, and standard error, the independent sample T-test was successfully carried out. Consider the statistical significance of the mean between the groups. Conclusion: According to the findings presented in this article, the Innovative Cluster Mining Algorithm is superior than the Real Time Data Stream Algorithm in terms of its level of precision. [ABSTRACT FROM AUTHOR]

The objective of this research is to develop a sophisticated method for predicting bankruptcy that makes use of support vector machines (SVM) rather than random forests and produces more accurate results. The Constituents and the Methods involved: In the case of group 2, which is sometimes referred to as Random Forest, the sample size includes twenty individuals. Support vector machines are utilised in order to facilitate the classification process (SVM). A check with one tail is carried out with a Confidential Interval (CI) of 95, and the price of importance is set at beta 1.00. In the support vector machine (SVM), the alpha price is believed to be 0.8, and the G-Power price is likewise considered to be 0.8. The results showed that the SVM algorithm acquired an accuracy rate of 97.3 percent, while the Random Forest community Algorithm achieved a recognition rate of 98.8 percent. This was determined after all of the assessments that were carried out inside the research were conducted and completed. As a result of the fact that the mean accuracy detection is 1.483 (±2SD) and the significance value is 0.00 (p<0.05), it is possible to take note of the fact that there is a difference between the two groups that is statistically significant. It is not even close to being a T-test for independent samples, despite the fact that the speculation is accurate. In reality, it is something quite different. In contrast to the Random Forest, the support vector machine (SVM) had a lesser degree of accuracy, which was 95.3 percent, according to the findings of the research that was conducted on bankruptcy (dt). The proposed studies of SVM, which produces lower accuracy than the RF and makes it easier for customers to accurately, are the means by which the loss of accuracy that is occurring within the current research is progressing. This loss of accuracy is occurring as a result of the advancement of the research. [ABSTRACT FROM AUTHOR]

Recently, numerous normalization layers for the purpose of stabilizing the training of deep neural networks (DNN) have been developed. Group normalization is one such technique that expands upon instance normalization and layer normalization by permitting some flexibility about the quantity of groups it employs. However, in order to ascertain the most effective number of groups, it is necessary to conduct time-consuming studies involving trial-and-error hyperparameter adjustment. For this study, we lay out a method that is both practical and effective for deciding how many groups to use. The initial observation is that the group normalization layer's gradient behavior is affected by the number of groups. By deducing the optimal group size, we may calibrate the gradient scale for optimization using gradient descent. For the first time, this research suggests a maximum group size that accounts for theoretical underpinnings, architectural concerns, and the capacity to independently produce adequate value for each layer. All sorts of neural network topologies, tasks, and datasets showed that the proposal method outperformed the state-of-the-art procedures. [ABSTRACT FROM AUTHOR]

In order to evaluate the sonolytic breakdown of sodium laureth sulphate under standard conditions in a synthetic aqueous medium to determine its effectiveness. Detailed instructions and materials: Recent efforts have been made to prepare and analyse a stock solution of sodium laureth sulphate for the purpose of determining the ideal temperature and pH range conditions. We compiled a record of the findings and evaluated them. We employed a starting concentration of 100 mg/L for the sample solution, and we tested a pH range that went from 4 to 7.9 during the experiment. An ultraviolet-visible spectrophotometer was utilised in order to determine the absorbance of the solution at a wavelength of 520 nanometers. A total of five samples were collected for each pH solution, and these samples were then split into two groups. A one-way analysis of variance (ANOVA) test was carried out with IBM SPSS Version 26, which allowed for a numerical comparison of the samples. There were a total of 10 people who took part in the study, and the confidence interval was set to 95 percent, the G-power was set to 80 percent, and the enrollment ratio was set to 1. Furthermore, the results of the demineralization that were achieved by the use of sonolysis were encouraging. An increase in the rate of degradation of sodium laureth sulphate occurs when the temperature is 70 degrees Celsius and the pH is 7.9. In order to assess the mean-variance between Group 2 and Group 1, a one-way analysis of variance (ANOVA) test was employed. The results of the test indicated a statistically significant value of p=0.004 (p<0.05) for statistical observation. From the findings, it was determined that sonolysis has the potential to degrade sodium laureth sulphate. It had been demonstrated that the sonolysis method, which is responsible for the destruction of dangerous chemicals in wastewater treatment, is both effective and simple to implement. [ABSTRACT FROM AUTHOR]

The purpose of this study is to investigate the hazardous properties of a new tartrazine dye and to compare its decolorization under various pH circumstances using sonolysis. Approach and methodology: Results were obtained at their greatest yields after ultrasonication was carried out for 2 hours at the appropriate temperatures in a tartrazine dye solution that had been produced under conditions of pH 4 and 8. We used IBM SPSS V 28.0 software to conduct a numerical analysis with a confidence interval, G-power of 80%, enrollment ratio of one (N=12), and an alpha error level of 0.05. For each pH solution, we divided the participants into two groups, each with six samples. We then used T-tests to compare the samples. As a result, the optimal temperature for a sonicated tartrazine dye solution was determined to be 45°C under pH 4 circumstances and 70°C under pH 8 conditions. The mean-variance under the significance level of 0.001 with a significant p<0.05 was used to conduct an independence-sample T-Test between Group 2 and Group 1, and the results were obtained. The results show that tartrazine solution may be employed in wastewater treatment operations after sonication breaks down its complicated structure under specific circumstances, within the restrictions of this study. [ABSTRACT FROM AUTHOR]

