Search

Attention-deficit/hyperactivity disorder (ADHD) is a complex neurodevelopmental disorder characterized by symptoms such as inattention, hyperactivity, and impulsivity. Early and accurate diagnosis of ADHD is crucial for effective intervention and management. This project presents a novel approach to detect ADHD in individuals using brain scan data and machine learning techniques. The study utilizes a brain scan-making machine, and the data is pre-processed and analysed using popular Python libraries, including NumPy and Pandas. To develop the predictive model, TensorFlow and TensorFlow Keras layers are employed for building and training the deep learning model. The proposed methodology aims to identify distinct brain activity patterns associated with ADHD, allowing for the reliable prediction of ADHD presence in individuals. The project emphasizes the potential of machine learning in neuroimaging research for ADHD detection, with practical implications for early diagnosis and personalized treatment strategies. [ABSTRACT FROM AUTHOR]

Analysis and forecasting of stock sentiment has always played a significant role in determining the stock markets' profitability. To examine and forecast the performance of specific stocks, ETFs, etc., these methods have been developed. Thus, it becomes essential to have a sophisticated system that could extract all relevant information about the Stock market in order to assess their performance and provide the traders with the necessary support. The goal of the work is to forecast trends by scraping online stock prices. This stock prediction model was developed using three distinct studies and approaches, including sentiment analysis, machine learning algorithms, and technical analysis using technical indicators, to ensure a more accurate prediction. Different people employ various strategies when it comes to investing in stocks. Our approach combines the outputs from the three prediction techniques into a single result. The results of these three approaches taken together will help the investor forecast the stock performance. [ABSTRACT FROM AUTHOR]

This study aims to compare Naive Bayes and Decision Tree in terms of their ability to forecast loan acceptance rates. The study is split into two groups, Naive Bayes and Decision Tree, with 10 participants in each. Using the pretest power value, the G Power program calculates the sample size. is 0.8 and alpha 0.05. The predictive model using the Naive Bayes algorithm shows a higher Accuracy of 83% than Decision Tree algorithm with Accuracy of 69% with the significance value <0.05. The outcome of the study concluded that the Naive Bayes based model provides more promising results in prediction of loan approval than Decision Tree based model. [ABSTRACT FROM AUTHOR]

The primary goal of this research was to examine how well Naive Bayes and Logistic Regression predicted loan acceptance. Ten participants were used for the Logistic Regression and Naive Bayes components of the research, respectively. Using the alpha level (0.05) and the pretest power rating (0.8), the Gpower program determines the sample size. The Naive Bayes algorithm-based prediction model outperforms the Logistic Regression approach with an 83% accuracy, while the significance value is less than 0.05. The research showed that when it came to forecasting whether or not a loan would be approved, the Naive Bayes model outperformed the Logistic Regression model. [ABSTRACT FROM AUTHOR]

In this study, hybrid approaches are proposed to deal with the complex features of nonstationary and nonlinear behaviour. The proposed model takes into account some features, such as the seasonal component and the trend as an input data. However, to the best of the authors' knowledge, statistical patterns, including trend, seasonality and nonlinear behaviour, have rarely been studied simultaneously in bitcoin data. Therefore, a hybrid of structural time series (STS) and nonlinear autoregressive with exogenous input (NARX) is developed to predict bitcoin closing price. The results reveal that the optimal model is the hybrid STS, combining Local Level with Drift Deterministic Seasonal and NARX, featuring two inputs, one output, two delay values, and ten neurons in a hidden layer. This model demonstrates superior performance with the lowest values across all three measurement errors when compared to a single linear STS in bitcoin price prediction. [ABSTRACT FROM AUTHOR]

The objective of this study is to improve the accurateness of the skin disease prediction using ML algorithm by comparing SVM and Extreme Gradient Boosting (XGBOOST) models. Skin disease prediction models developed using ML method have improved the accuracy of diagnosis. The methods currently available for building accurate skin disease prediction models require extensive data collection. This study compared the enactment of two popular ML algorithms, SVM and XGBOOST, in predicting skin disease using a dataset obtained from Kaggle consisting of clinical and demographic variables. The dataset was preprocessed, and both algorithms trained and tested both algorithms and found that XGBOOST outperformed SVM. The findings of this study indicate that machine learning algorithms have the capability to provide precise predictions of Skin disease. The outcomes of the autonomous samples t-test indicated that the mean accurateness of XGBoost was expressively higher than the mean accuracy of SVM (p < 0.001, p < 0.05), a statistically noteworthy variation between the algorithms related to classification accuracy. By this learning accurateness of skin disease prediction is improved while analyzing the results of two ML algorithms SVM and XGBoost. The study compares the accuracy of both algorithms using a dataset. The conclusion is that the proposed model of XGBOOST 94% outperformed the SVM 59.6% within this dataset. [ABSTRACT FROM AUTHOR]

Diabetic disease is the mostly affected and massive disease on a global level. Diagnosing the diabetic earlier will help the medicalist to give the improved and latest clinical treatment. The healthcare specialist unit uses many machine learning techniques, methodologies and tools for decision making in diabetic field. The machine learning techniques are utilized for the prediction of the diabetic diseases in the initial level. To eliminate such issues, optimized detection techniques are proposed. First of all, the training samples are increased using the sliding window protocol. Further, class imbalanced training data classes are balanced and resolved using the adaptive and gradient booster technique. Further, the diabetic feature selection process is improved by the Intensity Weighted Firefly Optimization firefly techniques (IWFO), in which irrelevant features are reduced based on the correlation between the features that deducts the unwanted features involved in the diabetic disease process. Then the feature transformation problem is faced by the PCA technique, which manages the several types of features. Finally, the improved and optimal hybrid random forest is applied into the normal and diabetes classes respectively. The proposed system predicts the diabetic disease efficiently and maximizes its precision of the prediction system. The present paper is compared with different classifiers to determine the efficiency of the work. Overall, the initiated system improved the present studies accuracy level. [ABSTRACT FROM AUTHOR]

