Search results for: predictive accuracy

Authors: Monika Chuchro

Abstract:

This paper presents a preliminary attempt to apply classification of time series using meta-clusters in order to improve the quality of regression models. In this case, clustering was performed as a method to obtain a subgroups of time series data with normal distribution from inflow into waste water treatment plant data which Composed of several groups differing by mean value. Two simple algorithms: K-mean and EM were chosen as a clustering method. The rand index was used to measure the similarity. After simple meta-clustering, regression model was performed for each subgroups. The final model was a sum of subgroups models. The quality of obtained model was compared with the regression model made using the same explanatory variables but with no clustering of data. Results were compared by determination coefficient (R2), measure of prediction accuracy mean absolute percentage error (MAPE) and comparison on linear chart. Preliminary results allows to foresee the potential of the presented technique.

Keywords: clustering, data analysis, data mining, predictive models

Procedia PDF Downloads 443

Authors: William Anderson, Kyle Kobold, Oleg Yakimenko

Abstract:

Green microgrids using mostly renewable energy (RE) for generation, are complex systems with inherent nonlinear dynamics. Among a variety of different optimization tools there are only a few ones that adequately consider this complexity. This paper evaluates applicability of two somewhat similar optimization tools tailored for standalone RE microgrids and also assesses a machine learning tool for performance prediction that can enhance the reliability of any chosen optimization tool. It shows that one of these microgrid optimization tools has certain advantages over another and presents a detailed routine of preparing input data to simulate RE microgrid behavior. The paper also shows how neural-network-based predictive modeling can be used to validate and forecast solar power generation based on weather time series data, which improves the overall quality of standalone RE microgrid analysis.

Keywords: microgrid, renewable energy, complex systems, optimization, predictive modeling, neural networks

Procedia PDF Downloads 258

Authors: Khadidja Bellifa, Sidi Mohamed Mekelleche

Abstract:

Phenols and especially halogenated phenols represent a substantial part of the chemicals produced worldwide and are known as aquatic pollutants. Quantitative structure–toxicity relationship (QSTR) models are useful for understanding how chemical structure relates to the toxicity of chemicals. In the present study, the acute toxicities of 45 halogenated phenols to Tetrahymena Pyriformis are estimated using no cost semi-empirical quantum chemistry methods. QSTR models were established using the multiple linear regression technique and the predictive ability of the models was evaluated by the internal cross-validation, the Y-randomization and the external validation. Their structural chemical domain has been defined by the leverage approach. The results show that the best model is obtained with the AM1 method (R²= 0.91, R²CV= 0.90, SD= 0.20 for the training set and R²= 0.96, SD= 0.11 for the test set). Moreover, all the Tropsha’ criteria for a predictive QSTR model are verified.

Keywords: halogenated phenols, toxicity mechanism, hydrophobicity, electrophilicity index, quantitative stucture-toxicity relationships

Procedia PDF Downloads 271

Authors: Shailaja Grover, Purvi Tiwari, Vigneshwaran S. R., U. Dinesh Kumar

Abstract:

In India, grapes are one of the most important horticulture crops. Grapes are most vulnerable to downy mildew, which is one of the most devasting diseases. In the absence of a precise weather-based advisory system, farmers spray pesticides on their crops extensively. There are two main challenges associated with using these pesticides. Firstly, most of these sprays were panic sprays, which could have been avoided. Second, farmers use more expensive "Preventive and Eradicate" chemicals than "Systemic, Curative and Anti-sporulate" chemicals. When these chemicals are used indiscriminately, they can enter the fruit and cause health problems such as cancer. This paper utilizes decision trees and predictive modeling techniques to provide grape farmers with customized advice on grape disease management. This model is expected to reduce the overall use of chemicals by approximately 50% and the cost by around 70%. Most of the grapes produced will have relatively low residue levels of pesticides, i.e., below the permissible level.

