Search results for: prediction model accuracy

Authors: Prashant Verma

Abstract:

Commercial banks are facing immense pressure, including financial disintermediation, interest rate volatility and digital ways of finance. Retaining an existing customer is 5 to 25 less expensive than acquiring a new one. This paper explores customer churn prediction, based on various statistical & machine learning models and uses under-sampling, to improve the predictive power of these models. The results show that out of the various machine learning models, Random Forest which predicts the churn with 78% accuracy, has been found to be the most powerful model for the scenario. Customer vintage, customer’s age, average balance, occupation code, population code, average withdrawal amount, and an average number of transactions were found to be the variables with high predictive power for the churn prediction model. The model can be deployed by the commercial banks in order to avoid the customer churn so that they may retain the funds, which are kept by savings bank (SB) customers. The article suggests a customized campaign to be initiated by commercial banks to avoid SB customer churn. Hence, by giving better customer satisfaction and experience, the commercial banks can limit the customer churn and maintain their deposits.

Keywords: savings bank, customer churn, customer retention, random forests, machine learning, under-sampling

Procedia PDF Downloads 117

Authors: Bitewulign Mekonnen

Abstract:

Context: This scientific paper focuses on the use of near-infrared (NIR) spectroscopy to determine glucose concentration in aqueous solutions accurately and rapidly. The study compares six different machine learning methods for predicting glucose concentration and also explores the development of a deep learning model for classifying NIR spectra. The objective is to optimize the detection model and improve the accuracy of glucose prediction. This research is important because it provides a comprehensive analysis of various machine-learning techniques for estimating aqueous glucose concentrations. Research Aim: The aim of this study is to compare and evaluate different machine-learning methods for predicting glucose concentration from NIR spectra. Additionally, the study aims to develop and assess a deep-learning model for classifying NIR spectra. Methodology: The research methodology involves the use of machine learning and deep learning techniques. Six machine learning regression models, including support vector machine regression, partial least squares regression, extra tree regression, random forest regression, extreme gradient boosting, and principal component analysis-neural network, are employed to predict glucose concentration. The NIR spectra data is randomly divided into train and test sets, and the process is repeated ten times to increase generalization ability. In addition, a convolutional neural network is developed for classifying NIR spectra. Findings: The study reveals that the SVMR, ETR, and PCA-NN models exhibit excellent performance in predicting glucose concentration, with correlation coefficients (R) > 0.99 and determination coefficients (R²)> 0.985. The deep learning model achieves high macro-averaging scores for precision, recall, and F1-measure. These findings demonstrate the effectiveness of machine learning and deep learning methods in optimizing the detection model and improving glucose prediction accuracy. Theoretical Importance: This research contributes to the field by providing a comprehensive analysis of various machine-learning techniques for estimating glucose concentrations from NIR spectra. It also explores the use of deep learning for the classification of indistinguishable NIR spectra. The findings highlight the potential of machine learning and deep learning in enhancing the prediction accuracy of glucose-relevant features. Data Collection and Analysis Procedures: The NIR spectra and corresponding references for glucose concentration are measured in increments of 20 mg/dl. The data is randomly divided into train and test sets, and the models are evaluated using regression analysis and classification metrics. The performance of each model is assessed based on correlation coefficients, determination coefficients, precision, recall, and F1-measure. Question Addressed: The study addresses the question of whether machine learning and deep learning methods can optimize the detection model and improve the accuracy of glucose prediction from NIR spectra. Conclusion: The research demonstrates that machine learning and deep learning methods can effectively predict glucose concentration from NIR spectra. The SVMR, ETR, and PCA-NN models exhibit superior performance, while the deep learning model achieves high classification scores. These findings suggest that machine learning and deep learning techniques can be used to improve the prediction accuracy of glucose-relevant features. Further research is needed to explore their clinical utility in analyzing complex matrices, such as blood glucose levels.

Keywords: machine learning, signal processing, near-infrared spectroscopy, support vector machine, neural network

Procedia PDF Downloads 64

Authors: S. Channgam, A. Sae-Tang, T. Termsaithong

Abstract:

In this research, the occurrences of large size events in various system sizes of the Bak-Tang-Wiesenfeld sandpile model are considered. The system sizes (square lattice) of model considered here are 25×25, 50×50, 75×75 and 100×100. The cross-correlation between the ratio of sites containing 3 grain time series and the large size event time series for these 4 system sizes are also analyzed. Moreover, a prediction method of the large-size event for the 50×50 system size is also introduced. Lastly, it can be shown that this prediction method provides a slightly higher efficiency than random predictions.

Keywords: Bak-Tang-Wiesenfeld sandpile model, cross-correlation, avalanches, prediction method

Procedia PDF Downloads 356

Authors: C. Manjula, Lilly Florence

Abstract:

Software technology is developing rapidly which leads to the growth of various industries. Now-a-days, software-based applications have been adopted widely for business purposes. For any software industry, development of reliable software is becoming a challenging task because a faulty software module may be harmful for the growth of industry and business. Hence there is a need to develop techniques which can be used for early prediction of software defects. Due to complexities in manual prediction, automated software defect prediction techniques have been introduced. These techniques are based on the pattern learning from the previous software versions and finding the defects in the current version. These techniques have attracted researchers due to their significant impact on industrial growth by identifying the bugs in software. Based on this, several researches have been carried out but achieving desirable defect prediction performance is still a challenging task. To address this issue, here we present a machine learning based hybrid technique for software defect prediction. First of all, Genetic Algorithm (GA) is presented where an improved fitness function is used for better optimization of features in data sets. Later, these features are processed through Decision Tree (DT) classification model. Finally, an experimental study is presented where results from the proposed GA-DT based hybrid approach is compared with those from the DT classification technique. The results show that the proposed hybrid approach achieves better classification accuracy.