Acute myocardial infarction, commonly referred to as a heart attack, stands as one of the most lethal medical conditions, highlighting the pressing necessity for the effective management of cardiovascular disease. This involves conducting comprehensive data analysis and extracting knowledge essential for diagnosis, regulation, and treatment. Anticipating the occurrence of heart attacks presents a formidable challenge for healthcare professionals, given the intricate nature of the condition that demands both experience and a profound understanding. In the contemporary landscape of medicine, the concealed data landscape conceals invaluable insights that can significantly shape critical decision-making processes. In this research endeavor, a dataset comprising patient records is harnessed to predict an individual's vulnerability to heart attacks. Advanced data visualization techniques are employed to identify pivotal trends and outliers, facilitating the extraction of meaningful and actionable conclusions. This study involves the development of three classifier models for heart attack prediction: Logistic Regression, K Nearest Neighbor, and Support Vector model. [ABSTRACT FROM AUTHOR]

A predictive mathematical model is presented to analyze whether the cases of lung cancer are increasing or decreasing after COVID-19. Further to investigate the age range in which it is more attacked by the disease and its modes of treatment. The relevant data collected from the previously published articles are utilized to obtain the simulation results for the mathematical model considered. The mathematical model considered for the present study is a result of the case study undertaken on the patient-specific data obtained from a deceased patient who was suffering from post-COVID fibrosis. The beginning and end of the first phase and the last phase of each phase pair's variability showed identical phase plots for each phase duo. Implementing control techniques to optimize each compartment had a big impact on every division. The graphical representations also provided evidence for the necessity of controlling the condition that causes post-COVID fibrosis. The mathematical model and the exploration of the results in terms of the orientation of different phases add novelty to the present research. [ABSTRACT FROM AUTHOR]

The economic crisis has hit countries in Southeast Asia including Malaysia resulting in many companies experiencing financial crisis. Therefore, companies seize the opportunity to take preventive measures against financial crisis or bankruptcy. The use of different bankruptcy prediction models with different economic environments in the companies involved shows different results. In general, each model will give ideas for each company to make changes in business to avoid bankruptcy. The objective of this study is to evaluate the suitability of the bankruptcy prediction model and improve the effectiveness among the Practice Note 17 (PN17) and healthy companies that listed in Bursa Malaysia Stock Market with the Altman, Springate, Grover and Zmijerski model. The data of this study consists of 27 of PN17 companies and 155 of healthy companies are collected from 2017 to 2021. The models employ the logistic regression method using SPSS to predict multiple financial ratios simultaneously to assess a company's financial distress. The result show that 100% accuracy of the Grover model suggested method has an acceptable efficiency to predict financial distress followed by Altman, Springate and Zmijewski model. Meanwhile, the significant financial ratio variables are (current asset-current liability)/total assets (X1), sales income/ total assets (X5) and earnings before interest and tax (EBIT)/total assets (X6) for Altman and Springate model. While return on assets (ROA) and debt ratio (DR) for the Zmijewski model. [ABSTRACT FROM AUTHOR]

The Details of accidents on roadways, in workplaces, or public areas will be addressed in this study paper, it can have severe repercussions. In recent years, there has been an increase in interest in studying mishap data in a bid to comprehend their causes as well as create effective protection strategies. This paper will investigate the different approaches and methods used in the analysis of mishap data and associated effects, emphasizing the advantages and disadvantages of this approach, and giving instances of effective uses in various contexts. Analyzing mishap data and associated factors aids in the identification of patterns and trends, the development of predictive models, and the creation of measures to reduce the risk of accidents as well as improve overall safety. [ABSTRACT FROM AUTHOR]

The world is still struggling with a substantial loss of human life as a outcome caused by COVID-19. For futuristic purpose of wellbeingness of human life it's important to study such type of pathogen in an authentic manner using mathematical modelling. In this study some mathematical prediction models based on data from nations, including China, Italy, France, India, Iran and Turkey are reviewed. It was observed that researchers presented a model in number of ways for projecting COVID findings. It was also observed that result from trained data correlate to a great extent albeit not fully to the test data. Finally, several potential futuristic techniques for improving model efficacy are proposed. [ABSTRACT FROM AUTHOR]

Due to their detrimental effects on the environment and human health, the rising amount of microplastics in wastewater has become a growing concern. To address this problem, a predictive model is required to analyze the presence and concentration of microplastics in wastewater with precision. This paper proposes the development of a VARMA (Vector Autoregressive Moving Average) autoencoder-based predictive model. To extract meaningful features from the input data, the model employs autoencoders, a type of neural network that can learn and compress complex data patterns. The VARMA model is then used to capture the dynamic relationship between the extracted features and the output variables of interest, namely the presence and concentration of microplastics in wastewater. The proposed model has multiple applications, including providing insights into the efficacy of various wastewater treatment methods in removing microplastics, predicting the potential sources of microplastics in wastewater, and assisting in the development of strategies to reduce the total amount of microplastics in the environment. The model proposed in this work has several advantages over existing models, including its ability to handle non-linear relationships and its efficiency in processing large datasets. In addition, the VARMA autoencoder-based model's capacity to capture dynamic relationships between features and target variables makes it ideal for analyzing the presence and concentration of microplastics in wastewater, where the data is frequently complex and nonlinear. [ABSTRACT FROM AUTHOR]