Keywords: analytics in agriculture, downy mildew, weather based advisory, decision tree, predictive modelling

Procedia PDF Downloads 47

Authors: Jaypalsinh B. Parmar, Pintu Nakrani, Ashish Chaurasia

Abstract:

Digital Elevation Model (DEM) is considered as an important data in GIS-based terrain analysis for many applications and assessment of processes such as environmental and climate change studies, hydrologic modelling, etc. Vertical accuracy of DEM having geographically dynamic nature depends on different parameters which affect the model simulation outcomes. Vertical accuracy assessment in Indian landscape especially in low-lying coastal urban terrain such as lower Tapi Basin is very limited. In the present study, attempt has been made to evaluate the vertical accuracy of 30m resolution open source Indian National Cartosat-1 DEM v3 for Lower Tapi Basin (LTB) from western India. The extensive field investigation is carried out using stratified random fast static DGPS survey in the entire study region, and 117 high accuracy ground control points (GCPs) have been obtained. The above open source DEM was compared with obtained GCPs, and different statistical attributes were envisaged, and vertical error histograms were also evaluated.

Keywords: CartoDEM, Digital Elevation Model, GPS, lower Tapi basin

Procedia PDF Downloads 336

Authors: Christina Adly, Meena Abdelmeseeh, Tamer Basha

Abstract:

This paper proposes a system for hand movement recognition using multichannel surface EMG(sEMG) signals obtained from 40 subjects using 40 different exercises, which are available on the Ninapro(Non-Invasive Adaptive Prosthetics) database. First, we applied processing methods to the raw sEMG signals to convert them to their amplitudes. Second, we used deep learning methods to solve our problem by passing the preprocessed signals to Fully connected neural networks(FCNN) and recurrent neural networks(RNN) with Long Short Term Memory(LSTM). Using intrasubject evaluation, The accuracy using the FCNN is 72%, with a processing time for training around 76 minutes, and for RNN's accuracy is 79.9%, with 8 minutes and 22 seconds processing time. Third, we applied some postprocessing methods to improve the accuracy, like majority voting(MV) and Movement Error Rate(MER). The accuracy after applying MV is 75% and 86% for FCNN and RNN, respectively. The MER value has an inverse relationship with the prediction delay while varying the window length for measuring the MV. The different part uses the RNN with the intersubject evaluation. The experimental results showed that to get a good accuracy for testing with reasonable processing time, we should use around 20 subjects.

Keywords: hand movement recognition, recurrent neural network, movement error rate, intrasubject evaluation, intersubject evaluation

Procedia PDF Downloads 112

Authors: Danni Cong, Meiping Wu, Hua Mu, Xiaofeng He, Junxiang Lian, Juliang Cao, Shaokun Cai, Hao Qin

Abstract:

Gravity field is of great significance in geoscience, national economy and national security, and gravitational gradient measurement has been extensively studied due to its higher accuracy than gravity measurement. Gravity gradient sensor, being one of core devices of the gravity gradient instrument, plays a key role in measuring accuracy. Therefore, this paper starts from analyzing the working principle of the gravity gradient sensor by Newton’s law, and then considers the relative motion between inertial and non-inertial systems to build a relatively adequate mathematical model, laying a foundation for the measurement error calibration, measurement accuracy improvement.

Keywords: gravity gradient, gravity gradient sensor, accelerometer, single-axis rotation modulation

Procedia PDF Downloads 300

Authors: Xiaobo Liu, Rakesh Mishra, Yun Zhang

Abstract:

Continuous monitoring of forest canopy height with large coverage is essential for obtaining forest carbon stocks and emissions, quantifying biomass estimation, analyzing vegetation coverage, and determining biodiversity. LiDAR can be used to collect accurate woody vegetation structure such as canopy height. However, LiDAR’s coverage is usually limited because of its high cost and limited maneuverability, which constrains its use for dynamic and large area forest canopy monitoring. On the other hand, optical satellite images, like Sentinel-2, have the ability to cover large forest areas with a high repeat rate, but they do not have height information. Hence, exploring the solution of integrating LiDAR data and Sentinel-2 images to enlarge the coverage of forest canopy height prediction and increase the prediction repeat rate has been an active research topic in the environmental remote sensing community. In this study, we explore the potential of training a Random Forest Regression (RFR) model and a Convolutional Neural Network (CNN) model, respectively, to develop two predictive models for predicting and validating the forest canopy height of the Acadia Forest in New Brunswick, Canada, with a 10m ground sampling distance (GSD), for the year 2018 and 2021. Two 10m airborne LiDAR-derived canopy height models, one for 2018 and one for 2021, are used as ground truth to train and validate the RFR and CNN predictive models. To evaluate the prediction performance of the trained RFR and CNN models, two new predicted canopy height maps (CHMs), one for 2018 and one for 2021, are generated using the trained RFR and CNN models and 10m Sentinel-2 images of 2018 and 2021, respectively. The two 10m predicted CHMs from Sentinel-2 images are then compared with the two 10m airborne LiDAR-derived canopy height models for accuracy assessment. The validation results show that the mean absolute error (MAE) for year 2018 of the RFR model is 2.93m, CNN model is 1.71m; while the MAE for year 2021 of the RFR model is 3.35m, and the CNN model is 3.78m. These demonstrate the feasibility of using the RFR and CNN models developed in this research for predicting large-coverage forest canopy height at 10m spatial resolution and a high revisit rate.

