Browse Items (2150 total)
Sort by:
-
Predicting of Credit Risk Using Machine Learning Algorithms
Credit risk management is one of the key processes for banks and is crucial to ensuring the banks stability and success. However, due to the need for more rigid forecasting models with strong mapping abilities, credit risk prediction has become challenging for the banking industry. Therefore, this paper attempts to predict commercial banks credit risk (CR) by using various machine learning algorithms. Machine learning algorithms, namely linear regression, KNN, SVR, DT, RF, XGB, and MLP, are compared with and without feature selection and feature extraction techniques to examine their prediction capabilities. Various determinants of credit risk (features) have been extracted to predict credit risk, and these features have been used to train machine learning models. Findings revealed that the decision tree algorithm had the highest performance, with the lowest mean absolute error (MSE) value of 0.1637 and the lowest root mean squared error (RMSE) value of 0.2158. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Predicting Player Engagement in Online Gaming: A Machine Learning Approach
The aim of this research is to make precise forecasts on player participation in online game using state-of-the-art machine learning algorithms. Player engagement plays a crucial element in determining the success of online games because it affects player retention, satisfaction and monetization. By understanding and predicting engagement levels, game developers and marketers can enhance the gaming experience and develop strategies to keep players invested. This research involves a comprehensive analysis of player behavior data from an online gaming platform. The dataset includes various demographic and behavioral features such as age, gender, location, game genre, playtime hours, in-game purchases, game difficulty, sessions per week, average session duration, player level, achievements unlocked, and engagement level. The data was preprocessed through handling missing values, normalizing numerical features, and encoding categorical variables. Exploratory Data Analysis (EDA) was conducted to understand the distribution and relationships between different features. Multiple machine learning models were evaluated to predict player engagement levels, including Random Forest, Gradient Boosting, XGBoost, and Support Vector Machine (SVM). These models were then compared through the accuracy, precision, recall, and F1-score metrics. In the comparison, XGBoost emerged as the best model. Since it is the best-performing model, we can make the feature importance analysis to identify the best factors for predicting engagement in the next step. The XGBoost model achieved the highest accuracy of 91%, demonstrating superior precision, recall, and F1-scores across all engagement levels (High, Medium, Low). Ensemble methods like XGBoost, Gradient Boosting, and Random Forest outperformed the SVM model, highlighting their effectiveness in handling complex datasets. 2024 IEEE. -
Predicting Price Direction of Bitcoin based on Hybrid Model of LSTM and Dense Neural Network Approach
Bitcoin is a rapidly growing but extremely risky cryptocurrency. It marks a watershed moment in the history of cash. These days, digital currency is preferred to actual money. Bitcoin has decentralized authority and placed it in the hands of its users. Many people are joining the largest and most well-known Bitcoin mining pools as the risk of working alone is too great. In order to enhance their chances of creating the next block in the Bitcoins blockchain and decrease the mining reward volatility, users can band together to form Bitcoin pools. This tendency toward consolidation may also be seen in the rise of large-scale mining farms equipped with powerful mining resources and speedy processing capability. Because of the risk of a 51% assault, this pattern shows that Bitcoin's pure, decentralized protocol is moving toward greater centralization in its distribution network. Not to be overlooked is the resulting centralization of the bitcoin network as a result of cloud wallets making it simple for new users to join. Because of the easily hackable nature of Bitcoin technologies, this could lead to a wide range of security vulnerabilities. The proposed approach uses normalization and filling missing values in preprocessing, PCA for feature Extraction and finally training the model using LSTM-DNN Models. The proposed approach outperforms other two models such as CNN and DNN. 2023 IEEE. -
Predicting Song Popularity Using Data Analysis
In today's music landscape, predicting a song's success is crucial for musicians, record labels, and streaming platforms. This paper introduces a methodology for estimating popularity using Spotify data, termed the 'Proxy Popularity Score.' Three models - Random Forest, LightGBM Regressor, and XGBoost Regressor - are utilized for prediction. Performance metrics including mean absolute error, mean squared error, root mean squared error, and R-squared error are employed to evaluate model accuracy. Correlation values of 99.85%, 99.87%, and 99.84% are achieved for XGBoost, LightGBM, and Random Forest respectively. The study concludes with a ranking of songs based on predicted popularity scores. 2024 IEEE. -