We are going to investigate the sonolytic breakdown of butylated hydroxyanisole and its effectiveness against cetyl alcohol by making use of a synthetic aqueous sample solution and conducting the experiment under well controlled circumstances. Detailed instructions and materials: A study was conducted in which the researchers investigated the effects of time, temperature, and pH on the butylated hydroxyanisole solution. At first, the sample solution had a concentration of one hundred milligrammes per litre and a pH that ranged from four and a half to eight and thirty. All of the changes in value were recorded at each stage of the process. In order to determine the absorbance of the solution, a UV-Vis spectrophotometer operated at 550 nm was utilised for the purpose of conducting the analysis. A numerical technique of analysis was utilised, and the confidence interval for each pH solution was set at 95 percent, the G-power was set at 80 percent, the enrollment ratio was set at one (N=10), and an alpha error level was set for each concentration. The analysis of each solution was performed on five samples that were collected. A one-way analysis of variance (ANOVA) test was used to compare the samples, and the software IBM SPSS Version 26 was used to carry out the analysis. In order to get effective results, the sonolytic demineralization approach was utilised. Similarly, an increase in pH results in an improvement in the efficiency of degradation (8.5). The bulk of the sonolytic breakdown of butylated hydroxyanisole was sped up by increasing the temperature of the sample solution to 90 degrees Celsius. Following the statistical analysis of the mean and multiple comparisons between time, temperature, and pH, a one-way analysis of variance (ANOVA) test was employed to assess the findings of a significant value of p=0.01 (p<0.05). Conclusion: Using the sonolysis technique, it was determined that the chemical breakdown of BHA takes place in a solution that has a pH of 8.5 and takes place at a temperature of 90 degrees Celsius. When carried out in the suitable manner, the approach appeared to be a very successful technology for adequately treating wastewater, and it was also the way that was easy to apply. [ABSTRACT FROM AUTHOR]

Appliances and consumer electronics (ACE) sector is one of the fastest growing industry in India. The study intends to examine the influence of service quality perception on customer satisfaction in home appliance sector with special focus on air conditioner. The moderating effect of consumer mindfulness on the relationship between service quality perception and customer satisfaction is also explored. Service quality perception was measured through SERVQUAL dimensions. Data were collected from 140 IT professionals in Kochi, Kerala, India using a purposive sampling approach. The results confirm that service quality perception has a positive effect on customer satisfaction and Consumer mindfulness has a positive relationship with customer satisfaction but does not have a moderating effect on the service quality-customer satisfaction relationship. The study offers new insights to service marketing research by exploring the influence of consumer mindfulness in enhancing customer satisfaction. [ABSTRACT FROM AUTHOR]

Sustainability practices have attracted many industries in recent years and now organizations are searching for opportunities to implement sustainability practices in their supply chains to develop a competitive advantage. The main challenge with the organization is to identify the economic factors which can help them to be profitable while meeting the environment and social objectives. In addition, the joint sustainability practices in supply chain with respect to service sector like tourism have largely remained an ignored area. A conceptual frame work is developed through many iterations and successive steps of data collection, analysis and refinement. Our proposed conceptual framework depicts that sustainability practices collaboratively undertaken by the supplier and focal organization of Service Supply Chain (SSC), which will contribute to sustainable development of service sectors. The research model is validated through a Confirmatory factor analysis. The identified sustainability practices are used for conceptualization of the model and contribute to the continuing research on sustainability performance in the supply chain. The results show strong structural relationship between sustainable practices and performance of focal organization. These identified factors will provide the opportunity for the researchers and practicing managers to logically use them in tourism supply chain. [ABSTRACT FROM AUTHOR]

Bengaluru is the fifth largest urban centre in India and one of the fastest growing cities of Asia. The city is unique with a distinction of having no perennial water resource in near vicinity. Both surface water and ground water act as the major sources of drinking water in many parts of the city. In the study, statistical analysis was accomplished to determine the ground water level, water level drop of Bengaluru metropolitan area and the major environmental factors affect the water level fluctuation. The study included 711 square kilometer BBMP area where BWSSB is supplying water through 9 divisions-North, West, South, East, North-East, South-West, Central, South-East and North-West. To meet the objectives, groundwater level (DTWL) data from 11bore wells and 64 dug wells were collected for the period 2010 to 2015 for three seasons (Pre-monsoon, Monsoon and Post-monsoon) together with the contributing factors viz; population density, altitude of a region, rainfall, municipal water supply coverage, well density etc. These factors were identified for having influence on ground water level and ground water level drop in the study area. The study gave an attempt to arrive at the predictor significance among these various contributing factors on water level and its drop. CRISP DM data mining process aided by the statistical tool (SPSS) was utilized to accomplish statistical analysis of ground water condition in the city. Present work included C5.0 algorithm which yielded best results in predicting the ground water scenario (ground water level classification and ground water level drop classification) in the entire study area. The results of statistical analysis were as follows: 1) Prediction accuracy of the ground water level classifications (good, moderate and bad) showed good accuracy in the range of 91% to 96%. 2) The prediction accuracy of ground water drops levels showed good accuracy at 83% both in moderate drop and safe drop. High drop showed around 70% accuracy. 3) Water level drop among borewells depends on Rainfall Previous, Population density, well density, K_rainfall and variation in rain fall to normal. However, the analysis showed rain fall has no immediate effect on water level fluctuation in bore wells. 4)Population density was the major factor deciding water level in bore wells whereas dug wells showed no direct correlation between ground water level & population density. Unless measures are taken to regulate population density, ground water exploitation and condition in the city will move from bad to worse inspite of the rainfall receipt by the city. Further GIS technique using Arc GIS was incorporated in the work to determine ground water condition and its availability with the help of water table movement analysis in the city. Based on the water level contours, seasonal variation of ground water levels (pre-monsoon, monsoon and post monsoon), ground water level fluctuation, progression of critical areas in terms of ground water availability were established between years 2010 and 2015. The results showed that ground water levels were receding at an alarming rate in 2015 compared to 2010. Parts of North-West, North and East were very critical zones for ground water. In addition, the paper gave an overview about the socio-economic significance of the study in Indian context. [ABSTRACT FROM AUTHOR]