Keywords: remote sensing, forest canopy height, LiDAR, Sentinel-2, artificial intelligence, random forest regression, convolutional neural network

Procedia PDF Downloads 61

Authors: Dmytro Zubov, Francesco Volponi

Abstract:

In this paper, D-Wave quantum computing Ising model is used for the forecasting of positive extremes of daily mean air temperature. Forecast models are designed with two to five qubits, which represent 2-, 3-, 4-, and 5-day historical data respectively. Ising model’s real-valued weights and dimensionless coefficients are calculated using daily mean air temperatures from 119 places around the world, as well as sea level (Aburatsu, Japan). In comparison with current methods, this approach is better suited to predict heat wave values because it does not require the estimation of a probability distribution from scarce observations. Proposed forecast quantum computing algorithm is simulated based on traditional computer architecture and combinatorial optimization of Ising model parameters for the Ronald Reagan Washington National Airport dataset with 1-day lead-time on learning sample (1975-2010 yr). Analysis of the forecast accuracy (ratio of successful predictions to total number of predictions) on the validation sample (2011-2014 yr) shows that Ising model with three qubits has 100 % accuracy, which is quite significant as compared to other methods. However, number of identified heat waves is small (only one out of nineteen in this case). Other models with 2, 4, and 5 qubits have 20 %, 3.8 %, and 3.8 % accuracy respectively. Presented three-qubit forecast model is applied for prediction of heat waves at other five locations: Aurel Vlaicu, Romania – accuracy is 28.6 %; Bratislava, Slovakia – accuracy is 21.7 %; Brussels, Belgium – accuracy is 33.3 %; Sofia, Bulgaria – accuracy is 50 %; Akhisar, Turkey – accuracy is 21.4 %. These predictions are not ideal, but not zeros. They can be used independently or together with other predictions generated by different method(s). The loss of human life, as well as environmental, economic, and material damage, from extreme air temperatures could be reduced if some of heat waves are predicted. Even a small success rate implies a large socio-economic benefit.

Keywords: heat wave, D-wave, forecast, Ising model, quantum computing

Procedia PDF Downloads 472

Authors: Mostafa ‘Neo’ Mohsenvand, Mohammad Rasool Izadi, Pattie Maes

Abstract:

EEG-based biometrics (user identification) has been explored on small datasets of no more than 157 subjects. Here we show that the accuracy of modern supervised methods falls rapidly as the number of users increases to a few thousand. Moreover, supervised methods require a large amount of labeled data for training which limits their applications in real-world scenarios where acquiring data for training should not take more than a few minutes. We show that using contrastive learning for pre-training, it is possible to maintain high accuracy on a dataset of 2130 subjects while only using a fraction of labels. We compare 5 different self-supervised tasks for pre-training of the encoder where our proposed method achieves the accuracy of 96.4%, improving the baseline supervised models by 22.75% and the competing self-supervised model by 3.93%. We also study the effects of the length of the signal and the number of channels on the accuracy of the user-identification models. Our results reveal that signals from temporal and frontal channels contain more identifying features compared to other channels.