Keywords: decision tree, genetic algorithm, machine learning, software defect prediction

Procedia PDF Downloads 310

Authors: Belkacem Selma, Boumediene Selma, Tourkia Guerzou, Abbes Labdelli

Abstract:

Natural disasters have occurred and will continue to cause human and material damage. Therefore, the idea of "preventing" natural disasters will never be possible. However, their prediction is possible with the advancement of technology. Even if natural disasters are effectively inevitable, their consequences may be partly controlled. The rapid growth and progress of artificial intelligence (AI) had a major impact on the prediction of natural disasters and risk assessment which are necessary for effective disaster reduction. The Earthquakes prediction to prevent the loss of human lives and even property damage is an important factor; that is why it is crucial to develop techniques for predicting this natural disaster. This present study aims to analyze the ability of artificial neural networks (ANNs) to predict earthquakes that occur in a given area. The used data describe the problem of high energy (higher than 10^4J) seismic bumps forecasting in a coal mine using two long walls as an example. For this purpose, seismic bumps data obtained from mines has been analyzed. The results obtained show that the ANN with high accuracy was able to predict earthquake parameters; the classification accuracy through neural networks is more than 94%, and that the models developed are efficient and robust and depend only weakly on the initial database.

Keywords: earthquake prediction, ANN, seismic bumps

Procedia PDF Downloads 104

Authors: Maryam Kheirollahpour, Mahmoud Danaee, Amir Faisal Merican, Asma Ahmad Shariff

Abstract:

Background: The presence of nonlinearity among the risk factors of eating behavior causes a bias in the prediction models. The accuracy of estimation of eating behaviors risk factors in the primary prevention of obesity has been established. Objective: The aim of this study was to explore the potential of a hybrid model of structural equation modeling (SEM) and Artificial Neural Networks (ANN) to predict eating behaviors. Methods: The Partial Least Square-SEM (PLS-SEM) and a hybrid model (SEM-Artificial Neural Networks (SEM-ANN)) were applied to evaluate the factors affecting eating behavior patterns among university students. 340 university students participated in this study. The PLS-SEM analysis was used to check the effect of emotional eating scale (EES), body shape concern (BSC), and body appreciation scale (BAS) on different categories of eating behavior patterns (EBP). Then, the hybrid model was conducted using multilayer perceptron (MLP) with feedforward network topology. Moreover, Levenberg-Marquardt, which is a supervised learning model, was applied as a learning method for MLP training. The Tangent/sigmoid function was used for the input layer while the linear function applied for the output layer. The coefficient of determination (R²) and mean square error (MSE) was calculated. Results: It was proved that the hybrid model was superior to PLS-SEM methods. Using hybrid model, the optimal network happened at MPLP 3-17-8, while the R² of the model was increased by 27%, while, the MSE was decreased by 9.6%. Moreover, it was found that which one of these factors have significantly affected on healthy and unhealthy eating behavior patterns. The p-value was reported to be less than 0.01 for most of the paths. Conclusion/Importance: Thus, a hybrid approach could be suggested as a significant methodological contribution from a statistical standpoint, and it can be implemented as software to be able to predict models with the highest accuracy.

Keywords: hybrid model, structural equation modeling, artificial neural networks, eating behavior patterns

Procedia PDF Downloads 123

Authors: Ajayi Olusola Olajide, Alonge Olaide Moses

Abstract:

Predicting the outcome of soccer matches poses an interesting challenge for which it is realistically impossible to successfully do so for every match. Despite this, there are lots of resources that are being expended on the correct prediction of soccer matches weekly, and all over the world. Soccer Match Result Prediction System Model (SMRPSM) is a system that is proposed whereby the results of matches between two soccer teams are auto-generated, with the added excitement of giving users a chance to test their predictive abilities. Soccer teams from different league football are loaded by the application, with each team’s corresponding manager and other information like team location, team logo and nickname. The user is also allowed to interact with the system by selecting the match to be predicted and viewing of the results of completed matches after registering/logging in.

Keywords: predicting, soccer match, outcome, soccer, matches, result prediction, system, model

Procedia PDF Downloads 470

Authors: O. Behar, A. Khellaf, K. Mohammedi, S. Ait Kaci

Abstract:

Up to now, most validation studies have been based on the MBE and RMSE, and therefore, focused only on long and short terms performance to test and classify solar radiation models. This traditional analysis does not take into account the quality of modeling and linearity. In our analysis we have tested 22 solar radiation models that are capable to provide instantaneous direct and global radiation at any given location Worldwide. We introduce a new indicator, which we named Global Accuracy Indicator (GAI) to examine the linear relationship between the measured and predicted values and the quality of modeling in addition to long and short terms performance. Note that the quality of model has been represented by the T-Statistical test, the model linearity has been given by the correlation coefficient and the long and short term performance have been respectively known by the MBE and RMSE. An important founding of this research is that the use GAI allows avoiding default validation when using traditional methodology that might results in erroneous prediction of solar power conversion systems performances.