This study compares different CNN architectures for identifying and categorizing brain tumors. Brain tumors are mainly of three types which are Glioma, Meningioma and pituitary adenoma. Tumors can appear in a variety of places, and the location of a tumor may reveal information about the type of cells that are responsible for it, aiding in a more accurate diagnosis. Here we will use MRI to identify and classify tumors. The main issues in MRI are illumination which can provide wrong result. To remove this the these images should be processed before training. To enhance the quality of the MRI, many image pre-processing techniques are used, such as histogram equalization, blurring and thresholding. This will help in faster classification. Six different architectures are compared in this study. These are ResNet50, VGG19, InceptionV3, MobileNet, DenseNet and NASNetMobile. By comparing this architectures we can find which one is more faster and accurate. Also maybe some architecture will be better than others for identifying certain type of tumor. That can also be identified from this. The model will be trained at different epoch to validate the result. About 20 percent of images will be used for testing and remaining are used for training. Also a part of this paper deals with calculation of brain tumor volume from MRI. Finding volume of brain tumor at two different period can help us to identify progression rate. The direction in which tumor is progressing is also identified. This will help with treatment of tumor. Volume is obtained from various slices of MRI of same brain at different locations. From each slices area of tumor is identified by segmentation. Segmentation allows us to obtain tumor part from MRI. Area will depend on number of pixels in tumor part. Two slices will be separated by a distance. This distance is multiplied by area of slice to give its volume. Summing up volume of each slice will give the approximate volume of the tumour. The main problem is each MRI should be of same resolution. This is a major design issue which we are yet to tackle as it depends on the MRI technology involved. [ABSTRACT FROM AUTHOR]

The cyclone separator is one among the most effectual and long-lasting dust collectors. Its resilience is aided by the lack of mechanical components and its capability to tolerate uneasy operating conditions. Also, cyclones are apt for applications involving high-pressure and high-temperatures. Cyclone separators utilize the centrifugal force created by the cyclonic motion for the separation of dust particles. For a normal cyclone, the dust gas enters tangentially, making the flow to rotate spirally. The centrifugal force by the cyclone makes the dust particles turn towards the cyclone's wall. Particles that collide with the wall drop into a hopper beneath the cyclone separator. The separator forces the dust-laden gas to a circular motion and the particles are thrown against the wall. And the downwardly directed gas flow near to the wall transports them to the exit at the bottom. The inlet has a rectangular cross section. The swirling gas moves axially downwards in the outer region of separation space. In the conical area of the cyclone, the gas is gradually directed into the cyclone inner zone, where the movement is directed axially upward. Cyclone separators have a downwardly directed outer vortex and upwardly directed inner vortex known as "double vortex". The gas departs through the vortex finder, which is extended from the roof top centre. This outlet is known by different names, in addition to the vortex finder, where the frequently used are vortex tube and dip-tube. Inlet gas particles are thrown outwards hitting the wall and made deposited through the dust exit by downwardly directed gas flow in the centrifugal. Existing studies reveals that the conversion of single cone into dual-inverse cone improves separation efficiency [1], and also suggests a better performing second cone length, but the generated pressure drop was of a higher value. Also, studies done in single cone square cyclone with different inlet angles [2], put forward the positive effect of inlet angle in reducing the pressure drop, where still an up gradation in collection efficiency is needed. While changing the shape of inlet from straight to incline and bend [4], the separation efficiency of the inclined was seen better. Incorporating the above results, the study comprises of the findings done computationally on the effects of inlet configurations in a square cyclone separator performance with dual-inverse cone. Four new models, with different inlet angles are developed and the performance based on separation efficiency and pressure drop is been analyzed. Pressure drop is been evaluated with five velocities between 12m/s and 28m/s. Separation efficiency is been determined in a range of particle sizes from 2 µm to 32 µm. Apart from this the separation efficiency is been compared in three different particle densities via, 1710 kg/m3, 1989.7 kg/m3 and 2320 kg/m3. The research objectives are; to model a dual-inverse square cyclone with different inlet configurations with double inlets of 0°, 30°, 45° and 60° inlet angles and observethe performance or fin particle collection. To analyze the variation of pressure, drop in four models for five inlet velocities such as 12m/s, 16m/s. 20m/s, 24m/s and 28m/s by maintaining the particle size constant. To analyse the flow field of four models at five different inlet velocities such as 12m/s, 16m/s. 20m/s, 24m/s and 28m/s by maintaining the particle size constant. To calculate the separation efficiency of the four models at two different velocities, 12m/s and 20m/s,while providing different particle sizes, such as 2 µm, 6 µm, 10 µm, 14 µm, 18 µm, 22 µm, 26 µm,30 µm and 32 µm. [ABSTRACT FROM AUTHOR]

This study is concerned with the numerical analysis of a completely submerged propeller operating at unsteady state in liquid water. The behaviour of the performance of the propeller is studied at increasing inclinations with respect to the horizontal and the nature of variations in its output torque and thrust are analysed. The simulation of the propeller for the purpose, which is a 101mm x 127mm mud motor blade, the type commonly seen in long-tail boat configurations, is conducted by utilising the k-epsilon realisable turbulence model. By implementing this numerical model, scalable wall functions for boundary conditions at normal working conditions, the output characteristics are studied. These results can be taken to consideration as useful inputs for optimal propeller mechanism systems for light outboard engine boats particularly those with long-tail configurations. [ABSTRACT FROM AUTHOR]

Multilevel inverters have gained prominence in domestic applications because of their benefits, such as improved efficiency and enhanced control. Current research is centered on creating economical and exceptionally efficient solutions. Multilevel inverters exhibit significant advantages over conventional two-level inverters due to their improved performance, ability of multilevel inverter in reducing the total harmonic distortion (THD), availability of various topology, reduction in the number of components are some of the features that make them preferable. A seven-level inverter has been designed with a reduced count of switches comparing to the conventional inverter. Here a new topology for a seven-level inverter that combines a cascaded H-bridge and asymmetric inverter topologies is presented. The inverter circuit achieves improved efficiency and reduced total harmonic distortion (THD). The developed seven-level inverter topology is suitable for various applications, including renewable energy systems, motor drives, fuel cell and grid-tied systems. Multilevel inverter (MLI) technologies are well-suited for converting renewable energy sources and meeting the demands of high-power industrial applications. They offer numerous advantages, including minimal power dissipation on power switches, reduced harmonic content, the capability to handle high power levels, and achieving impressive efficiency levels without the need for transformers. Results and observations are obtained and analyzed by simulating the inverter in MATLAB/SIMULINK R2017a. A prototype of inverter is implemented and is controlled using dSPACE DS1104 controller with two different voltage sources and attained seven level output with a gain of Vdc1 + Vdc2 [ABSTRACT FROM AUTHOR]