Predicting Stock Market Price Movement Using Machine Learning Technique: Evidence from India
The stock market is uncertain, volatile, and multidimensional. Stock prices have been difficult to predict since they are influenced by a variety of factors. In order to make critical investment and financial decisions, investors and analysts are interested in predicting stock prices. Predicting a stock's price entails developing price pathways that a stock might take in the future. ANN and mathematical Geometric Brownian movement technique were employed in this study to forecast a stock market closing price of Indian companies. The comparative analysis indicates that the Geometric Brownian Method is better than ANN in giving better MAPE and RMSE Values. 2022 IEEE. -
Predicting Stock Market Trends: Machine Learning Approaches of a Possible Uptrend or Downtrend
This paper delves into a statistical analysis of the stock market, emphasizing the significance of accuracy in stock predictions. Large data sets can be handled by machine learning algorithms, which can also forecast outcomes based on past data and spot intricate patterns in financial data. They assist control risks, automate decision-making procedures, and adjust to changing circumstances. Multi-source data can be combined by ML models to provide a comprehensive picture of market circumstances. They can manage intricate, nonlinear interactions, provide impartial analysis, and lessen human bias. Models are able to adjust to shifting market conditions through ongoing learning and retraining. They must, however, exercise caution when deploying models in real-world situations and ensure that they are validated. Although machine learning has advantages for stock market analysis, it must be carefully evaluated for dangers and validated before being used in practical situations. The traditional machine learning model, Logistic Regression has been used in order to predict stock prices. It focuses on binary classification based on the trend of the stock. Through the model training and evaluation and additional analysis done on the results, this research contributes towards obtaining predictions and studying reasons of a possible uptrend or downtrend to further assist companies. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Predicting the Cerebral Blood Flow Change Condition during Brain Strokes using Feature Fusion of FMRI Images and Clinical Features
By fusing clinical information with functional magnetic resonance imaging (fFMRI) pictures, this study describes a novel method for predicting changes in cerebral blood flow during brain strokes. The FMRI data and patient-specific variables, such as age, gender, and medical history, are combined via feature fusion in the proposed technique. As a result, the model developed can accurately forecast changes in cerebral blood flow that occur during brain strokes. The efficiency of the suggested strategy is shown by experimental findings. The performance of the model is greatly enhanced when FMRI data and clinical characteristics are combined as opposed to just one data source. The findings of this study have important ramifications for increasing the accuracy of stroke diagnosis and treatment and, eventually, for bettering patient outcomes. The experimental results showed that the proposed method a high level of accuracy in predicting changes in cerebral blood flow after brain strokes. The performance of the model was much enhanced by combining clinical characteristics with FMRI data as opposed to using only one of these data sources. This emphasizes the value of including pertinent clinical information in the diagnosis and management of stroke. 2023 IEEE. -
Predicting the Stock Markets Using Neural Network with Auxiliary Input
Predicting the stock market has always been a challenging task and has always had a certain appeal for researchers all around the world. Stock markets are supposed to be quite random and people with experience in the market strongly agree to the fact. Thus, predicting the stock market accurately paves the way for endless money. To date, no such algorithm has been devised that could even predict the stock market with a 90% accuracy rate. The difficulty lies in the randomness of the markets, and the various complexities involved in modeling market dynamics. Nevertheless, there have been algorithms with a decent success rate and researchers around the world have been in a constant attempt to improve over them. Thus, through this paper we attempt at predicting the return of a stock over a period of 10days after a particular news was out regarding the stock using the headlines of the news and certain other features important in determining the direction of a stock. The model was implemented with a sigma score of 0.81. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Predicting the Thyroid Disease Using Machine Learning Techniques