Keywords: solar radiation model, parametric model, performance analysis, Global Accuracy Indicator (GAI)

Procedia PDF Downloads 326

Authors: Salman Mohamadi, Seyed Mohammad Ali Tayaranian Hosseini, Hamidreza Amindavar

Abstract:

In this paper, we provide a procedure to analyze and model EEG (electroencephalogram) signal as a time series using ARIMA-GARCH to predict an epileptic attack. The heteroskedasticity of EEG signal is examined through the ARCH or GARCH, (Autore- gressive conditional heteroskedasticity, Generalized autoregressive conditional heteroskedasticity) test. The best ARIMA-GARCH model in AIC sense is utilized to measure the volatility of the EEG from epileptic canine subjects, to forecast the future values of EEG. ARIMA-only model can perform prediction, but the ARCH or GARCH model acting on the residuals of ARIMA attains a con- siderable improved forecast horizon. First, we estimate the best ARIMA model, then different orders of ARCH and GARCH modelings are surveyed to determine the best heteroskedastic model of the residuals of the mentioned ARIMA. Using the simulated conditional variance of selected ARCH or GARCH model, we suggest the procedure to predict the oncoming seizures. The results indicate that GARCH modeling determines the dynamic changes of variance well before the onset of seizure. It can be inferred that the prediction capability comes from the ability of the combined ARIMA-GARCH modeling to cover the heteroskedastic nature of EEG signal changes.

Keywords: epileptic seizure prediction , ARIMA, ARCH and GARCH modeling, heteroskedasticity, EEG

Procedia PDF Downloads 381

Authors: Ajai Singh

Abstract:

Simulation of hydrological processes at the watershed outlet through modelling approach is essential for proper planning and implementation of appropriate soil conservation measures in Damodar Barakar catchment, Hazaribagh, India where soil erosion is a dominant problem. This study quantifies the parametric uncertainty involved in simulation of stream flow using Soil and Water Assessment Tool (SWAT), a watershed scale model and Radial Basis Neural Network (RBNN), an artificial neural network model. Both the models were calibrated and validated based on measured stream flow and quantification of the uncertainty in SWAT model output was assessed using ‘‘Sequential Uncertainty Fitting Algorithm’’ (SUFI-2). Though both the model predicted satisfactorily, but RBNN model performed better than SWAT with R2 and NSE values of 0.92 and 0.92 during training, and 0.71 and 0.70 during validation period, respectively. Comparison of the results of the two models also indicates a wider prediction interval for the results of the SWAT model. The values of P-factor related to each model shows that the percentage of observed stream flow values bracketed by the 95PPU in the RBNN model as 91% is higher than the P-factor in SWAT as 87%. In other words the RBNN model estimates the stream flow values more accurately and with less uncertainty. It could be stated that RBNN model based on simple input could be used for estimation of monthly stream flow, missing data, and testing the accuracy and performance of other models.

Keywords: SWAT, RBNN, SUFI 2, bootstrap technique, stream flow, simulation

Procedia PDF Downloads 332

Authors: Haifeng Wang, Haili Zhang

Abstract:

Most movie recommendation systems have been developed for customers to find items of interest. This work introduces a predictive model usable by small and medium-sized enterprises (SMEs) who are in need of a data-based and analytical approach to stock proper movies for local audiences and retain more customers. We used classification models to extract features from thousands of customers’ demographic, behavioral and social information to predict their movie genre preference. In the implementation, a Gaussian kernel support vector machine (SVM) classification model and a logistic regression model were established to extract features from sample data and their test error-in-sample were compared. Comparison of error-out-sample was also made under different Vapnik–Chervonenkis (VC) dimensions in the machine learning algorithm to find and prevent overfitting. Gaussian kernel SVM prediction model can correctly predict movie genre preferences in 85% of positive cases. The accuracy of the algorithm increased to 93% with a smaller VC dimension and less overfitting. These findings advance our understanding of how to use machine learning approach to predict customers’ preferences with a small data set and design prediction tools for these enterprises.

Keywords: computational social science, movie preference, machine learning, SVM

Procedia PDF Downloads 239

Authors: Chunhua Liao, Jinfei Wang, Bo Shan, Yang Song, Yongjun He, Taifeng Dong

Abstract:

Winter wheat is one of the main crops in Canada. Forecasting of within-field variability of yield in winter wheat at the early stages is essential for precision farming. However, the crop yield modelling based on high spatial resolution satellite data is generally affected by the lack of continuous satellite observations, resulting in reducing the generalization ability of the models and increasing the difficulty of crop yield forecasting at the early stages. In this study, the correlations between Sentinel-2 data (vegetation indices and reflectance) and yield data collected by combine harvester were investigated and a generalized multivariate linear regression (MLR) model was built and tested with data acquired in different years. It was found that the four-band reflectance (blue, green, red, near-infrared) performed better than their vegetation indices (NDVI, EVI, WDRVI and OSAVI) in wheat yield prediction. The optimum phenological stage for wheat yield prediction with highest accuracy was at the growing stages from the end of the flowering to the beginning of the filling stage. The best MLR model was therefore built to predict wheat yield before harvest using Sentinel-2 data acquired at the end of the flowering stage. Further, to improve the ability of the yield prediction at the early stages, three simple unsupervised domain adaptation (DA) methods were adopted to transform the reflectance data at the early stages to the optimum phenological stage. The winter wheat yield prediction using multiple vegetation indices showed higher accuracy than using single vegetation index. The optimum stage for winter wheat yield forecasting varied with different fields when using vegetation indices, while it was consistent when using multispectral reflectance and the optimum stage for winter wheat yield prediction was at the end of flowering stage. The average testing RMSE of the MLR model at the end of the flowering stage was 604.48 kg/ha. Near the booting stage, the average testing RMSE of yield prediction using the best MLR was reduced to 799.18 kg/ha when applying the mean matching domain adaptation approach to transform the data to the target domain (at the end of the flowering) compared to that using the original data based on the models developed at the booting stage directly (“MLR at the early stage”) (RMSE =1140.64 kg/ha). This study demonstrated that the simple mean matching (MM) performed better than other DA methods and it was found that “DA then MLR at the optimum stage” performed better than “MLR directly at the early stages” for winter wheat yield forecasting at the early stages. The results indicated that the DA had a great potential in near real-time crop yield forecasting at the early stages. This study indicated that the simple domain adaptation methods had a great potential in crop yield prediction at the early stages using remote sensing data.

Keywords: wheat yield prediction, domain adaptation, Sentinel-2, within-field scale

Procedia PDF Downloads 40

Authors: Jamerson Felipe Pereira Lima, Jeane Cecília Bezerra de Melo

Abstract:

The difficulty and cost related to obtaining of protein tertiary structure information through experimental methods, such as X-ray crystallography or NMR spectroscopy, helped raising the development of computational methods to do so. An approach used in these last is prediction of tridimensional structure based in the residue chain, however, this has been proved an NP-hard problem, due to the complexity of this process, explained by the Levinthal paradox. An alternative solution is the prediction of intermediary structures, such as the secondary structure of the protein. Artificial Intelligence methods, such as Bayesian statistics, artificial neural networks (ANN), support vector machines (SVM), among others, were used to predict protein secondary structure. Due to its good results, artificial neural networks have been used as a standard method to predict protein secondary structure. Recent published methods that use this technique, in general, achieved a Q3 accuracy between 75% and 83%, whereas the theoretical accuracy limit for protein prediction is 88%. Alternatively, to achieve better results, support vector machines prediction methods have been developed. The statistical evaluation of methods that use different AI techniques, such as ANNs and SVMs, for example, is not a trivial problem, since different training sets, validation techniques, as well as other variables can influence the behavior of a prediction method. In this study, we propose a prediction method based on artificial neural networks, which is then compared with a selected SVM method. The chosen SVM protein secondary structure prediction method is the one proposed by Huang in his work Extracting Physico chemical Features to Predict Protein Secondary Structure (2013). The developed ANN method has the same training and testing process that was used by Huang to validate his method, which comprises the use of the CB513 protein data set and three-fold cross-validation, so that the comparative analysis of the results can be made comparing directly the statistical results of each method.

Keywords: artificial neural networks, protein secondary structure, protein structure prediction, support vector machines

Procedia PDF Downloads 590

Authors: Joseph C. Chen, Venkata Mohan Kudapa

Abstract:

Development of two real-time surface roughness (Ra) prediction systems for milling operations was attempted. The systems used not only cutting parameters, such as feed rate and spindle speed, but also the cutting current generated and corrected by a clamp type energy sensor. Two different approaches were developed. First, a fuzzy inference system (FIS), in which the fuzzy logic rules are generated by experts in the milling processes, was used to conduct prediction modeling using current cutting data. Second, a neuro-fuzzy system (ANFIS) was explored. Neuro-fuzzy systems are adaptive techniques in which data are collected on the network, processed, and rules are generated by the system. The inference system then uses these rules to predict Ra as the output. Experimental results showed that the parameters of spindle speed, feed rate, depth of cut, and input current variation could predict Ra. These two systems enable the prediction of Ra during the milling operation with an average of 91.83% and 94.48% accuracy by FIS and ANFIS systems, respectively. Statistically, the ANFIS system provided better prediction accuracy than that of the FIS system.

Keywords: surface roughness, input current, fuzzy logic, neuro-fuzzy, milling operations

Procedia PDF Downloads 120

Authors: Sun-Young Jang, Sung-Ah Kim, Dongyoun Shin

Abstract:

The Intelligent transportation system is essential to build smarter cities. Machine learning based transportation prediction could be highly promising approach by delivering invisible aspect visible. In this context, this research aims to make a prototype model that predicts transportation network by using big data and machine learning technology. In detail, among urban transportation systems this research chooses bus system.  The research problem that existing headway model cannot response dynamic transportation conditions. Thus, bus delay problem is often occurred. To overcome this problem, a prediction model is presented to fine patterns of bus delay by using a machine learning implementing the following data sets; traffics, weathers, and bus statues. This research presents a flexible headway model to predict bus delay and analyze the result. The prototyping model is composed by real-time data of buses. The data are gathered through public data portals and real time Application Program Interface (API) by the government. These data are fundamental resources to organize interval pattern models of bus operations as traffic environment factors (road speeds, station conditions, weathers, and bus information of operating in real-time). The prototyping model is designed by the machine learning tool (RapidMiner Studio) and conducted tests for bus delays prediction. This research presents experiments to increase prediction accuracy for bus headway by analyzing the urban big data. The big data analysis is important to predict the future and to find correlations by processing huge amount of data. Therefore, based on the analysis method, this research represents an effective use of the machine learning and urban big data to understand urban dynamics.