Change is an unavoidable aspect of life, affecting different areas such as product configurations, designs, and characteristics due to evolving environments and customer demands. Adapting to these changes necessitates considering available resources and manufacturing technology. This study aims to utilize locally accessible resources to create roofing tiles that possess desirable properties, while ensuring cost-effectiveness in the manufacturing process. The research focuses on combining Epoxy Resin LY556 and HY951 with bi-directional banana fiber to develop composite material. The properties of the composite were analysed using various techniques, including tensile testing, impact testing, and moisture absorption testing, following ASTM standards. The findings demonstrate that banana fiber-reinforced epoxy composites show promise in low strength applications such as vehicle mud flaps, engine hoods, and dashboards, providing a lightweight and strong alternative to conventional materials mainly synthetic materials. This research showcases the feasibility of developing composite specimens that integrate desirable properties of Epoxy Resin LY556 and HY951 with banana fiber, offering a cost-effective solution for roofing needs. Moreover, the mechanical attributes and potential applications of banana fiber-reinforced epoxy composites pave the way for their utilization in diverse low-strength applications, providing a novel and eco-friendly alternative to conventional materials. [ABSTRACT FROM AUTHOR]

Machine learning has been a revolutionary method that enables machines to learn from various datasets and make decisions and predictions according to its evaluation. This method has been extensively used in medical field for the prediction of various diseases using different kind of datasets available. It has also been vastly used in the medical field to develop systems to analyse various graphs and diagnose diseases. While there exist various algorithms to implement machine learning, the efficiency and complexities vary from one algorithm to another. This project aims to conduct a comparative study on some of the important algorithms and their working on datasets to predict the occurrences of non-communicable diseases in an area. This project helps in finding the best algorithm in terms of their complexities in prediction of diseases that helps in the development of an efficient prediction system which further facilitates in taking preventive measures. A novel classifier algorithm will also be developed and would be tested and compared against the existing algorithms to calculate its efficiency. [ABSTRACT FROM AUTHOR]

Nutmeg seed quality grading is an essential process that plays a crucial role in ensuring the consistency and safety of nutmeg-based products. It helps to promote fair trade practices and sustainability in the nutmeg industry, making it a vital component of the global spice trade. The seed quality grading primarily considers various characteristics of seed images. Shape, color, and texture are among the key image attributes that are extracted for analysis. The features that are created rely heavily on manual labor and are somewhat limited in scope. As a result, these features often do not perform well when applied to a broader set of data. This can lead to significant discrepancies in the final discrimination results. In the present research, a complex Convolutional Neural Network (CNN) was proposed, which integrates a classifier based on Support Vector Machine (SVM). The primary goal of this approach is to improve the accuracy of classifying dried nutmegs into either good or bad categories. [ABSTRACT FROM AUTHOR]

Human resource and management are one of the basic elements to maintain basic industrial operations. The selection of capable human assets is crucial for the overall effective industrial output. There are various factors that are responsible for the recruitment process. The factors vary from industries to industries depending on the requirement. Selecting an individual is challenging process where scrutinization of the individuals are executed through various stages. Therefore, this paper discusses about the HR recruitment process and the challenges faced of the multi-national automobile corporate company during the recruitment processes. This paper also stipulates the remedial measures for the HR team to overcome their difficulties. This study further recommends other future researchers to take over and apply the findings in other industries. [ABSTRACT FROM AUTHOR]

This study investigates the mechanical and flexural property of the Fiber Reinforced Polymer concrete (FRPC). In this study of mechanical property of concrete, the grade used in this experiment was M30. The polymer used in the concrete is Acrylic polymer the percentage added in the concrete are 2.5%, 5%, 10% and 15%. The fiber added to the concrete is E-glass fiber the percentage added are 1%, 2% and 3%. The water cement ratio initially is 0.40 as the polymer percentage increased the water cement ratio decreases. According to the compression strength and split tensile strength test conventional concrete compared to the polymer with 2.5% and fiber with 2% had a significantly higher value. The flexural analysis of the FRPC is done on the bases of the formulas which were collected from the various journals to give a broad understanding about the FRPC concrete. [ABSTRACT FROM AUTHOR]

This study looks at how well a recurrent Neural Network and a logistic regression classifier can detect cyberbullying. Methods and Equipment Used in the Study: This approach makes use of two datasets. To find out how effective new cyberbullying deductions are, twenty instances were divided into two groups: one using logistic regression and the other using recurrent neural networks. This sample was calculated using G power and an 80% pretest power. Data pertaining to cyberbullying was used for training purposes using logistic regression and RNN. The two approaches, RNN and logistic regression, vary significantly in their accuracy levels; 94.58% vs. 94.55% (p>0.05). Lastly, the Recurrent Neural Network approach outperforms the logistic regression classifier. [ABSTRACT FROM AUTHOR]

A study that pits a novel recurrent neural network-based cyberbullying inference approach against a support vector machine classifier reveals that the new method outperforms the latter. Approach and Substances Employed: This strategy is used by two firms. Twenty instances were assigned to each group so that they could test their cyberbullying deduction skills. While Group 1 made use of recurrent neural networks, Group 2 opted to employ support vector machines. We estimated the sample size using a fixed pretest power of 80% and a G power of 80%. In terms of accuracy, the data shows that RNNs are 94.43% better than support vector machines. There is a statistically significant difference, as shown by the p-value of 0.18 (p>0.05). The results show that recurrent neural network techniques are much superior than support vector machine classifiers. [ABSTRACT FROM AUTHOR]

This study compares the accuracy of a random forest classifier with a revolutionary recurrent neural network-based cyber bullying deduction. Supplies and Methods: Two businesses employ this strategy. This research used the Random Forest technique in Group 2 & Recurrent Neural Networks in Group 1 to analyze 20 samples from each group in order to evaluate the validity of the new Deduction of Cyber bullying. G power was used to calculate the sample size, and the pretest power was fixed at 80%. Findings: Random forest has an accuracy of 92.36%, whereas RNN has 94.55%. A statistically significant difference of 0.18 (p>0.05) has been discovered. To sum up, recurrent neural network methods outperform random forest classifiers in terms of accuracy. [ABSTRACT FROM AUTHOR]