Keywords: brainprint, contrastive learning, electroencephalo-gram, self-supervised learning, user identification

Procedia PDF Downloads 134

Authors: Megan E. Davies

Abstract:

With the intention of identifying cognitive processes uniquely involved in the dark triad personality traits of psychopathy, Machiavellianism, and narcissism, this study aimed to determine further potential differences and parameters of individual traits by explaining a statistically significant amount of variance between the constructs of manipulativeness, impulsiveness, grit, and need for cognition within the dark triad. Applying a cross-sectional design, N = 96 participants self-reported using the MACH-IV, SRP-III, NFC-S, and Grit Scale for Perseverance and Passion for Long-Term Goals. Hierarchical regression analyses showed that only manipulativeness predicted Machiavellianism, whereas manipulativeness and impulsiveness were found to have predictive qualities for psychopathy. Overall, these results found areas of discrepancy and overlap between manipulation and impulsivity regarding psychopathy and Machiavellianism. Additionally, this study serves to preliminarily eliminate the Need for Cognition and grit as predictive variables for Machiavellianism and psychopathy.

Keywords: Machiavellianism, psychopathy, manipulation, impulsiveness, need for cognition, grit, dark triad

Procedia PDF Downloads 77

Authors: Klara Smith-Etxeberria

Abstract:

main goal of this study was to analyze the predictive ability of some variables associated with the parental divorce process alongside attachment history with parents on both, mother-child and father-child relationship quality. Our sample consisted of 173 undergraduate and vocational school students from the Autonomous Community of the Basque Country. All of them belonged to a divorced family. Results showed that adequate maternal strategies during the divorce process (e.g.: stable, continuous and positive role as a mother) was the variable with greater predictive ability on mother-child relationships quality. In addition, secure attachment history with mother also predicted positive mother-child relationships. On the other hand, father-child relationship quality was predicted by adequate paternal strategies during the divorce process, such as his stable, continuous and positive role as a father, along with not badmouthing the mother and promoting good mother-child relationships. Furthermore, paternal negative emotional state due to divorce was positively associated with father-child relationships quality, and both, history of attachment with mother and with father predicted father-child relationships quality. In conclusion, our data indicate that both, paternal and maternal strategies for children´s adequate adjustment during the divorce process influence on mother-child and father-child relationships quality. However, these results suggest that paternal strategies during the divorce process have a greater predictive ability on father-child relationships quality, whereas maternal positive strategies during divorce determine positive mother-child relationships among young adults.

Keywords: father-child relationships quality, mother-child relationships quality, parental divorce process, young adulthood

Procedia PDF Downloads 239

Authors: Mohammed Al-Zubaidi, Katherine Ong, Pravin Viswambaram, Steve McCombie, Oliver Oey, Jeremy Ong, Richard Gauci, Ronny Low, Dickon Hayne

Abstract:

Objective: Lymph node involvement along with distant metastasis in a patient with invasive bladder cancer determines the disease survival, therefeor, it is an essential determinant of the therapeutic management and outcome. This retrospective study aims to determine the accuracy of FDG PET scan in detecting lymphatic involvement and distant metastatic urothelial cancer compared to conventional CT staging. Method: A retrospective review of 76 patients with UC or BC who underwent surgery or confirmatory biopsy that was staged with both CT and 18F-FDG-PET (up to 8 weeks apart) between 2015 and 2020. Fifty-sevenpatients (75%) had formal pelvic LN dissection or biopsy of suspicious metastasis. 18F-FDG-PET reports for positive sites were qualitative depending on SUV Max. On the other hand, enlarged LN by RECIST criteria 1.1 (>10 mm) and other qualitative findings suggesting metastasis were considered positive in CT scan. Histopathological findings from surgical specimens or image-guided biopsies were considered the gold standard in comparison to imaging reports. 18F-FDG-avid or enlarged pelvic LNs with surgically proven nodal metastasis were considered true positives. Performance characteristics of 18F-FDG-PET and CT, including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (PPV), were calculated. Results: Pelvic LN involvement was confirmed histologically in 10/57 (17.5%) patients. Sensitivity, specificity, PPV and NPV of CT for detecting pelvic LN metastases were 41.17% (95% CI:18-67%), 100% (95% CI:90-100%) 100% (95% CI:59-100%) and 78.26% (95% CI:64-89%) respectively. Sensitivity, specificity, PPV and NPV of 18F-FDG-PET for detecting pelvic LN metastases were 62.5% (95% CI:35-85%), 83.78% (95% CI:68-94%), 62.5% (95% CI:35-85%), and 83.78% (95% CI:68-94%) respectively. Pre-operative staging with 18F-FDG-PET identified the distant metastatic disease in 9/76 (11.8%) patients who were occult on CT. This retrospective study suggested that 18F-FDG-PET may be more sensitive than CT for detecting pelvic LN metastases. 7/76 (9.2%) patients avoided cystectomy due to 18F-FDG-PET diagnosed metastases that were not reported on CT. Conclusion: 18F-FDG-PET is more sensitive than CT for pelvic LN metastases, which can be used as the standard modality of bladder cancer staging, as it may change the treatment by detecting lymph node metastasis that was occult in CT. Further research involving randomised controlled trials comparing the diagnostic yield of 18F-FDG-PET and CT in detecting nodal and distant metastasis in UC or BC is warranted to confirm our findings.