Keywords: big data, machine learning, smart city, social cost, transportation network

Procedia PDF Downloads 236

Authors: Isaac Mugume, Charles Basalirwa, Daniel Waiswa, Mary Nsabagwa, Triphonia Jacob Ngailo, Joachim Reuder, Sch¨attler Ulrich, Musa Semujju

Abstract:

The Numerical weather prediction (NWP) models are considered powerful tools for guiding quantitative rainfall prediction. A couple of NWP models exist and are used at many operational weather prediction centers. This study considers two models namely the Consortium for Small–scale Modeling (COSMO) model and the Weather Research and Forecasting (WRF) model. It compares the models’ ability to predict rainfall over Uganda for the period 21st April 2013 to 10th May 2013 using the root mean square (RMSE) and the mean error (ME). In comparing the performance of the models, this study assesses their ability to predict light rainfall events and extreme rainfall events. All the experiments used the default parameterization configurations and with same horizontal resolution (7 Km). The results show that COSMO model had a tendency of largely predicting no rain which explained its under–prediction. The COSMO model (RMSE: 14.16; ME: -5.91) presented a significantly (p = 0.014) higher magnitude of error compared to the WRF model (RMSE: 11.86; ME: -1.09). However the COSMO model (RMSE: 3.85; ME: 1.39) performed significantly (p = 0.003) better than the WRF model (RMSE: 8.14; ME: 5.30) in simulating light rainfall events. All the models under–predicted extreme rainfall events with the COSMO model (RMSE: 43.63; ME: -39.58) presenting significantly higher error magnitudes than the WRF model (RMSE: 35.14; ME: -26.95). This study recommends additional diagnosis of the models’ treatment of deep convection over the tropics.

Keywords: comparative performance, the COSMO model, the WRF model, light rainfall events, extreme rainfall events

Procedia PDF Downloads 239

Authors: Vinayaka Gude Divya Sampath

Abstract:

The research introduces an integrated prediction model to assess the susceptibility of roads in a future flooding event. The model consists of deep learning algorithm for forecasting gauge height data and Flood Inundation Mapper (FIM) for spatial flooding. An optimal architecture for Long short-term memory network (LSTM) was identified for the gauge located on Tangipahoa River at Robert, LA. Dropout was applied to the model to evaluate the uncertainty associated with the predictions. The estimates are then used along with FIM to identify the spatial flooding. Further geoprocessing in ArcGIS provides the susceptibility values for different roads. The model was validated based on the devastating flood of August 2016. The paper discusses the challenges for generalization the methodology for other locations and also for various types of flooding. The developed model can be used by the transportation department and other emergency response organizations for effective disaster management.

Keywords: deep learning, disaster management, flood prediction, urban flooding

Procedia PDF Downloads 119

Authors: Yuwaratu Syafira, Wahyu K. Y. Putra, Kusharisupeni Djokosujono

Abstract:

Background/Objective: Body Mass Index (BMI) serves various purposes, including measuring the prevalence of obesity in a population, and also in formulating a patient’s diet at a hospital, and can be calculated with the equation = body weight (kg)/body height (m)². However, the BMI of an individual with difficulties in carrying their weight or standing up straight can not necessarily be measured. The aim of this study was to form a prediction model for the BMI of young adult students of Public Health Faculty of University of Indonesia. Subject/Method: This study used a cross sectional design, with a total sample of 132 respondents, consisted of 58 males and 74 females aged 21- 30. The dependent variable of this study was BMI, and the independent variables consisted of sex and anthropometric measurements, which included ulna length, arm length, tibia length, knee height, mid-upper arm circumference, and calf circumference. Anthropometric information was measured and recorded in a single sitting. Simple and multiple linear regression analysis were used to create the prediction equation for BMI. Results: The male respondents had an average BMI of 24.63 kg/m² and the female respondents had an average of 22.52 kg/m². A total of 17 variables were analysed for its correlation with BMI. Bivariate analysis showed the variable with the strongest correlation with BMI was Mid-Upper Arm Circumference/√Ulna Length (MUAC/√UL) (r = 0.926 for males and r = 0.886 for females). Furthermore, MUAC alone also has a very strong correlation with BMI (r = 0,913 for males and r = 0,877 for females). Prediction models formed from either MUAC/√UL or MUAC alone both produce highly accurate predictions of BMI. However, measuring MUAC/√UL is considered inconvenient, which may cause difficulties when applied on the field. Conclusion: The prediction model considered most ideal to estimate BMI is: Male BMI (kg/m²) = 1.109(MUAC (cm)) – 9.202 and Female BMI (kg/m²) = 0.236 + 0.825(MUAC (cm)), based on its high accuracy levels and the convenience of measuring MUAC on the field.