The effectiveness of recurrent neural networks and adaptive boosting classifiers in deducing novel cyberbullying conclusions is the focus of this study. This approach employs a pair of sets. Group 1 utilized RNNs (Recurrent Neural Networks) to test the accuracy of creative cyber bullying deductions, whereas Group 2 used Ada boost. Each group was given twenty samples to evaluate. An 80% pretest power and G power were used to determine the sample size. Statistical analysis shows that RNN produces a much higher accuracy of 94.7% compared to Ada boost's 93.3% (p>0.05). The Recurrent Neural Network approach outperforms the Ada boost classifier in terms of accuracy. [ABSTRACT FROM AUTHOR]

This study evaluates two machine learning methods for improved fruit disease diagnosis: one that uses innovative convolutional neural networks and the other that prioritizes accuracy and sensitivity improvement. From the many fruit disease photographs on Kaggle, three thousand samples were chosen. For the sake of testing, two datasets will be utilized: one containing 800 photos (or 30% of the total) and another including 1800 shots (or 70% of the total). Honesty, precision, and reliability by doing numerical calculations, we may assess how well the convolution neural network approach works. The novel CNN produced 97.5% accuracy, 5.1% sensitivity, and 99.4% precision, as opposed to the SVM method's 83.3% accuracy, 87.9% sensitivity, and 88.3% precision. For both groups, the p-value is 0.000. It seems that the Novel CNN approach outperforms the SVM method when it comes to detecting fruit quality. [ABSTRACT FROM AUTHOR]

So that you may find out how well You Only Look Once works with OpenCV for speed-based car detection. This investigation involves two groups; one of them is the YOLO over OpenCV group. In each group, there are 10 participants, and the study settings for Glower are (α=0.05) and (power=0.85), both set simultaneously. When it comes to car detection, YOLO produces results that are 91% better than OpenCV's 84% accuracy. The accuracy of the two methods differs by a statistically significant amount of p=0.639 when assessed with two tails. When it comes to identifying new Vehicle Detection, the SVM performs far better than the You Only Look Once model. One may argue that it's the top choice for vehicle detection as well. [ABSTRACT FROM AUTHOR]

Magnesium halide and nickel halide batteries, then lithium ion batteries, ruled for several decades. Polymer batteries are the subject of the latest research, and they are of considerable scientific significance. In comparison to their liquid analogues, polymer electrolytes typically have poor ionic transmission. Consequently, numerous attempts are made to enhance its ionic conductivity by different means such as blending, adding lithium ions, plasticizers, ionic liquid, and/or inorganic additives. This study focuses on the processing and analysis of nano-polymer electrolytes made from PVC-PBMA blends for use in lithium batteries. The impact of adding rare-earth(RE) to a nano-polymer electrolyte is studied during its preparation using solution casting method. The prepared RE-PVC-PBMA blended polymer electrolytes are characterized by XRD, FTIR, ac impedance, TG/DTA, SEM, and mechanical analysis with regard to their structural characters, complex formation, ionic conductivity, thermal properties, morphological and mechanical behavior and discussed in detail. [ABSTRACT FROM AUTHOR]

The popularity of inorganic glasses is growing due to their wide range of applications. Therefore, this study focuses on creating a new type of zinc phosphate glasses (ZnP) that contain Eu3+ ions by using the melt-quenching method. The glasses were made in the following composition: (60-x)P2O5-20ZnO-10SrO-10LiF-xEu2O3, with x ranging from 0.1 to 2.0%. The XRD profiles of the fabricated glasses revealed their amorphous nature, and the photoluminescence spectra of the prepared glasses displayed five distinct emission bands at an excitation wavelength of 394 nm. Judd-Ofelt (JO) parameters were calculated from the photoluminescence spectra and showed a trend Ω2 > Ω4. The radiative total transition probabilities (AT), stimulated emission cross-sections (σ), and quantum efficiencies (η) were also calculated. The 5D0→7F2 transition at 612 nm exhibits high intensity. The characteristic color emission of the ZnP glasses was determined by means of the Commission International de l'éclairage (CIE) 1931 chromaticity coordinates (x, y) and lies in the red region. These glasses exhibited strong red luminescence. Zinc phosphate glasses (ZnP) doped with Eu3+ ions have potential applications as red lighting components. [ABSTRACT FROM AUTHOR]

Relaxor ferroelectric single crystals of (1-x)Pb(Zn1/3Nb2/3)O3−xPbTiO3, (PZN-PT) of two different compositions (x=8% & 9%) were grown by high temperature self flux method. The compositions of 'x' were chosen to be near the morphotropic phase boundary. Different orientations of rhombohedral and tetragonal domains and their coexisting phases for PZN-9%PT were confirmed from the images of polarized light microscopy. The presence of domain walls and strip like structures on the surface was observed using atomic force microscope. Refractive index was found to be 2.6 and 2.57 for PZN-9%PT and PZN-8%PT crystals respectively. Temperature dependent Raman scattering studies clearly indicate the structural phase transitions from rhombohedral to tetragonal and finally to the cubic phase upon heating. Curie temperature was found to be around 452 K and 446 K PZN-9%PT and PZN-8%PT crystals respectively. Presence of Raman bands even after the phase transition in both the samples evidence that no long-range ferroelectric order is established. Ferroelectric behaviour of PZN-PT crystals was confirmed from the butterfly like C-V curves for frequencies ranging from 10 kHz to 100 kHz. It was observed that PZN-9%PT exhibits higher capacitance when compared to PZN-8%PT. [ABSTRACT FROM AUTHOR]

The research work is investigating of sintering behavior of ZrO2−TiO2−Al2O3 Ternary system, the composite was prepared by precipitation of hydroxides from their water soluble salts followed by calcinations and sintering, the sintering behavior studied in 1:1:1 mole ratio of ZrO2, TiO2, Al2O3 between 1550°C to 1600°C. Densification depends on the function of temperature and MgO addition. It shows in significant change in densification as well as micro-structure development upto a 1mole percentage. XRD study showed that MgO addition facilitated the stabilization of tetragonal and cubic zirconia together with aluminum-titanate and Magnesio-aluminate with a dense microstructure. This Ternary system we used to mention as composite in the proceeding orders to understand clearly and this composite (Ternary system) was used at High temperature applications, it resists to mechanical (erosion) and chemical attack and thermal properties. The role of MgO in matrix is stabilized the zirconia. The result is expected to develop the formation of MgAl2O4 spinel phase and ZrTiO4 spinel phase and produce the cubic-zirconia phase with alumina Titania reinforcement. [ABSTRACT FROM AUTHOR]