Keywords: FDG PET, CT scan, urothelial cancer, bladder cancer

Procedia PDF Downloads 96

Authors: Jaan-Willem Simon, Bertram Stier, Brett Bednarcyk, Evan Pineda, Stefanie Reese

Abstract:

Textile composites, in which the reinforcing fibers are woven or braided, have become very popular in numerous applications in aerospace, automotive, and maritime industry. These textile composites are advantageous due to their ease of manufacture, damage tolerance, and relatively low cost. However, physics-based modeling of the mechanical behavior of textile composites is challenging. Compared to their unidirectional counterparts, textile composites introduce additional geometric complexities, which cause significant local stress and strain concentrations. Since these internal concentrations are primary drivers of nonlinearity, damage, and failure within textile composites, they must be taken into account in order for the models to be predictive. The macro-scale approach to modeling textile-reinforced composites treats the whole composite as an effective, homogenized material. This approach is very computationally efficient, but it cannot be considered predictive beyond the elastic regime because the complex microstructural geometry is not considered. Further, this approach can, at best, offer a phenomenological treatment of nonlinear deformation and failure. In contrast, the mesoscale approach to modeling textile composites explicitly considers the internal geometry of the reinforcing tows, and thus, their interaction, and the effects of their curved paths can be modeled. The tows are treated as effective (homogenized) materials, requiring the use of anisotropic material models to capture their behavior. Finally, the micro-scale approach goes one level lower, modeling the individual filaments that constitute the tows. This paper will compare meso- and micro-scale approaches to modeling the deformation, damage, and failure of textile-reinforced polymer matrix composites. For the mesoscale approach, the woven composite architecture will be modeled using the finite element method, and an anisotropic damage model for the tows will be employed to capture the local nonlinear behavior. For the micro-scale, two different models will be used, the one being based on the finite element method, whereas the other one makes use of an embedded semi-analytical approach. The goal will be the comparison and evaluation of these approaches to modeling textile-reinforced composites in terms of accuracy, efficiency, and utility.

Keywords: multiscale modeling, continuum damage model, damage interaction, textile composites

Procedia PDF Downloads 326

Authors: Rashi Bhalla, Russel Pears, M. Asif Naeem

Abstract:

In this 'Information Explosion Era' analyzing data 'a critical commodity' and mining knowledge from vertically distributed data stream incurs huge communication cost. However, an effort to decrease the communication in the distributed environment has an adverse influence on the classification accuracy; therefore, a research challenge lies in maintaining a balance between transmission cost and accuracy. This paper proposes a method based on Bayesian inference to reduce the communication volume in a heterogeneous distributed environment while retaining prediction accuracy. Our experimental evaluation reveals that a significant reduction in communication can be achieved across a diverse range of dataset types.