Keywords: body mass index, mid-upper arm circumference, prediction model, ulna length

Procedia PDF Downloads 198

Authors: Muhammet Baldan, Emel Timuçin

Abstract:

Molecular solubility prediction plays a crucial role in various fields, such as drug discovery, environmental science, and material science. In this study, we compare the performance of five machine learning algorithms—linear regression, support vector machines (SVM), random forests, gradient boosting machines (GBM), and neural networks—for predicting molecular solubility using the AqSolDB dataset. The dataset consists of 9981 data points with their corresponding solubility values. MACCS keys (166 bits), RDKit properties (20 properties), and structural properties(3) features are extracted for every smile representation in the dataset. A total of 189 features were used for training and testing for every molecule. Each algorithm is trained on a subset of the dataset and evaluated using metrics accuracy scores. Additionally, computational time for training and testing is recorded to assess the efficiency of each algorithm. Our results demonstrate that random forest model outperformed other algorithms in terms of predictive accuracy, achieving an 0.93 accuracy score. Gradient boosting machines and neural networks also exhibit strong performance, closely followed by support vector machines. Linear regression, while simpler in nature, demonstrates competitive performance but with slightly higher errors compared to ensemble methods. Overall, this study provides valuable insights into the performance of machine learning algorithms for molecular solubility prediction, highlighting the importance of algorithm selection in achieving accurate and efficient predictions in practical applications.

Keywords: random forest, machine learning, comparison, feature extraction

Procedia PDF Downloads 4

Authors: Suzan Arslanturk, Mohammad-Reza Siadat, Theophilus Ogunyemi, Ananias Diokno

Abstract:

Missing data is a common challenge in statistical analyses of most clinical survey datasets. A variety of methods have been developed to enable analysis of survey data to deal with missing values. Imputation is the most commonly used among the above methods. However, in order to minimize the bias introduced due to imputation, one must choose the right imputation technique and apply it to the correct type of missing data. In this paper, we have identified different types of missing values: missing data due to skip pattern (SPMD), undetermined missing data (UMD), and genuine missing data (GMD) and applied rough set imputation on only the GMD portion of the missing data. We have used rough set imputation to evaluate the effect of such imputation on prediction by generating several simulation datasets based on an existing epidemiological dataset (MESA). To measure how well each dataset lends itself to the prediction model (logistic regression), we have used p-values from the Wald test. To evaluate the accuracy of the prediction, we have considered the width of 95% confidence interval for the probability of incontinence. Both imputed and non-imputed simulation datasets were fit to the prediction model, and they both turned out to be significant (p-value < 0.05). However, the Wald score shows a better fit for the imputed compared to non-imputed datasets (28.7 vs. 23.4). The average confidence interval width was decreased by 10.4% when the imputed dataset was used, meaning higher precision. The results show that using the rough set method for missing data imputation on GMD data improve the predictive capability of the logistic regression. Further studies are required to generalize this conclusion to other clinical survey datasets.

Keywords: rough set, imputation, clinical survey data simulation, genuine missing data, predictive index

Procedia PDF Downloads 145

Authors: David Percy

Abstract:

Landslides are a geologic phenomenon that affects a large number of inhabited places and are constantly being monitored and studied for the prediction of future occurrences. Reconstructability analysis (RA) is a methodology for extracting informative models from large volumes of data that work exclusively with discrete data. While RA has been used in medical applications and social science extensively, we are introducing it to the spatial sciences through applications like landslide prediction. Since RA works exclusively with discrete data, such as soil classification or bedrock type, working with continuous data, such as porosity, requires that these data are binned for inclusion in the model. RA constructs models of the data which pick out the most informative elements, independent variables (IVs), from each layer that predict the dependent variable (DV), landslide occurrence. Each layer included in the model retains its classification data as a primary encoding of the data. Unlike other machine learning algorithms that force the data into one-hot encoding type of schemes, RA works directly with the data as it is encoded, with the exception of continuous data, which must be binned. The usual physical and derived layers are included in the model, and testing our results against other published methodologies, such as neural networks, yields accuracy that is similar but with the advantage of a completely transparent model. The results of an RA session with a data set are a report on every combination of variables and their probability of landslide events occurring. In this way, every combination of informative state combinations can be examined.

Keywords: reconstructability analysis, machine learning, landslides, raster analysis

Procedia PDF Downloads 44

Authors: A. K. M. Kamrul Islam, Abdelhamid Bouchachia, Suang Cang, Hongnian Yu

Abstract:

Forecasting model has a great impact in terms of prediction and continues to do so into the future. Although many forecasting models have been studied in recent years, most researchers focus on different forecasting methods based on fuzzy time series to solve forecasting problems. The forecasted models accuracy fully depends on the two terms that are the length of the interval in the universe of discourse and the content of the forecast rules. Moreover, a hybrid forecasting method can be an effective and efficient way to improve forecasts rather than an individual forecasting model. There are different hybrids forecasting models which combined fuzzy time series with evolutionary algorithms, but the performances are not quite satisfactory. In this paper, we proposed a hybrid forecasting model which deals with the first order as well as high order fuzzy time series and particle swarm optimization to improve the forecasted accuracy. The proposed method used the historical enrollments of the University of Alabama as dataset in the forecasting process. Firstly, we considered an automatic clustering algorithm to calculate the appropriate interval for the historical enrollments. Then particle swarm optimization and fuzzy time series are combined that shows better forecasting accuracy than other existing forecasting models.