The primary audience for this project is the younger generation, which frequently enters the stock market and makes investments based solely on advice from others, despite the fact that the stock price genuinely depends on the stock's back graph, or previous ups and downs. The goal of this project is to build a web application that will allow users to view all stocks by stock code. After creating an account, the user can sign into the website. Each member will be taken to the stock prediction page after logging into their accounts, where they can view the most recent trends and enter the stock code. Prior to use, the indicators are converted to binary data after being initially derived using stock trading values. Three metrics are used to evaluate each prediction model in accordance with the input techniques. According to the evaluation results, RNN and LSTM greatly outperform other prediction models for continuous data. Also, the findings show that those deep learning methods are the best for analyzing binary data, albeit the gap between them is gradually narrowing due to the second method's obviously enhanced model performance. [ABSTRACT FROM AUTHOR]

In this era, technology has revolutionized the health industry to a great extent. The proposed model aims to design a diagnostic system for various diseases based on their symptoms. The Disease Prediction System has implemented different ML prediction models for the prediction of the user's disease based on various symptoms inputted by the user. Machine learning classification algorithms analyse the inputs given by the user and then predict the disease and probability of occurrence of the disease as output. The proposed system predicts the diseases such as i) Diabetes, ii) Kidney, iii) Cancer, iv) Heart and v) Liver. Four prediction models, Naive Bayes, Decision Tree, Random Forest and Logistic Regression, has been implemented in the proposed system for various disease. The dataset "Disease Prediction Using Machine Learning," with a count of 132 symptoms, has been used in the proposed model. The main goal of the proposed model is to predict the disease; however, the user doesn't need a medical report to use this system as the prediction is based on the symptoms, which will save time and money. The system also has an easy-to-use user interface, and all the users can use it to predict genetic diseases. [ABSTRACT FROM AUTHOR]

Wine is an alcoholic drink, which is produced by fermenting grapes in a controlled environment. Different varieties ofgrapes and strains of yeasts are major factors in different styles of wine. Wine has been produced for thousands of years and has a long history and culture in many regions of the world. Wine can be classified into different types, such as red, white, rosé, sparkling, dessert, etc. based on the color, sweetness, carbonation, and other factors. The global wine market is expected to growfrom $340 billion in 2021 to $456 billion in 2028. As wine production is part of such a huge industry, quality checks involved consists of sophisticated procedures, which include the participation of distinguished wine quality checking officials, called Wine Sommeliers. These Sommeliers require years of rigorous training and practice to give a valid verdict about the quality and classification of different wines. [ABSTRACT FROM AUTHOR]

Stock market forecasts are always attracting the attention of multiple analysts and researchers. Many Popular theories observe that the stock market is, in its essence, a random game of chance and that it is a mindless game to try to predict a thin item. Predicting a stock'sprice is a challenging problem due to the number of external variables present. The market behaves in the short term like a ballot or a voting machine, but in the long run, it acts like a weighing machine and may thus forecast market movements overa more extended time period. Stock Price prediction's integration with modern technology – especially Machine Learning Algorithms (Quant models as often referred to in the financial sector) isrecently becoming a growing idea for research. Research has shown that Machine Learning Models, particularly with the use of Recursive Neural Networks(RNN) and Long-Term Short Memory (LSTM), when applied to historical data of shares, can be utilized to predict the short-term price of the share. Our research aims to find the best model that uses companies' projections and sector performances and how the given company fares accordingly to correctly predict equity share pricesfor both short- and long-term goals. [ABSTRACT FROM AUTHOR]

Online advertising has a big impact on whether your business succeeds or fails. Because of this, it is crucial to assess your advertisement's effectiveness before posting it online. Finding the Click-Through Rate (CTR) allows for this. Unfortunately, because you must gather user clicks before determining Click-Through Rate, this method is not environmentally friendly. In this situation, CTR prediction is helpful. For forecasting ad Click-Through Rate, user click data is a crucial source of information. Accurate Click-Through Rate prediction for contemporary e-advertising platforms is a challenging and crucial undertaking. Click-Through Rate prediction employs machine learning methods to determine how many times a potential consumer has clicked on an online ad. The more clicks an advertisement receives, the more successful it is. In this paper, we create a machine learning-based Click-Through Rate prediction model. Finding the Click-Through Rate allows for this. Unfortunately, because you must gather user clicks before determining Click-Through Rate, this method is not environmentally friendly. In this situation, CTR prediction is helpful. For forecasting ad Click-Through Rate, user click data is a crucial source of information. Accurate CTR prediction for contemporary e-advertising platforms is a challenging and crucial undertaking. Click-Through Rate prediction employs machine learning methods to determine how many times a potential consumer has clicked on an online ad. The more clicks an advertisement receives, the more successful it is. In this paper, we create a machine learning-based Click-Through Rate prediction model. The proposed study defines a model that produces accurate results with minimal use of computational resources. Three classification methods were used namely logistic regression, decision tree classifier and random forest classifier. Awasu dataset was used for analysis. The click data is generated over a 10-day period and sorted chronologically. This study answers the following question: Considering a user and the page they visit. What is the likelihood that they will click on a particular ad? The Random Forest classifier proved to be the best model with an accuracy score of 96%. [ABSTRACT FROM AUTHOR]

Due to depletion in raw material quality, blast furnace (BF) aerodynamics, hence the permeability uses to be disturbed in BF#8, Bhilai Steel Plant (BSP). Deviation in middle zone permeability of BF#8 suggested the deviation in cohesive zone shape and position. For better understanding and monitoring of the lower zone permeability of BF#8, BSP, two models were developed. The models were Cohesive Zone (CZ) shape and position prediction model and online presentation of Dead-man clinginess Index (DCI). Use of CZ model is helping in assessment of type of cohesive zone shape through which burden distributions as well as blast parameters arebeing regulated. Hearth cleanliness condition is monitored through online DCI model. Use of DCI is helping in devising the optimum casting strategy for proper drainage of liquid products and also to maintain healthy and active hearth. [ABSTRACT FROM AUTHOR]