Keywords: big data, bayesian inference, distributed data stream mining, heterogeneous-distributed data

Procedia PDF Downloads 138

Authors: Sara Hasani

Abstract:

This research focuses on natural sudden onset disasters characterised as ‘occurring with little or no warning and often cause excessive injuries far surpassing the national response capacities’. Based on the panel analysis of the historic record of 4,252 natural onset disasters between 1980 to 2015, a predictive method was developed to predict the human impact of the disaster (fatality, injured, homeless) with less than 3% of errors. The geographical dispersion of the disasters includes every country where the data were available and cross-examined from various humanitarian sources. The records were then filtered into 4252 records of the disasters where the five predictive variables (disaster type, HDI, DRI, population, and population density) were clearly stated. The procedure was designed based on a combination of pattern recognition techniques and rule-based clustering for prediction and discrimination analysis to validate the results further. The result indicates that there is a relationship between the disaster human impact and the five socio-economic characteristics of the affected country mentioned above. As a result, a framework was put forward, which could predict the disaster’s human impact based on their severity rank in the early hours of disaster strike. The predictions in this model were outlined in two worst and best-case scenarios, which respectively inform the lower range and higher range of the prediction. A necessity to develop the predictive framework can be highlighted by noticing that despite the existing research in literature, a framework for predicting the human impact and estimating the needs at the time of the disaster is yet to be developed. This can further be used to allocate the resources at the response phase of the disaster where the data is scarce.

Keywords: disaster management, natural disaster, pattern recognition, prediction

Procedia PDF Downloads 133

Authors: Budhvin T. Withana, Sulochana Rupasinghe

Abstract:

The educational system faces a significant concern with regards to Dyslexia and Dysgraphia, which are learning disabilities impacting reading and writing abilities. This is particularly challenging for children who speak the Sinhala language due to its complexity and uniqueness. Commonly used methods to detect the risk of Dyslexia and Dysgraphia rely on subjective assessments, leading to limited coverage and time-consuming processes. Consequently, delays in diagnoses and missed opportunities for early intervention can occur. To address this issue, the project developed a hybrid model that incorporates various deep learning techniques to detect the risk of Dyslexia and Dysgraphia. Specifically, Resnet50, VGG16, and YOLOv8 models were integrated to identify handwriting issues. The outputs of these models were then combined with other input data and fed into an MLP model. Hyperparameters of the MLP model were fine-tuned using Grid Search CV, enabling the identification of optimal values for the model. This approach proved to be highly effective in accurately predicting the risk of Dyslexia and Dysgraphia, providing a valuable tool for early detection and intervention. The Resnet50 model exhibited a training accuracy of 0.9804 and a validation accuracy of 0.9653. The VGG16 model achieved a training accuracy of 0.9991 and a validation accuracy of 0.9891. The MLP model demonstrated impressive results with a training accuracy of 0.99918, a testing accuracy of 0.99223, and a loss of 0.01371. These outcomes showcase the high accuracy achieved by the proposed hybrid model in predicting the risk of Dyslexia and Dysgraphia.

Keywords: neural networks, risk detection system, dyslexia, dysgraphia, deep learning, learning disabilities, data science

Procedia PDF Downloads 36

Authors: Jyotismita Talukdar

Abstract:

In this paper, a study has been made on the possibility and accuracy of early prediction of several Heart Disease using Artificial Neural Network. (ANN). The study has been made in both noise free environment and noisy environment. The data collected for this analysis are from five Hospitals. Around 1500 heart patient’s data has been collected and studied. The data is analysed and the results have been compared with the Doctor’s diagnosis. It is found that, in noise free environment, the accuracy varies from 74% to 92%and in noisy environment (2dB), the results of accuracy varies from 62% to 82%. In the present study, four basic attributes considered are Blood Pressure (BP), Fasting Blood Sugar (FBS), Thalach (THAL) and Cholesterol (CHOL.). It has been found that highest accuracy(93%), has been achieved in case of PPI( Post-Permanent-Pacemaker Implementation ), around 79% in case of CAD(Coronary Artery disease), 87% in DCM (Dilated Cardiomyopathy), 89% in case of RHD&MS(Rheumatic heart disease with Mitral Stenosis), 75 % in case of RBBB +LAFB (Right Bundle Branch Block + Left Anterior Fascicular Block), 72% for CHB(Complete Heart Block) etc. The lowest accuracy has been obtained in case of ICMP (Ischemic Cardiomyopathy), about 38% and AF( Atrial Fibrillation), about 60 to 62%.