Smart grid (SG) solutions that include distributed generators (DGs) are offered to address the shortcomings of the existing grid infrastructure. Recent decades have seen a decline in power quality (PQ) due to the injection of non-sinusoidal components into the grid, caused by the widespread use of power electronics devices and the expansion of non-linear loads. In a power grid, a major problem is caused by voltage fluctuations and system harmonics. Reactive power, PQ management, voltage sag, swell, flicker, harmonics, and negative sequence current are only some of the grid-connected problems that may be alleviated with the use of devices like the Active Low Pass Filter, STATCOM, SVC, capacitor bank, etc. As research methods advance, UPQC has become more effective in addressing almost every challenge that may arise. Maintaining PQ is the most crucial condition for improved process output in the sensitive load and manufacturing business, when failure to do so leads to manufacturing failures. The combination of a series converter and a shunt converter into a single UPQC makes it a switching device in the field of power electronics, and this may cause harmonics or distortion to be injected into the grid, destabilizing the system. Security and Power Quality (PQ) enhancement are two common concerns addressed by the Smart Grid (SG) technology. Because of the prevalence of power electronics and nonlinear loads, harmonics are introduced into the system. Popular Unified Power Quality Conditioners (UPQC) from the Flexible AC Transmission System (FACTS) are often used to address voltage sag, swell, flicker, power quality (PQ), and neutral current reduction in distribution networks. System stability for sensitive loads is impacted by the harmonics generated by a UPQC. In this work, we provide a new controller for the modified UPQC coupled with SG, along with strategies for eliminating harmonics. Harmonics of higher orders are suppressed by infusing harmonics of the same order, of comparable amplitude, but opposite phase from the other converter, while lower order harmonics are removed by careful consideration of switching angles. Modified UPQC converters use energy from a PV (Photo-Voltaic) panel as a source of excitation. Through the use of pre-stored firing angles in the microcontroller's memory, the series-shunt converter's firing angles may be determined instantly. [ABSTRACT FROM AUTHOR]

In this study, a unique method for balancing bipolar DC charging stations for electric automobiles is presented. To access both DC buses and establish complimentary balance, the suggested solution employs a grid-tied neutral point clamped converter with an energy storage step. This approach is compatible with various types of energy storage systems, thereby reducing hardware requirements. To account for DC currents, the system uses a three-level DC-DC interface. To avoid even-order harmonics during asymmetrical operation, the central converter switches via a different switching sequence. Unlike traditional methods that require a balancing circuit, this proposed technique eliminates the need for one, allowing the use of off-the-shelf rectifiers and fast chargers without significantly altering the charging process of the energy storage system. This study also shows the comparison between the ANN hybrid controller and the PI controller in regulating the DC-link voltage to the desired level using Vector Control Technique. [ABSTRACT FROM AUTHOR]

Tall building growth has been expanding at an alarming rate across the globe, posing new issues that must be addressed via engineering decisions. In the present analysis of work, an outrigger system is suggested in order to improve the performance of the structure when subjected to seismic stress. The modelling of the structure is carried out with the help of the "ETABS" software. Analysis of the model is carried out using two methods: the equivalent static approach and the response spectrum method. In order to provide stiffness against static and dynamic loads, the stiffness and efficiency characteristics of a structure are quantified in terms of lateral displacement, storey drift, base shear, and basic natural period for various kinds of structures. [ABSTRACT FROM AUTHOR]

This research paper delves into the exploration of Jute/Epoxy composites, a material of growing interest in the automotive industry. Jute, a natural fiber, is gaining preference due to its cost-effectiveness, renewability, and satisfactory mechanical properties. These attributes make it a compelling alternative to synthetic fibers in automotive applications. The study also emphasizes the importance of hole drilling, a secondary but crucial process in manufacturing. This process is integral to the assembly of various automobile parts, and its impact on the mechanical properties of composite materials is often underestimated. The core of the research investigates the effects of open-hole tensile testing on Jute/Epoxy composites, with a particular focus on the impact of varying hole diameters. The diameters examined in the study were 6mm, 9mm, and 12mm, with a constant loading rate of 1 mm/min maintained throughout the testing process. The primary finding of the research was a significant decrease in tensile strength with an increase in hole diameter while keeping the thickness constant. This inverse relationship between hole diameter and tensile strength is a vital consideration in the design and manufacturing processes of composite materials. The findings of this study underscore the need for careful consideration of hole diameter during the hole-drilling process. This factor directly influences the tensile strength of the composite material, thereby affecting the overall performance and safety of the automobile. The insights provided by this research are invaluable for manufacturers and engineers in the automotive industry, contributing to the development of safer and more efficient vehicle parts. [ABSTRACT FROM AUTHOR]

There is a correlation between our posture and our physical and emotional wellness. The detection of various human postures has been approached in a variety of ways. Determine a patient's resting position, for example, using posture analysis in the medical industry. Image processing using OpenCV and mediapipe python library for human posture estimation. Analyzing standing and sitting postures is made easier with an image-processing-based technique. The benefits of fitness activities to a person's health are remarkable, but if they are done improperly, they may be ineffective and even detrimental. Exercise blunders happen when someone doesn't adopt the right posture. This suggested application makes use of pose estimation to identify the user's workout posture and then offers specific, tailored advice on what the user can do to correct their posture. A pose estimation module in python called mediapipe helps to detect all major joints of the human body. Then it calculates the angle between the joints and increases the counter for each repetition. Pose estimate uses a picture or video of the subject to determine the precise locations of the body's major joints. This computer vision technology recognizes human posture in films and displays important regions, such as the elbow or shoulder, in the finished product. Another way to work out at home is through interactive games. The proposed application has an interactive boxing game built using pygame, which helps all age groups to work out and take care of their health. Overall the user can track all their details on a workout on this application, like how many calories are burned and the daily goals, and different workout planners for each individual. [ABSTRACT FROM AUTHOR]