Electrochemical chemical discharge machining (ECDM) is the hybrid machining process as an alternative in machining process of material that require precision dimension and complex structure. The demand for the manufacture of tungsten carbide products is a challenge due to its high strength. Technological innovation in metal cutting processes such as hybrid manufacturing-based technology is independent of the hardness of the workpiece. This study aims to predict the response value and identify the combination of process parameters to achieve the required performance characteristics in the ECDM process of tungsten carbide using the ANN method. The four process parameters selected were voltage, gap width, electrode type, and electrolyte type, with each parameter having two levels and full factorial 24 selected as the design of experiment. R-squared is then used to determine how accurate the model is. In comparison to the linear regression model, the Artificial Neural Network model produces more precise and reliable results. Using the desirability function, the input parameters were improved for a higher MRR. The neural network model was used to validate the results for the best solution. The validation serves as an example of the impact artificial intelligence-focused machining techniques have on optimizing progressions. [ABSTRACT FROM AUTHOR]

There is a large amount of public data available that is spread across various sources. However, these public data is spread across different government bodies' websites and repository platforms. Data must be collected and stored in a warehouse for utilization. Hence, this research is conducted to design and implement an appropriate data warehouse that is able to merge all public data from different government bodies and public websites in one data warehouse. This research started by locating relevant sources from various open data platforms such as Public Info Banjir and the Official Portal iDengue for community version 3.0. Then, significant data were collected such as river water level data, rainfall data, dengue locality cases and weather data. A public data warehouse was constructed using the bottom-up approach of the Kimball methodology. Following this, a galaxy schema model was identified to design the fact and dimension tables, which is then converted into a logical model. Finally, the structured data warehouse was developed. The public data warehouse, containing data such as water level, rainfall, dengue locality cases and weather data will be of great benefit to both government bodies and the public. For future study, the data warehouse enables the development of advanced analytics and predictive models, which helps to enhance decision-making and optimize infrastructure management in relation to flood mitigation efforts. [ABSTRACT FROM AUTHOR]

This work addresses the urgent demand for exact sales estimates in the automotive industry by utilising Random Forest Regression, a potent predictive modelling technique. Compared to conventional forecasting techniques, this approach gives a more detailed grasp of market trends and consumer preferences. It works especially well in the dynamic automotive business, which is characterised by quick technological change and shifting consumer preferences. For manufacturers and dealerships to plan production schedules, manage supply chains, and handle inventory, it is stressed how important accurate sales forecasts are. In order to provide a thorough overview of the industry, the study takes into account a wide range of vehicle characteristics, including engine characteristics, safety features, decorative components, and technology improvements. The methodology uses the RandomForestRegressor model because it can handle complicated interactions and non-linear patterns, as well as data collection, preprocessing, feature engineering, model selection, training, and evaluation. The study also includes a user-friendly web interface that allows users to enter specific automobile details and get sales projections. The project aims to improve the accuracy and granularity of automobile sales predictions by merging cutting-edge machine learning algorithms with substantial car characteristic data, giving industry stakeholders useful insights for informed decision-making in this quickly evolving sector. [ABSTRACT FROM AUTHOR]

In telecommunication industries, as customers have the alternative of deciding on from numerous telecom groups,they should effortlessly circulate from one provider to another. On this kind of an extremely aggressive marketplace, telecom organizations have an average client churn price of 10-20% every yr, that's drastically excessive. as the expenses of acquiring a new patron is simply too a good deal extra in comparison to that of retaining a cutting-edge client, retaining high worthwhile customers is a critical enterprise cause. Due to the direct impact on the income of the businesses, they are looking for out a means to expect customers who are expected to depart. So, this paper, the statistics of customers of a telecom organization is tested, and predictive models have been made to recognize the clients at high hazard of churn. [ABSTRACT FROM AUTHOR]

The problem of water pollution become more prominent due to its accelerated industrialization and development of society. Environmental management department has established sequences of pollutant discharge standard for preventing real time water pollution to stop the diffusion and gathering of harmful substance in water bodies, resulting in more serious environmental problems and further deterioration of water quality. Common testing method is the TDS (total dissolved solids) value testing and colorimetric methods that is typically through water bodies that comprise salt, acid, alkali, and further indications of the concentration test, for producing a water quality prediction. But the classical techniques of water quality testing, whether in the accuracy of the test or in the measurement time, are some drawbacks. The treatment and timely response of extremely polluted water bodies are essential to rapidly identify the water quality indicator concentration in water bodies. This study designs a new Quantum Cat Swarm Optimization with Micro Neural Network for Water Quality Index Prediction and Classification (QCSOMNN-WQIC) technique. The presented QCSOMNN-WQIC technique follows machine learning based classification with hyperparameter tuning process for WQI prediction. To accomplish this, the QCSOMNN-WQIC technique undergoes data normalization process at the initial stage. Next, the classification of WQI occurs by employing the MNN model, which categorizes the WQI into several classes. The QCSO method is utilized to improve the classification achievement of the MNN model. A sequence of experiments was performed to demonstrate the superior classifier outputs of the QCSOMNN-WQIC technique. The outputs identified improved achievement of the QCSOMNN-WQIC technique over other recent approaches. [ABSTRACT FROM AUTHOR]

The paper presents the modelling and assessment of a road embankment body after different construction phases using the finite element method (FEM) and the limit equilibrium method (LEM) for slope stability assessment. Different material models of soil (Mohr-Coulomb, Modified Mohr-Coulomb and Drucker-Prager) were used for modelling and assessment by FEM method. The deformation of the subsoil and the embankment was determined for each construction phases and the stability of the embankment slopes was assessed. The results obtained were compared with respect to the soil material model used as well as with respect to the calculation method used. In conclusion, it was found that it is more appropriate to use non-linear material models for stability calculations. [ABSTRACT FROM AUTHOR]