An endocrine gland that is allocated in the front of the neck is called the thyroid, which produces thyroid hormones as its main job. Thyroid hormone may be produced insufficiently or excessively as a result of its potential malfunction. There are various thyroid types including Hyperthyroidism, Hypothyroidism, Thyroid Cancer Thyroiditis, swelling of the thyroid. A goiter is an enlarged thyroid gland. When your thyroid gland produces more thyroid hormones than your body requires, you have hyperthyroidism. When the thyroid gland in our body doesnt provide enough thyroid hormones, then our body has hypothyroidism; when you have euthyroid sick, your thyroid function tests during critical illness taken in an inpatient or intensive care setting show alterations. Hypothyroid, hyperthyroid, and euthyroid conditions are expected from these thyroid conditions. The Three similarly used machine learning algorithms are: Support Vector Machine (SVM), Logistic Regression, and Random Forest methods, were evaluated from among the various machine learning techniques to forecast and evaluate their performance in terms of accuracy. Random forest can perform both regression and classification tasks. Logistic Regression is used to calculate or predict the probability of a binary (yes/no) event occurring. SVM classifiers offers great accuracy and work well with high dimensional space. A thyroid data set from Kaggle is used for this. This study has demonstrated the use of SVM, logistic regression, and random forest as classification tools, as well as the understanding of how to forecast thyroid disease. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Predicting Wind Energy: Machine Learning from Daily Wind Data
This paper offers a comprehensive review of the advancements in the realm of renewable energy, specifically focusing on solid oxide fuel cells and electrolysers for green hydrogen production. The review delves into the significance of wind energy as a pivotal renewable energy source and underscores the importance of precise forecasting for efficient energy management and distribution. The integration of machine learning-based approaches, such as Support Vector Regression and Random Forest Regression, has shown promising results in enhancing the accuracy of wind energy production forecasts. Furthermore, the paper explores the broader landscape of renewable energy generation forecasting, emphasizing the rising prominence of machine learning and deep learning techniques. As the penetration of renewable energy sources into the electricity grid intensifies, the need for accurate forecasting becomes paramount. Traditional methods, while valuable, have encountered limitations, paving the way for advanced algorithms capable of deciphering intricate data relationships. The review also touches upon the inherent challenges and prospective research avenues in the domain, including addressing uncertainties in renewable energy generation, ensuring data availability, and enhancing model interpretability. The overarching goal remains the seamless integration of renewable sources into the grid, propelling us towards a greener future. The Authors, published by EDP Sciences, 2024. -
Predicting Work Environment and Job Environment Among Employees using Transfer Learning Approach
Today's enterprises face numerous challenges as a result of the world's rapid evolution. Maintaining a content workforce is crucial to a company's success and survival in today's fast-paced business environment. The efficacy, productivity, efficiency, and dedication of the company's staff are directly associated with the company's capacity to meet the needs of its employees in the workplace. The focus of this system is to identify the factors that contribute to a satisfying work environment for the participants. Preprocessing, feature selection, and model training are the first three steps in the suggested methodology. Data mining systems should get in the habit of normalizing data as a preliminary processing step. The multiple elements assessing company culture and worker satisfaction were consolidated using Principal Components Analysis (PCA) in the feature selection phase. Once features have been selected, KNN-SVM is utilized for model training. When compared to the two most popular alternatives, SVM and KNN, the proposed technique performs better. 2023 IEEE. -
Prediction of Depression in Young Adults Using Supervised Learning Algorithm
Over the years, mental health has achieved an essential role in the pertinent development of a human being, and a large part of the population is affected by it. The most commonly affected community being college-going students, and the most common disorders being Anxiety and Depression. Depression is a leading cause of suicide in individuals, where suicide is the second most prevailing reason for death among 1529-year-olds. This study aims to identify the different reasons and other factors associated with depression to predict and determine whether an individual faces depressive disorders. For this research purpose, the most appropriate classifier is selected. The absolute accuracy of the proposed model is 91.17%, i.e., the model can correctly predict whether an individual has depression 91.17% of the time. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Prediction of football players performance using machine learning and deep learning algorithms