Keywords: coronary heart disease, chronic stable angina, sick sinus syndrome, cardiovascular disease, cholesterol, Thalach

Procedia PDF Downloads 156

Authors: Nikki Berrange, Gizelle Willows

Abstract:

Business rescue has become increasingly popular since its inclusion in the Companies Act of South Africa in May 2011. The Alternate Exchange (AltX) of the Johannesburg Stock Exchange has experienced a marked increase in the number of companies entering business rescue. This study sampled twenty companies listed on the AltX to determine whether Altman’s Z-score model for emerging markets (ZEM) or Taffler’s Z-score model is a more accurate model in predicting financial distress for small to medium size companies in South Africa. The study was performed over three different time horizons; one, two and three years prior to the event of financial distress, in order to determine how many companies each model predicted would be unlikely to succeed as well as the predictive ability and accuracy of the respective models. The study found that Taffler’s Z-score model had a greater ability at predicting financial distress from all three-time horizons.

Keywords: Altman’s ZEM-score, Altman’s Z-score, AltX, business rescue, Taffler’s Z-score

Procedia PDF Downloads 332

Authors: Chad Goldsworthy, B. Rajeswari Matam

Abstract:

The use of deep learning for species identification in camera trap images has revolutionised our ability to study, conserve and monitor species in a highly efficient and unobtrusive manner, with state-of-the-art models achieving accuracies surpassing the accuracy of manual human classification. The high imbalance of camera trap datasets, however, results in poor accuracies for minority (rare or endangered) species due to their relative insignificance to the overall model accuracy. This paper investigates the use of Focal Loss, in comparison to the traditional Cross Entropy Loss function, to improve the identification of minority species in the “255 Bird Species” dataset from Kaggle. The results show that, although Focal Loss slightly decreased the accuracy of the majority species, it was able to increase the F1-score by 0.06 and improve the identification of the bottom two, five and ten (minority) species by 37.5%, 15.7% and 10.8%, respectively, as well as resulting in an improved overall accuracy of 2.96%.

Keywords: convolutional neural networks, data imbalance, deep learning, focal loss, species classification, wildlife conservation

Procedia PDF Downloads 156

Authors: Mia Françoise

Abstract:

This report aims to develop a strategy for assisting short-term investors to benefit from the current economic climate in South Africa by utilizing technical analysis techniques and predictive analytics. As part of this research, value investing and technical analysis principles will be combined to maximize returns for South African investors while optimizing volatility. As an emerging market, South Africa offers many opportunities for high growth in sectors where other developed countries cannot grow at the same rate. Investing in South African companies with significant growth potential can be extremely rewarding. Although the risk involved is more significant in countries with less developed markets and infrastructure, there is more room for growth in these countries. According to recent research, the offshore market is expected to outperform the local market over the long term; however, short-term investments in the local market will likely be more profitable, as the Johannesburg Stock Exchange is predicted to outperform the S&P500 over the short term. The instabilities in the economy contribute to increased market volatility, which can benefit investors if appropriately utilized. Price prediction and portfolio optimization comprise the two primary components of this methodology. As part of this process, statistics and other predictive modeling techniques will be used to predict the future performance of stocks listed on the Johannesburg Stock Exchange. Following predictive data analysis, Modern Portfolio Theory, based on Markowitz's Mean-Variance Theorem, will be applied to optimize the allocation of assets within an investment portfolio. By combining different assets within an investment portfolio, this optimization method produces a portfolio with an optimal ratio of expected risk to expected return. This methodology aims to provide a short-term investment with a stock portfolio that offers the best risk-to-return profile for stocks listed on the JSE by combining price prediction and portfolio optimization.

Keywords: financial stocks, optimized asset allocation, prediction modelling, South Africa

Procedia PDF Downloads 67

Authors: Deokho Kim, Wonwoo Lee, Jaewoong Lee, Teresa Ng, Gun-Ill Lee, Jiwon Jeong

Abstract:

This paper presents a method for improving object search accuracy using a deep learning model. A major limitation to provide accurate similarity with deep learning is the requirement of huge amount of data for training pairwise similarity scores (metrics), which is impractical to collect. Thus, similarity scores are usually trained with a relatively small dataset, which comes from a different domain, causing limited accuracy on measuring similarity. For this reason, this paper proposes a deep learning model that can be trained with a significantly small amount of data, a clustered data which of each cluster contains a set of visually similar images. In order to measure similarity distance with the proposed method, visual features of two images are extracted from intermediate layers of a convolutional neural network with various pooling methods, and the network is trained with pairwise similarity scores which is defined zero for images in identical cluster. The proposed method outperforms the state-of-the-art object similarity scoring techniques on evaluation for finding exact items. The proposed method achieves 86.5% of accuracy compared to the accuracy of the state-of-the-art technique, which is 59.9%. That is, an exact item can be found among four retrieved images with an accuracy of 86.5%, and the rest can possibly be similar products more than the accuracy. Therefore, the proposed method can greatly reduce the amount of training data with an order of magnitude as well as providing a reliable similarity metric.