A Balinese script is one of the cultural relics of our ancestors and can be found in palm leaves. However, today Balinese script is abandoned along with the increasing use of the Roman script. Some efforts are needed for continuous socialization by developing the Balinese script learning media. We proposed the Balinese script identification using a machine learning approach. The backpropagation neural network algorithm is applied to create a predictive model recognizing Balinese scripts. Our study used 1001 Balinese character images extracted with the intensity of character algorithm, mark direction, and length and width of the character images methods. The result is promising. The training data accuracy is 90.31%, and the single testing data accuracy is 94%. The predictive model result shows the opportunity to develop a transliteration of Balinese script apps using the machine learning technique. [ABSTRACT FROM AUTHOR]

In the present era most of the people are effected with hypothyroid disease. People who are of age 13 and above are more effected with this disease and day by day it is transforming into a dangerous disease. The prediction of disease at earlier stage is very crucial so superior treatment is contributed by doctors. In this paper, hypothyroid disease prediction is implemented through Ensemble Machine Learning Algorithms and through RapidMiner Tool. The algorithms like Random Forest, AdaBoost, and Gradient Boosting and Bagging are implemented. These models were compared and evaluated using evaluation metrics like Accuracy, Precision, and Recall. As multiple models are combined so ensemble method is more robust and accurate. It removes unrelated data in the medical dataset and produces accurate and precise data. It proves that the ensemble method is very efficient than using a single classifying method. Hence implementing the ensemble method, we can predict the patient's disease efficiently. [ABSTRACT FROM AUTHOR]

The rate of penetration (ROP) prediction is carried out using a recorded drilling dataset from 02-Well. The prediction approaches use machine learning random forest regressor model and artificial neural network, especially the MLP regressor. The aim is to make the best machine learning model accurately predict the ROP parameter at 02-Well. The method used is designing the architecture of the machine learning model, which is divided into five stages: exploratory data analysis, data pre-processing, prediction and modeling using the selected algorithm, hyper-parameter tuning, and model evaluation. The 02-Well dataset would be divided into a 70% training set and a 30% test set as the base case. The model evaluation results show that modeling using a random forest regressor has a mean absolute percentage error (MAPE) score of 19.81%, which belongs to the "Good Forecasting" criteria. Meanwhile, modeling using MLP regressors has a MAPE score of 22.84% with the "Reasonable Forecasting" criteria. [ABSTRACT FROM AUTHOR]

This preliminary research article presents an alternative formulation of machine learning (ML) models for the prediction of the higher heating value (HHV) of biomass. The proposed method aims to improve the accuracy of the decision tree model (DT) while preserving its computational performance. It used 21 HHV prediction correlations augmented with the ML model formulation process rather than the sole use of raw experimental data as in a conventional approach. The method was also tested and compared with other two ML models, including support vector machine (SVM) and artificial neural network (ANN), with 148 experimental data sets. The test results showed that integration of the proposed method with DT had 2.72 times higher accuracy than the conventional model based on the R2 criterion with comparative computational performance. [ABSTRACT FROM AUTHOR]

The goal of this project is to use support vector clustering to enhance the performance of a coronary heart disease detection model in comparison to the aXgboost Classifier. Substances and Techniques: The Support Vector Clustering Method and Xgboost Classifier Algorithms are used for cardiac disease prediction. In this study, we split our participants evenly into two groups of five. Identification and forecasting are performed using a cardiac data set. There are 14 indicators in the dataset that could be used to determine if a patient has cardiac problems. Results: Two statistically reliable classifiers were used to produce the results: The observed significance level for accuracy is 0.002 (two-tailed) (p0.05). When it comes to cardiac disease prediction, Support Vector Clustering Technique appears to far outperform Xgboost Classifier. The xgboost Classifier Algorithm has an accuracy of 79.60 percent, while the Support Vector Clustering Method achieves an accuracy of 82.84 percent. [ABSTRACT FROM AUTHOR]

The purpose of this work is to contrast the effectiveness of the Logistic Regression algorithm with that of Novel Random Forest when predicting device failures. Two groups, alpha (0.05), power (80%), and the environment ratio are utilised in the G power analysis to determine the necessary sample size. Twenty thousand datasets are taken from Kaggle as a representative sample. There is a 75% training dataset and a 25% test dataset created from the collected samples (25 percent). For the purpose of measuring Novel Random Forest's efficacy, accuracy, precision, and sensitivity scores are computed. The model is significant at the (p 0.05) level. The Novel Random Forest model outperformed the Logistic Regression method by a wide margin, with an accuracy of 88.23%, precision of 87.43%, and sensitivity of 88.95%. The Logistic Regression method only managed 86.33% accuracy, 96.10% precision, and 83.30% sensitivity. It is determined that G power is 0. In this study, the Random Forest classifier outperforms the Logistic Regression classifier in terms of reducing costs for datasets that include failed devices. [ABSTRACT FROM AUTHOR]

Currently, the quality of the output product in the production of formalin is determined by the laboratory, which is an expensive, complex process and takes a long period of time. This paper examines the analysis of methods for constructing predictive models of an industrial technological facility using least squares (LSM) and regressions in order to increase the efficiency of safe process control. An approach to the construction of predictive models of the output product quality based on regression models is presented. It is shown that the results obtained do not contradict the existing methods of choosing regression modeling methods for constructing predictive models. An approach to the construction of a predictive model for decision-making on the operational management of multi-stage and multidimensional continuous technological processes of formalin is described. [ABSTRACT FROM AUTHOR]