Keywords: fuzzy time series (fts), particle swarm optimization, clustering algorithm, hybrid forecasting model

Procedia PDF Downloads 224

Authors: Hyon I. Paek, Sang Rim Kim, Hyon A. Ryu

Abstract:

In functional linear regression, one typical problem is to reduce dimension. Compared with multivariate linear regression, functional linear regression is regarded as an infinite-dimensional case, and the main task is to reduce dimensions of functional response and functional predictors. One common approach is to adapt functional principal component analysis (FPCA) on functional predictors and then use a few leading functional principal components (FPC) to predict the functional model. The leading FPCs estimated by the typical FPCA explain a major variation of the functional predictor, but these leading FPCs may not be mostly correlated with the functional response, so they may not be significant in the prediction for response. In this paper, we propose a supervised functional principal component analysis method for a functional response model with FPCs obtained by considering the correlation of the functional response. Our method would have a better prediction accuracy than the typical FPCA method.

Keywords: supervised, functional principal component analysis, functional response, functional linear regression

Procedia PDF Downloads 42

Authors: Soheila Sadeghi

Abstract:

Accurate cost prediction is essential for effective project management and decision-making in the construction industry. This study aims to develop a cost prediction model for construction projects using Machine Learning techniques and Particle Swarm Optimization (PSO). The research utilizes a comprehensive dataset containing project cost estimates, actual costs, resource details, and project performance metrics from a road reconstruction project. The methodology involves data preprocessing, feature selection, and the development of an Artificial Neural Network (ANN) model optimized using PSO. The study investigates the impact of various input features, including cost estimates, resource allocation, and project progress, on the accuracy of cost predictions. The performance of the optimized ANN model is evaluated using metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R-squared. The results demonstrate the effectiveness of the proposed approach in predicting project costs, outperforming traditional benchmark models. The feature selection process identifies the most influential variables contributing to cost variations, providing valuable insights for project managers. However, this study has several limitations. Firstly, the model's performance may be influenced by the quality and quantity of the dataset used. A larger and more diverse dataset covering different types of construction projects would enhance the model's generalizability. Secondly, the study focuses on a specific optimization technique (PSO) and a single Machine Learning algorithm (ANN). Exploring other optimization methods and comparing the performance of various ML algorithms could provide a more comprehensive understanding of the cost prediction problem. Future research should focus on several key areas. Firstly, expanding the dataset to include a wider range of construction projects, such as residential buildings, commercial complexes, and infrastructure projects, would improve the model's applicability. Secondly, investigating the integration of additional data sources, such as economic indicators, weather data, and supplier information, could enhance the predictive power of the model. Thirdly, exploring the potential of ensemble learning techniques, which combine multiple ML algorithms, may further improve cost prediction accuracy. Additionally, developing user-friendly interfaces and tools to facilitate the adoption of the proposed cost prediction model in real-world construction projects would be a valuable contribution to the industry. The findings of this study have significant implications for construction project management, enabling proactive cost estimation, resource allocation, budget planning, and risk assessment, ultimately leading to improved project performance and cost control. This research contributes to the advancement of cost prediction techniques in the construction industry and highlights the potential of Machine Learning and PSO in addressing this critical challenge. However, further research is needed to address the limitations and explore the identified future research directions to fully realize the potential of ML-based cost prediction models in the construction domain.

Keywords: cost prediction, construction projects, machine learning, artificial neural networks, particle swarm optimization, project management, feature selection, road reconstruction

Procedia PDF Downloads 15

Authors: Ehsan Mehryaar, Seyed Armin Motahari Tabari

Abstract:

Without a doubt, soil shear strength is the most important property of the soil. The majority of fatal and catastrophic geological accidents are related to shear strength failure of the soil. Therefore, its prediction is a matter of high importance. However, acquiring the shear strength is usually a cumbersome task that might need complicated laboratory testing. Therefore, prediction of it based on common and easy to get soil properties can simplify the projects substantially. In this paper, A hybrid model based on the classification and regression tree algorithm and logistic regression is proposed where each leaf of the tree is an independent regression model. A database of 189 points for clay soil, including Moisture content, liquid limit, plastic limit, clay content, and shear strength, is collected. The performance of the developed model compared to the existing models and equations using root mean squared error and coefficient of correlation.

Keywords: model tree, CART, logistic regression, soil shear strength

Procedia PDF Downloads 171

Authors: Nishank Raisinghani

Abstract:

Efficiently predicting drug response remains a challenge in the realm of drug discovery. To address this issue, we propose four model architectures that combine graphical representation with varying positions of multiheaded self-attention mechanisms. By leveraging two types of multi-omics data, transcriptomics and genomics, we create a comprehensive representation of target cells and enable drug response prediction in precision medicine. A majority of our architectures utilize multiple transformer models, one with a graph attention mechanism and the other with a multiheaded self-attention mechanism, to generate latent representations of both drug and omics data, respectively. Our model architectures apply an attention mechanism to both drug and multiomics data, with the goal of procuring more comprehensive latent representations. The latent representations are then concatenated and input into a fully connected network to predict the IC-50 score, a measure of cell drug response. We experiment with all four of these architectures and extract results from all of them. Our study greatly contributes to the future of drug discovery and precision medicine by looking to optimize the time and accuracy of drug response prediction.