In modern days the margin of error for football game is low, therefore the ultimate aim of the game is to win the match. The performance of the players in the match affects the results of the game. Due to this it is very important to evaluate the player and know his weakness. Manual evaluation tends to generate many errors and take more time. In the current research the statistical model is proposed to predict the stats of the football player based on previous session data by considering various aspects of the game. Through literature reviews it is observed that machine learning and deep learning algorithms can be used predict the performance of football player. But which model would be more efficient considering the positions of the player is not considered in any article. The proposed model has designed separate model as per the position of the player during the game. This can help to predict the player's performance as per their playing position. The current study has successfully implemented various machine learning and deep learning models and provide comparative analysis of the same. Each position has considered different variables associated with that position. The performance of these models is compared for further clarification 2021 IEEE. -
Prediction of Friction Stir Welding Parameters Using Response Surface Methodology
The Friction Stir Welding (FSW) technique results in mixing and densification of weld joint in a more accurate and localized manner. FSW has been used to create a more significant weld with more structural integrity. In this research work, to join AA 3103 and AA 7075 was carried out. These alloys were preferred due to their wide variety of applications varying from aluminium fabrication to the aerospace industry. AA 7075, being a costlier metal, can be partially replaced with AA 3103, which can be economically justifiable for this research. The study tries to reveal the regression model by considering the FSW parameters like speed feed and offset. Various mechanical tests, impact tests and hardness tests were used for determining the most suitable weld joint. After conducting the tests, the results were analyzed using Minitab 18 software. The mathematical equations were derived out of Response Surface Methodology, which proved to be efficient. The report thus discusses the details in the analysis and study of FSW. 2023 American Institute of Physics Inc.. All rights reserved. -
Prediction of Hazardous Asteroids Using Machine Learning
As the need for early detection and mitigation of potential threats from near-Earth objects continues to grow, this study presents a comprehensive approach to predicting hazardous asteroids through the application of machine learning techniques. With the increasing interest in safeguarding our planet from potential impact events, the accurate classification and prediction of hazardous asteroids is of paramount importance. This research leverages a diverse dataset comprising a wide array of asteroid characteristics, including orbital parameters, physical properties, and historical impact data, to train and validate machine learning models. The study employs a combination of feature engineering, data preprocessing, and state-of-the-art machine learning algorithms to assess the risk posed by asteroids in near-Earth space. 2024 IEEE. -
Prediction of heart disease using XGB classifier
Predicting heart disease in advance could be a significant medical breakthrough because it is widespread. A reliable strategy that can be utilized to do this is machine learning. Decision tree classifiers, random forests, and multilayer perceptron have all been used in studies to predict heart disease. However, several of these techniques could be improved, like poor precision. In our research, we have taken the South African heart Disease dataset and implemented a few models, which include Support Vector Machine (SVM), K Neighbors (KNN), Artificial neural network and XG Boost Classifier. We have used different methods for measuring performance. SVM with 69.0 accuracy, KNN with 86.0 accuracy, and ANN with 80.0 accuracy. However, the XGB classifier has shown some promising results in predicting heart disease with an accuracy of 90%. Further, when the hyperparameters were tuned using the random search method, the accuracy increased to 92.8%. The benefit of this work is that it uses machine-learning approaches to enhance the performance of coronary heart disease prediction. 2024 Author(s). -
Prediction of Rainfall Using Seasonal Auto Regressive Integrated Moving Average and Transductive Long Short-Term Model