Renewable energy generators are widely used in distribution networks to help meet consumption needs. Photovoltaic systems are common renewable generators, widely used in distribution networks and microgrids; however, due to the uncertainties of renewable generation, this leads to a possibility that the network may have problems in terms of generating the power required by the network at a given moment. To better control distribution networks, algorithms may thus be used to predict and model work conditions throughout the day and night. Model action control methods offer one of the more effective ways of controlling networks under uncertainties, which is important where, in addition to photovoltaic generation uncertainty, the network load is also subject to uncertainty. In this article, the optimal control of photovoltaic resources was done using the Model Predictive Control (MPC) method, a multivariable control method that offers dynamic control as opposed to direct control. The implementation of the predictive control controller of the model was done using MATLAB simulation of the microgrid, and the results highlight the advantages of the proposed method in terms of improving the total harmonic distortion indices and reducing rise time and settling time. Overall, the steady-state control method is confirmed to reduce steady-state errors. [ABSTRACT FROM AUTHOR]

Any water treatment plant (WTP) must use a reliable forecasting model to estimate treated water quality in order to establish a foundation for process control. The costs of operation and analysis are thus reduced, and the stability of WTP performance rendered evaluable. This paper's primary aim was thus to predict the Al Wahda WTP's treated water quality by means of an artificial neural network (ANN), using a linear neural network approach. The prediction model was developed using raw water quality data for seven years, from 2015 to 2021. The dataset for the study was obtained from the municipality of Baghdad for Al-Wahda WTP. Prediction models were then developed for turbidity, alkalinity, total hardness, pH, total dissolved solids, and electrical conductivity. The results showed that the ANN can accurately forecast the effluent parameters of water quality with the factor of determination (R2) between the variables of output predicted and those observed ranging from 0.65 to 0.99. As well as being a useful tool for comprehending and simulating water's nonlinear behaviours, the predicting ANN model can thus also serve as an excellent analysis and diagnosis tool. This should allow operators of WTPs to assist policy decision-makers by using established predictive models to report issues and potential for improvement. [ABSTRACT FROM AUTHOR]

This study aims to determine the best equation method and model for prediction unemployment in North Sumatra in 2022. This study uses dataset from the Central Statistics Agency (BPS) of North Sumatra where dataset is taken for 22 years from 2000 until 2021. The results of this study are the best parameters of each method were for the Double Moving Average method with an average time parameter of 3rd periods and the Double Exponential Smoothing method with alpha value = 0.5. The 3rd-period Double Moving Average accuracy test shows a MAPE value of 13.97% while Double Exponential Smoothing with alpha value=0.5 indicates a MAPE value of 16.45%. The study shows that the best forecast model for North Sumatra unemployment in 2022 is to use the 3rd period using the DMA method to forecast 2022 unemployment up to 519305 people. [ABSTRACT FROM AUTHOR]

Peat is problematic soil which has high compression and low bearing capacity. Soil improvement method for peat soil must pay attention to environmental issues because this soil easily releases greenhouse gas emissions. Preloading is one kind of environment friendly soil improvement method for peat. Conventional methods of settlement prediction could not be used for peat soil because one-dimensional time-compression curves of peat usually differ from those of inorganic clays. The peat compression curve shows short time of primary compression and significant long-term compression. Therefore, the long-term compression prediction model shall be verified. This paper presents a study of long-term compression prediction of tropical fibrous peat from Bereng Bengkel, Palangkaraya, Indonesia. The Lan model and Gibson and Lo model were chosen as the long-term compression prediction method. All model prediction will be verified by time-compression curve based on increment load and single load consolidation test in laboratory. The study result shows that the prediction of long-term compression is affected by the laboratory consolidation test method. The long-term compression prediction using Lan Model shows good agreement with laboratory test, especially on load increment consolidation test. [ABSTRACT FROM AUTHOR]

The world is now a field for perpetual rat race in every sphere of life and it has grown to be a space where about one thirds of the population suffers from anxiety disorder, proving it to be the most prevalent mental disorder. Keeping all concerns and the greatly developing technologies in mind, this study uses an intelligent analytics method for the responses of the DASS 21 Questionnaire, to train a K-Nearest Neighbour(KNN) model to predict psychological disorders and works on increasing the accuracy of the model. Machine learning models are of immense significance to numerous clinical issues as it may be helpful in diagnosing an illness, to analyze its symptoms and to forecast the result of a diagnosis. K-Nearest Neighbors algorithm (KNN) is found to generate the best performance in predictive analysis of this study. Different models of KNN like 1-NN, 3-NN, 5-NN and 7-NN are examined for the model among which 5-NN was found to be the best. Average accuracy, precision, recall and f-score noted with 5-NN were 89.6, 91.3, 0.89, 0.8975 respectively. Also a latency of 4.4 seconds was found with 5-NN. In the process, Emotional process, linguistic style, temporal process were also utilized for different KNN techniques to improve the results and predictions of the model. [ABSTRACT FROM AUTHOR]

The agricultural industry is the most prominent sector for the economic growth of the country. Cotton is an important crop that brings out major returns to the economy. In the current era, smart agriculture technology is incorporated which includes various modern techniques such as sensors, artificial intelligence, and computer vision to digitalize the whole process. This digitalization through modern technology enhances the fast and intelligent productivity of crops. In this paper convolutional neuralnetwork-based methods have been used to predict the maturity of the cotton balls using image features of mature and immature images. Real-time maturity detection is a challenging task for cotton fiber because of the many constraints in natural light conditionsand environmental effects. The proposed scheme uses annotation to classify the cotton balls. The classification is labeled as matureand immature images then the annotated images are used for training the deep learning models Resnet 50 and VGG 16. The averageaccuracy achieved from these methods is 92.47% for Resnet 50 and 96.58% for VGG16. The loss of the model is 21.05% for Resnet50 and 14.91% for VGG16. [ABSTRACT FROM AUTHOR]