Image classification is a supervised machine learning that can be used to classify the images based on certain factors. Classification takes an image as input and then generates one or more labels that correspond to what it sees in the image. Image classification takes the image as input and interprets what it sees. The major goal of this work involves the use of Deep Learning (DL) approaches. This research work identifies COVID-19 dataset as a critical issue for medical practitioners nowadays because of its rapid dissemination. To address this issue, medical practitioners have tried a variety of tools and approaches to detect and restrict the spread of COVID-19. The best method for identifying COVID-19 is currently using CT (Computed Tomography) scan and X-ray imaging. This diagnostic procedure is particularly accurate since it can see organs in three dimensions. Unfortunately, this procedure requires a radiologist to detect the sickness and takes a long time, medical practitioners will lose crucial time. This research creates a method to detect the coronavirus automatically using DL algorithms. The purpose of this study is to give medical professionals a search tool for CT scans and X-ray images of lungs with COVID-19 infection. DL algorithms such as AlexNet, ResNet, Inception and VGG16 are addressed in this work that helps to automate the classification process. This research work incorporates the CT scans and X-ray images that helps to detect the COVID-19. Experimental result shows that VGG obtains good result compared to other algorithms. VGG obtains 94 % accuracy for X-ray images and 93% accuracy for CT-scan dataset and outperforms best result compared to other algorithms. [ABSTRACT FROM AUTHOR]

The task scheduling and data security issue in the cloud is well-studied, and different approaches exist around the problem. Most approaches consider makespan, throughput, delay, and others in task scheduling. The existing methods need better performance in various quality of service metrics as they consider only limited features of task scheduling. To overcome these issues, an efficient multi-factor adaptive task scheduling (MATS) is presented in this article. The method considers the cost, delay, and utilization of various services in selecting a service for task scheduling. According to the factors mentioned above, the method applies MFTS scheduling which approximates the factors toward maximizing the QoS performance. The MFTS algorithm computes Cost Support Measure (CSM), Delay Centric QoS Support Measure (DCQSM), and Utilization Support Measure (USM). Using all these measures, the value of Selection Weight (SW) is measured to support scheduling. Further to maximize data security, the proposed MATS model discovers the routes and estimates Quality Centric Transmission Support (QCTS) to identify an optimal route to support QoS maximization. The MATS scheduling approach achieves higher QoS performance with data security. [ABSTRACT FROM AUTHOR]

About one person dies every minute from heart disease, consequently, it has surpassed war as the largest cause of death in the twenty-first century. This includes both the male and female demographics, as well as the ratio may fluctuate depending on age bracket & sex. Nothing here suggests that persons of various ages are immune to cardiovascular disease. Classifying the underlying causes and diseases associated with this condition has become more difficult in recent years. In this study, we explore the different classification methods and the Flask Framework used for cardiac illness. As a result, it is drawing attention as a "fatal disease" that may kill a person in the absence of any outward signs of illness. Particularly important in cardiology, early and accurate diagnosis of heart illness is a cornerstone of effective healthcare. Classifying cardiac illness might be difficult at times because of a general lack of funds in the healthcare profession. The medical community and their patients stand to gain a great deal by making use of appropriate technological help in this area. This problem may be handled by using Data Science methodologies. There are now established tendencies in Data Science, and a wide range of research is being conducted in this area. This research seeks to apply the CNN, RF, Naive Bayes, SVM algorithm techniques to create a classifier model for the of Cardiovascular disease. To put our plan into action, we've chosen to work with the Flask framework. Categorizes the desired value in-page using HTML and CSS to connect our models to variables. Given that it's a user-friendly website. An advantage of this framework is that it makes it simple for users to make sense of the values that have been categorised. Users may learn the status of an illness at a preliminary phase by using our website. [ABSTRACT FROM AUTHOR]

Implementation of a Delayed LMS algorithm(D-LMS) based adaptive filter using in Verilog HDL. The structure of the LMS algorithm-based adaptive filter has a longer critical path, an efficient structure for the execution of the LMS algorithm having a lower critical path, called D-LMS is discussed. The D-LMS will be implemented in Verilog HDL. For the design of the D-LMS adaptive filter, necessary building blocks, such as block carry look-ahead adder (BCLA) & multiplier are discussed. These circuits are efficiently used to implement D-LMS adaptive filter. The whole system is designed using Verilog HDL and synthesized using Xilinx ISE software. The detailed synthesis report along with the RTL schematic will be presented. [ABSTRACT FROM AUTHOR]

Dyslipidemias, the major cause for various lifestyle-disorders, especially CVD accounting for one-third of total deaths worldwide, is drastically increasing. Long-term use of conventional medications causes adverse effects. Hence, there is a need of safe interventions to prevent and manage this disease entity. In the tradition of Ayurveda, dyslipidemias are a consequence of an error in Ahara-parinama, or the process of digestion, absorption, and assimilation of food. Low digestion strength, or Jatharagni-mandya, creates inappropriate Sara-Kitta Vibhajanam (by-products of digestion), which stores up in the body and could be recognized as an excess of circulating lipids. Certain clinical issues may arise when this malarupi-kapha localizes across various srothas (channels). The most important objective of treatment is to prevent the vitiated Kapha-Dosa (body humor that governs bodily fluids and keeps the structural coherence of the organism) from advancing through the stages of Sancaya (stage of accumulation), Prakopa (stage of aggravation), or Prasara (phase of the liquefaction and disseminating of doshas) to a fully manifested disease. To achieve Samprapti-Vighatana (intervening pathology) in this instance, disorganized Agni must be rectified by Pācana - Dīpana(digestive-carminative) with Gandharvahastādi Kaṣāya and Aṣṭacūrṇa, after that Kapha-Dosa shall be comforted by Dosapratyanika-Chikitsa(treatment corresponding to respective body humors)with help of Viḍaṅgataṇṭhulādi Cūrṇa. Viḍaṅgataṇṭhulādi Cūrṇa with Pācana - Dīpana protocol is safe and effective in Dyslipidemias. [ABSTRACT FROM AUTHOR]

Ayurvedic medicines are gaining importance worldwide. The main challenge faced by the Ayurvedic drug manufacturing industry is with respect to the stability of the products. For keeping the products stable till the expiry period, Ayurvedic industry also uses preservatives of various kinds, just as those used in food industry. There are certain legally permissible limits which are to be followed while using these preservatives and even though most of these preservatives are generally recognized as safe by Food and Drug Administration (FDA), many studies have suggested the interactions of these with the ingredients of the food. The present review aims at giving an insight to the interactions and health hazards of preservatives used in Ayurvedic medicines. [ABSTRACT FROM AUTHOR]