One of the most crucial parts of the practical application in recent years has been the analysis of time series data for forecasting. Because of the extreme climate variations, it is now harder than ever to estimate rainfall accurately. It is possible to forecast rainfall using a number of time series models that uncover hidden patterns in past meteorological data. Choosing the right Time Series Analysis Models for predicting is a challenging task. This study suggests using a Seasonal Auto Regressive Integrated Moving Average (SARIMA) to forecast values that are similar to historical values that exhibit seasonal patterns. Twelve years of historical weather data for the city of Lahore (from 2005 to 2017) and Blora Regency are taken into account for the prediction. The dataset underwent pre-processing operations like cleaning and normalisation before to the classification procedure. For classification, Transductive Long Short-Term Model (TLSTM) is employed which has learned the dependency values where the memory blocks are recurring and capable of learning long-term dependencies on this model. Further, TLSTM's goal is to increase accuracy close to the test point, where test points are selected as a validation group. The performance of the models has been assessed based on accuracy (99%), precision (98%), recall (96%) and fl-score (98%). Proposed SARIMA model showed optimistic results when compared to existing models. 2023 IEEE. -
Prediction of Stock Prices using Prophet Model with Hyperparameters tuning
As part of the data analytical process, predicting and time - series are crucial. In academics and financial research, anticipating share prices is a prominent and significant subject. A share market would be an uncontrolled environment for anticipating shares since there are no fundamental guidelines for evaluating or anticipating share prices there. As a result, forecasting share prices is a difficult time-series issue. fundamental, technical, time series predictions and analytical strategies are just a few of the various techniques and approaches that machine learning uses to execute stock value predictions. This article implements the stock price prediction, Researchers compared the model of the prophet with the tuned model of the prophet. By utilizing the tuning of hyperparameters using parameter grid search to improve the performance of the model accuracy for the best prediction. The findings of the study demonstrated that tuned model of the prophet with hyperparameters tuning which results in model accuracy and based on the experimental findings mean squared error (MSE) and mean absolute percentage error (MAPE) has significant improvement. 2022 IEEE. -
Predictive Analysis of Academic Performance Among Students using A-CNN-BiLSTM Approach
The number of possibilities to analyze educational data using data mining techniques is expanding, with the goal of improving learning outcomes. There is an explosion in data produced by online and virtual education, e-learning platforms, and institutional IT. Using these statistics, teachers could gain valuable insights into their students' learning habits. Academic performance of students and other useful information can be analyzed with the help of educational data mining. Model training consists of three primary steps: data preprocessing, feature selection, and training the model. To eliminate unwanted problems like noise and redundant attributes, data preparation is necessary. By prioritizing which features to calculate, the mRMR algorithm lowers calculation costs. Feature selection plays a crucial role in training A-CNN-BiLSTM models. The suggested approach routinely outperforms BiLSTM and CNN, two state-of-the-art algorithms. With a data accuracy percentage of 96.57%, it's clear that there was a significant improvement. 2024 IEEE. -
Predictive Analysis of Sleep Disorders Using Machine Learning: A Comprehensive Analysis
The diagnosis of sleep disorders often relies on subjective patient reports, sleep diaries, and potentially cumbersome polysomnography (PSG) tests. However, these methods have limitations such as subjectivity, sleep diaries require meticulous effort, and expensive PSG tests are expensive, resource-intensive, and may not accurately capture sleep patterns in a non-clinical setting. Sleep disorders pose significant health risks and can impair overall well-being. Predictive analysis plays a crucial role in identifying individuals at risk of developing sleep disorders, enabling timely interventions and personalized treatment plans. In this paper, a comparative analysis of regression and classification models for sleep disorders prediction using machine learning (ML) techniques on insomnia and sleep apnea are discussed. Through extensive experimentation and comparative analysis, XGBoost and AdaBoost demonstrated as the most effective predictive models for insomnia and sleep apnea. AdaBoost and XGBoost classifiers are displaying 93.49% and 92.73% respectively. It is therefore possible to draw the conclusion that AdaBoost and XGBoost are doing well based on the findings as a whole, as indicated by the results. Our findings contribute to advancing the understanding and application of ML techniques in sleep disorder prediction, paving the way for more accurate and timely diagnosis based on ML techniques and personalized interventions in clinical practices. 2024 IEEE.