The article presents an algorithm for the optimal control generation of an unmanned aerial vehicle that provides flight along a trajectory in the presence of air obstacles. Control optimization is carried out using a model modified by the authors based on the spiral forecast equations with an analytical solution. [ABSTRACT FROM AUTHOR]

The article presents the experience of building a system for early prediction of learning success based on the Markov model. The simulation is based on data from the LMS and the solution of the Kolmogorov system of equations. The advantage of this model is that it makes it possible to solve the problems of managing the educational process in real time. The difference between the model proposed in the work and most works, in which learning success is considered from the point of view of predicting student dropouts, is the early prediction of learning success in a particular discipline. [ABSTRACT FROM AUTHOR]

Energy-water efficiency forecasting is important to ensure a reliable supply of water for future generations as well as reduce energy consumption. Uncertainty, fuzziness, and hesitation in the information about energy consumption and water production might affect the accuracy of predictions and the effectiveness of models. Forecasting inaccuracy and model efficiency may both be improved by using Intuitionistic Fuzzy Sets, which are able to deal with the uncertainty, fuzziness, and hesitancy inherent in using past data. However, there is no standard technique in intuitionistic fuzzy time forecasting that can handle uncertain data and be utilized to create forecasts that are significantly more accurate than classical time series forecasting. In this paper, the triangular membership function and the trapezoidal membership function of the Intuitionistic Fuzzy Set, which is used to manage the water industry, were compared. This study looks at which of these two shapes of membership functions can make more accurate predictions. This becomes a core procedure in transforming the crisp value into fuzzy values by recognizing uncertainties or variations in the crisp values, and the process is known as fuzzification. Three forecasting errors were used in order to justify the best shape, namely Absolute Percentage Forecasting Error (APFE), Mean Squared Forecasting Error (MSFE), and Mean Absolute Percentage Forecasting Error (MAPFE). The result reveals that the triangular membership function is more accurate compared to the trapezoidal membership function for imprecise data representation in intuitionistic fuzzy time series forecasting. [ABSTRACT FROM AUTHOR]

The prediction of future retweet counts for tweets shared on Twitter has been a topic of immense interest recently. Numerous models have been proposed for such prediction, with their accuracy being assessed using certain choices of evaluation metrics. Admittedly, the majority of predictive models involved have overlooked the problem on the use of theoretically optimal functionals as point predictions and resort to employing options which are more accessible like the predictive mean. This motivates our discussion wherein the practicality of using theoretically consistent functionals with respect to the evaluation metrics considered is put forth. We discuss how the median of order (–1) and harmonic median are optimal in theory relative to the mean and median absolute percentage errors respectively, followed by highlighting in contrast how predictive models extant in the literature may suggest otherwise. Specifically, using a Poisson model supported by a large corpus of Twitter data, our numerical experiments indicate that predictions based on different functionals derived from the predictive distribution do not vary materially across the different metrics used, although predictions stemming from the predictive mean are slightly yet consistently more accurate than those based on the other functionals. We further outline how consistent functionals can be obtained accordingly under the settings of more complex predictive models. [ABSTRACT FROM AUTHOR]

Predictive analytics is a new frontier sector of higher education in today's world of data science, similar to other businesses such as marketing, financial, fraud detection, and demographic trends. Predictive analytics can provide beneficial information to educators and potentially assist them in enhancing students' performance by analyzing historical data using a variety of approaches from data mining and machine learning. The e-learning practiced in today's education system are unfortunately cause the dropout rates among students. Dropouts may cause the big and negative issues for university system and the stakeholders as well. Based on the literature review, studies on machine learning and predictive analytics to improve student performance are still scarce in Malaysian higher education. Therefore, the objective of this quantitative research is to develop the best predictive model for predicting students' performance at Pahang Islamic University College using machine learning techniques such as Decision Tree, Random Forest, AdaBoost, and Gradient Boosting. Students who have taken the Business Statistics course from the years 2013 to 2021 will be the subjects of the study. Data retrieved through a Learning Management System were used. From the analysis that has been done, Random Forest is the best method to be used in the predictive model for students' final grades. [ABSTRACT FROM AUTHOR]

Many automotive companies are uncertain about factors that affect their customers' car purchasing decisions. This research aims to identify what factors that affect the customers' car purchasing decisions. At the same time, it will help many potential customers to own the car of their dream at a better price. Based on the data that has been collected from Kaggle, a predictive model based on survey analysis, correlation, and linear regression are suggested which can identify the factors that affect the car price. From the car price, comparisons can be made by the car manufacturer to identify what is needed and should be prioritized by potential customers. A car manufacturer can also gain insights into the price of the future car based on the factors that are prioritized by customers. The results indicate that certain factors have a strong relationship with the price and the suggested predictive model can determine whether the price is within an acceptable range for customers purchasing decisions. [ABSTRACT FROM AUTHOR]

Vehicular ad-hoc networks (VANETs) are widely contributing to the development of intelligent transport systems (ITS). VANETs are seeking to connect vehicles together to build services that are mainly appropriate for a vehicular environment. They try to do so with/without relying on Roadside Units (RSU) to assist the process of network topology management. To test or suggest a developed real application, a suitable vehicle mobility model must be proposed to be implemented in the tested simulation experiments. General Motor (GM) mobility model provided a realistic description of vehicular mobility at both macroscopic and microscopic levels. In this paper, in this paper, the GM car following model is modified based on one real traffic dataset (highD). The used dataset considers real vehicle traffic data. It is used in estimating the GM model parameter values. A developed prediction GM model is utilized in predicting driver behavior and its effect on the traffic environment. [ABSTRACT FROM AUTHOR]