Herbal remedies are known for their widespread availability and lower toxicity compared to modern medicines, Medicinal plants have been historically utilized in the treatment of various illness, with cancer emerging as a significant global cause of mortality. In spite of the numerous treatment methods, the high incidence rate, diagnostic challenges, and adverse effects associated with existing treatments present formidable obstacles in effectively addressing this disease. Given the complexity of combating cancer, there is a pressing need for innovative strategies in its treatment. Natural products, especially those derived from medicinal plants, play a pivotal role in the battle against cancer cells. These plants contain a rich array of bioactive compounds with proven anticancer properties. This review concentrates on exploring lesser-known ethnobotanicals found in the Indian subcontinent and adjoining regions of Southeast Asia, which show promising potential in the fight against cancer. By offering an overview of the active phytochemicals present in these plants and elucidating their mechanisms of action within cancer cells, this review aims to provide valuable insights. This information serves as a foundation for further studies, facilitating the exploration of additional plant-based products with therapeutic potential in the treatment of cancer. [ABSTRACT FROM AUTHOR]

Early diagnosis and treatment outcomes for skin cancer depend on these factors. Recent studies have shown that deep learning models with dermoscopic images can classify skin cancer with good accuracy rates. In this article, we develop and evaluate deep-learning models for skin cancer classification using a large dataset of dermoscopic pictures. There are more than 200 different types of cancer. The first stage in the diagnosis of cancer is often clinical screening, which is then followed by dermoscopic examination and histological analysis (biopsy). Melanoma is one type of skin cancer that can be treated if found in its early stages. Using high-speed cameras, dermatologists take dermoscopic images of the skin lesions, which are subsequently evaluated. For the early identification of skin cancer, the use of suitable automated examination models can be highly advantageous. In this study, labels indicating the presence of one of six distinct types of skin cancer—were applied to over 30,000 dermoscopic images of skin lesions (vasc). The photos are preprocessed in order to standardize the intensities, resize them to a uniform size, and maybe enhance them in order to expand the dataset. For the categorization of skin cancer, we examine convolutional neural networks (CNNs) and transfer learning techniques. The models are tested on a different test set after being trained on a subset of the dataset to determine how accurate they are at classifying the various forms of skin cancer. [ABSTRACT FROM AUTHOR]

It was discovered that the ionosphere is a major contributor to the fact that satellite positioning and measurement systems can at times produce inaccurate results. In addition to investigating the ionosphere's other properties, it is essential to investigate the ionosphere's variability in relation to the activity of the sun. This is because the solar extreme ultraviolet (XUV) radiations are primarily responsible for the fonnation of the ionosphere. To precisely determine the amount of time lost due to the ionosphere's influence on the receiver, one must be aware of the ionosphere's total electron concentration. This research will largely concentrate on the development of models for ionospheric delay correction. It is imperative that you have a thorough comprehension of the challenges posed by ionospheric delay time correction models as well as the opportunities they present. [ABSTRACT FROM AUTHOR]

Plastics have changed the way we live our lives. It's conceivable that the operation's periodic product will surpass 300 million by every instance. We want to evaluate the PET bottles that contain air, hydrogen, helium, and neon since it is obvio us that plastic brings many societal advantages as well as undeveloped technical and coastal areas. It's purpose is to float on water. In order to create the blocky box, the bottles were coated with nylon net, polyester, and glass fibre. In order to gauge not only weight but also capacity. Its primary users were the coast guard and fishermen. And thus, at various feasts, to put some weight in a blocky box. The purpose of this design is to evaluate the buoyancy test in various stages to choose the most fashionable gas for use in safety operations. [ABSTRACT FROM AUTHOR]

In this work, analysis of complementary split ring-based resonator (CSRR) structure and a crossed-slot structure are the main emphasis. The conventional patch antenna is designed to operate at 10.8GHz frequency and is tuned to operate at three resonant frequencies at 7.8 GHz (7.64 - 8.1GHz), 10.8 GHz (10.36 - 11.63GHz), and 18GHz (17.28 - 19.17GHz) which can be significantly useful for wireless applications. And the realized gain obtained at 7.8 GHz, 10.8 GHz, and 18GHz using High Frequency Structure Simulator (HFSS) is 3.17 dB, 6.25 dB, and 7.62 dB. The antenna is designed on a FR4 epoxy dielectric substrate with a relative permittivity of 4.4 and thickness of 1.6mm, and it is fed using an inset feeding approach. The outcome demonstrates that the size minimization can be accomplished through CSRR inclusion. [ABSTRACT FROM AUTHOR]

In today's world, binary logic is implanted utilizing CMOS. However, CMOS has several drawbacks, including high leakage power and short channel effects. CNFET is used to implement ternary logics to avoid these drawbacks. The main benefit of ternary logic is that it takes up less space on the device and requires less memory. Half adder circuit for ternary logic is proposed in this study. To begin, a decoder is employed to convert ternary to binary signals. After going through a binary half adder, the binary signals are transformed to ternary using an encoder. The disadvantage of utilizing a simple encoder is that it has a low resistance route and so uses a lot of power. To address this, numerous encoders have been devised, including the basic encoder and the enhanced encoder. To save energy, encoders with low power consumption are employed. To eliminate latency, high-speed encoders are used. Variables such as chirality, diameter, pitch, number of tubes, oxide thickness, and dielectric materials are used to do a complete study of the encoder's power consumption and latency. This improved Encoder is used in conjunction with a half adder. [ABSTRACT FROM AUTHOR]

To create a screening tool that uses deep learning to recognize designs of ILD. For the semantic separation of different ILD designs, a fully convolutional network was employed. ILD designs were better segmented using inter morphological operations. To keep feature maps resolution and broaden the receptive area, dilated convolution was applied. This article presents an edge depth CNN for classifying ILDs designs. The proposed framework comprises of regularization, peak, and four largely owing besides a thickness equivalent to a final output fully explore. It also includes a sparse stack auto encoder and decoder with convolutional neural networks, vgg16, vgg19 and resnet50 activation functions. [ABSTRACT FROM AUTHOR]