Keywords: drug discovery, transformers, graph neural networks, multiomics

Procedia PDF Downloads 119

Authors: Rohini Hariharan, Yazhini R, Bhamidipati Naga Shrikarti

Abstract:

In the realm of cricket, particularly within the context of the Indian Premier League (IPL), the ability to predict team scores accurately holds significant importance for both cricket enthusiasts and stakeholders alike. This paper presents a comprehensive study on IPL score prediction utilizing various machine learning algorithms, including Support Vector Machines (SVM), XGBoost, Multiple Regression, Linear Regression, K-nearest neighbors (KNN), and Random Forest. Through meticulous data preprocessing, feature engineering, and model selection, we aimed to develop a robust predictive framework capable of forecasting team scores with high precision. Our experimentation involved the analysis of historical IPL match data encompassing diverse match and player statistics. Leveraging this data, we employed state-of-the-art machine learning techniques to train and evaluate the performance of each model. Notably, Multiple Regression emerged as the top-performing algorithm, achieving an impressive accuracy of 77.19% and a precision of 54.05% (within a threshold of +/- 10 runs). This research contributes to the advancement of sports analytics by demonstrating the efficacy of machine learning in predicting IPL team scores. The findings underscore the potential of advanced predictive modeling techniques to provide valuable insights for cricket enthusiasts, team management, and betting agencies. Additionally, this study serves as a benchmark for future research endeavors aimed at enhancing the accuracy and interpretability of IPL score prediction models.

Keywords: indian premier league (IPL), cricket, score prediction, machine learning, support vector machines (SVM), xgboost, multiple regression, linear regression, k-nearest neighbors (KNN), random forest, sports analytics

Procedia PDF Downloads 24

Authors: Habtamu Ayenew Asegie

Abstract:

Wealth, as opposed to income or consumption, implies a more stable and permanent status. Due to natural and human-made difficulties, households' economies will be diminished, and their well-being will fall into trouble. Hence, governments and humanitarian agencies offer considerable resources for poverty and malnutrition reduction efforts. One key factor in the effectiveness of such efforts is the accuracy with which low-income or poor populations can be identified. As a result, this study aims to predict a household’s wealth status using ensemble Machine learning (ML) algorithms. In this study, design science research methodology (DSRM) is employed, and four ML algorithms, Random Forest (RF), Adaptive Boosting (AdaBoost), Light Gradient Boosted Machine (LightGBM), and Extreme Gradient Boosting (XGBoost), have been used to train models. The Ethiopian Demographic and Health Survey (EDHS) dataset is accessed for this purpose from the Central Statistical Agency (CSA)'s database. Various data pre-processing techniques were employed, and the model training has been conducted using the scikit learn Python library functions. Model evaluation is executed using various metrics like Accuracy, Precision, Recall, F1-score, area under curve-the receiver operating characteristics (AUC-ROC), and subjective evaluations of domain experts. An optimal subset of hyper-parameters for the algorithms was selected through the grid search function for the best prediction. The RF model has performed better than the rest of the algorithms by achieving an accuracy of 96.06% and is better suited as a solution model for our purpose. Following RF, LightGBM, XGBoost, and AdaBoost algorithms have an accuracy of 91.53%, 88.44%, and 58.55%, respectively. The findings suggest that some of the features like ‘Age of household head’, ‘Total children ever born’ in a family, ‘Main roof material’ of their house, ‘Region’ they lived in, whether a household uses ‘Electricity’ or not, and ‘Type of toilet facility’ of a household are determinant factors to be a focal point for economic policymakers. The determinant risk factors, extracted rules, and designed artifact achieved 82.28% of the domain expert’s evaluation. Overall, the study shows ML techniques are effective in predicting the wealth status of households.

Keywords: ensemble machine learning, households wealth status, predictive model, wealth status prediction

Procedia PDF Downloads 17

Authors: Ejaz Ahmed, Huang Yong

Abstract:

The operation of aero engines has a critical importance in the vicinity of lean blowout (LBO) limits. Lefebvre’s model of LBO based on empirical correlation has been extended to flame volume concept by the authors. The flame volume takes into account the effects of geometric configuration, the complex spatial interaction of mixing, turbulence, heat transfer and combustion processes inside the gas turbine combustion chamber. For these reasons, flame volume based LBO predictions are more accurate. Although LBO prediction accuracy has improved, it poses a challenge associated with Vf estimation in real gas turbine combustors. This work extends the approach of flame volume prediction previously based on fuel iterative approximation with cold flow simulations to reactive flow simulations. Flame volume for 11 combustor configurations has been simulated and validated against experimental data. To make prediction methodology robust as required in the preliminary design stage, reactive flow simulations were carried out with the combination of probability density function (PDF) and discrete phase model (DPM) in FLUENT 15.0. The criterion for flame identification was defined. Two important parameters i.e. critical injection diameter (Dp,crit) and critical temperature (Tcrit) were identified, and their influence on reactive flow simulation was studied for Vf estimation. Obtained results exhibit ±15% error in Vf estimation with experimental data.

Keywords: CFD, combustion, gas turbine combustor, lean blowout

Procedia PDF Downloads 246

Authors: A. Brahma

Abstract:

The drying of hydraulics materials is unavoidable and conducted to important spontaneous deformations. In this study, we show that it is possible to describe the drying shrinkage of the high-performance concrete by a simple expression. A multiple regression model was developed for the prediction of the drying shrinkage of the high-performance concrete. The assessment of the proposed model has been done by a set of statistical tests. The model developed takes in consideration the main parameters of confection and conservation. There was a very good agreement between drying shrinkage predicted by the multiple regression model and experimental results. The developed model adjusts easily to all hydraulic concrete types.

Keywords: hydraulic concretes, drying, shrinkage, prediction, modeling

Procedia PDF Downloads 344