Experimental measurements of the ear canal pressure (ECP) and cochlear microphonics at the round window (RWCM) in response to a pure tone exhibit quasi-periodic ripples in the amplitude. This fine structure, which is hypothesized to be due to a linear reflection mechanism caused by inhomogeneities in the properties of the cochlear partition, can be used to provide noninvasive information about cochlear function. In this work, a previously developed model of the gerbil cochlea is used to study the generation of the fine structure on the ECP and RWCM. The model includes cochlear roughness due to inhomogeneities in the properties of outer hair cells (OHCs) and a longitudinal cable model to represent the propagation of electrical currents in the cochlear ducts. The fine structure predicted in both ECP and RWCM is in line with experimental measurements in other species. Model simulations are used to derive the reflection components of the ECP and RWCM by taking the difference between predictions with a model with roughness and a smooth model. Results from simulations are analyzed to determine the cochlear regions that contribute to the RWCM fine structure. [ABSTRACT FROM AUTHOR]

Coronary heart disease (CHD), colloquially referred to as cardiovascular disease (CVD), is the leading cause of death worldwide. Each individual may experience different symptoms or none at all, which means they may be unaware they have CHD until they suffer from chest pain, a heart attack, or cardiac arrest. These circumstances may be averted if we can forecast the early diagnosis of heart disease and identify the disease's major risk factors. Currently, prediction accuracy is insufficient, and the most essential risk factors remain mysterious. This article addresses numerous risk factors for coronary heart disease and gives prediction models for the disease utilizing supervised machine learning algorithms, namely the Random Forest and XGBoost algorithms, as well as the Artificial Neural Network (ANN), a deep-learning-based method. It makes use of a publicly available dataset of coronary heart disease patients from the Cleveland database of the UCI. The methodology consists of the preprocessing steps, feature engineering which involves the correlations between features, data splitting with a ratio of 70:30, and model training. The trained models are further optimized using the Grid Search optimization algorithm. The parameters of the Random Forest model are the number of trees = 1000, the maximum depth of the tree = None, and the minimum samples split = 2. The parameters of the XGBoost model are learning rate = 0.5, maximum depth of the tree = 10, and the number of trees = 200. The ANN is built using ReLU activation function, 5 layers consisting of 26 nodes, a learning rate of 0.001, and 378 parameters in total, as well as an epoch of 50. The models are evaluated using accuracy, precision, recall, and F-measure numbers. The results show that the Random Forest, XGBoost, and ANN algorithms have accuracies of 81.11%, 82.22%, and 86.67%, respectively. Equally important, the results of the feature importance signify the importance of maximum heart rate and nuclear stress test in predicting the early diagnosis of the disease. [ABSTRACT FROM AUTHOR]

One-way traffic networks maybe force drivers to bypass blockages to reach their destinations, potentially increasing trip distances. However, it is beneficial for anticipating some traffic problems in most urban traffic conditions—only a few studies concerned the regression approach for predicting road network performance models. As a result, this research aims to use the regression method to forecast the performance model of the road network before and after one-way traffic system implementation. The derived model equations are compared to see which ones substantially impact the road network's performance. This study relies on secondary data from performance assessments of 26 roads before and after the system during morning peak hours of 06:30–07:30 AM. The independent variables (X) are data from segment analysis used, i.e., Volume (X1), Speed (X2), and Volume (X4), whereas the dependent variable (Y) is Level of Service. By attempting the model prediction multivariate linear regression modeling before the system, Y = 5.647 – 4.376X1+0.009X2+0.0003X4 tend better than others. It provided a more significant impact on LOS than other models provided by R-value 0.951 is higher than the other models. The R-square value is 0.904, which means that 90.4% of LOS is influenced by DS, V, and Q factors. Further research is needed to conduct in the form of additional other factors that can affect LOS. [ABSTRACT FROM AUTHOR]

Model predictive control is one of the advanced methods to solve trajectory following problems of autonomous ground vehicles due to its predictive capability. It can deal with multiple inputs, outputs, and set-point signals. However, this technique causes a heavy computational burden to the controller due to its multi-objective optimization approach, limiting its real-time applications. In this study, an explicit nonlinear model predictive controller (E-NMPC) is proposed to control front steering angle and rear wheel tractive torques simultaneously to provide trajectory tracking of a vehicle in NATO double lane change (DLC) maneuvers with consideration of reduced computation time. Artificial neural networks enabling offline learning processes are used to generate a computationally-efficient control law without going through extensive optimization methods. Namely, main contribution of this study is establishment of a neural-network based E-NMPC technique for the considered control problem that removes the need of online optimization and replace the classical NMPC that requires online-optimization with high computational load. The simulated results revealed that time elapsed during one execution cycle can be significantly reduced with the proposed method for various cases, compared to the classical online NMPC method. [ABSTRACT FROM AUTHOR]

The mechanical and thermal characteristics of the glasses are very crucial in order to select a glass for a particular application. In the present work, a mathematical model has been developed to predict the Vickers hardness (Hv) and glass transition temperature (Tg) of silicate glasses with alkali and alkaline earth oxides as network modifier. Single source data on glass properties has been used to understand their composition-property relationship. Multiple linear regression was employed to develop a model which takes selected composition of the glasses as input and gives an estimated value of Hv and Tg. Dependency of these properties of the glasses on the mol% of their constituent oxide has been explored. The formulated model was validated by obtaining the R2 value using the regression analysis between the calculated and the experimental values. Hv values of known compositions of glasses could be more reliably predicted and were close to the experimentally observed values. [ABSTRACT FROM AUTHOR]