Keywords: visual search, deep learning, convolutional neural network, machine learning

Procedia PDF Downloads 193

Authors: Mikiyas Adugna

Abstract:

Effort estimation is among the activities carried out in software development processes. An accurate model of estimation leads to project success. The method of agile effort estimation is a complex task because of the dynamic nature of software development. Researchers are still conducting studies on agile effort estimation to enhance prediction accuracy. Due to these reasons, we investigated and proposed a model on LASSO and Elastic Net regression to enhance estimation accuracy. The proposed model has major components: preprocessing, train-test split, training with default parameters, and cross-validation. During the preprocessing phase, the entire dataset is normalized. After normalization, a train-test split is performed on the dataset, setting training at 80% and testing set to 20%. We chose two different phases for training the two algorithms (Elastic Net and LASSO) regression following the train-test-split. In the first phase, the two algorithms are trained using their default parameters and evaluated on the testing data. In the second phase, the grid search technique (the grid is used to search for tuning and select optimum parameters) and 5-fold cross-validation to get the final trained model. Finally, the final trained model is evaluated using the testing set. The experimental work is applied to the agile story point dataset of 21 software projects collected from six firms. The results show that both Elastic Net and LASSO regression outperformed the compared ones. Compared to the proposed algorithms, LASSO regression achieved better predictive performance and has acquired PRED (8%) and PRED (25%) results of 100.0, MMRE of 0.0491, MMER of 0.0551, MdMRE of 0.0593, MdMER of 0.063, and MSE of 0.0007. The result implies LASSO regression algorithm trained model is the most acceptable, and higher estimation performance exists in the literature.

Keywords: agile software development, effort estimation, elastic net regression, LASSO

Procedia PDF Downloads 33

Authors: Thidathit Prachanukool, Chaiyaporn Yuksen, Welawat Tienpratarn, Sorravit Savatmongkorngul, Panvilai Tangkulpanich, Chetsadakon Jenpanitpong, Yuranan Phootothum, Malivan Phontabtim, Promphet Nuanprom

Abstract:

Background: Ruptured appendicitis has a high morbidity and mortality and requires immediate surgery. The Alvarado Score is used as a tool to predict the risk of acute appendicitis, but there is no such score for predicting rupture. This study aimed to developed the prediction score to determine the likelihood of ruptured appendicitis in an Asian population. Methods: This study was diagnostic, retrospectively cross-sectional and exploratory model at the Emergency Medicine Department in Ramathibodi Hospital between March 2016 and March 2018. The inclusion criteria were age >15 years and an available pathology report after appendectomy. Clinical factors included gender, age>60 years, right lower quadrant pain, migratory pain, nausea and/or vomiting, diarrhea, anorexia, fever>37.3°C, rebound tenderness, guarding, white blood cell count, polymorphonuclear white blood cells (PMN)>75%, and the pain duration before presentation. The predictive model and prediction score for ruptured appendicitis was developed by multivariable logistic regression analysis. Result: During the study period, 480 patients met the inclusion criteria; of these, 77 (16%) had ruptured appendicitis. Five independent factors were predictive of rupture, age>60 years, fever>37.3°C, guarding, PMN>75%, and duration of pain>24 hours to presentation. A score > 6 increased the likelihood ratio of ruptured appendicitis by 3.88 times. Conclusion: Using the Ramathibodi Welawat Ruptured Appendicitis Score. (RAMA WeRA Score) developed in this study, a score of > 6 was associated with ruptured appendicitis.

Keywords: predictive model, risk score, ruptured appendicitis, emergency room

Procedia PDF Downloads 143

Authors: Suzan Anwar, Utkarsh Goyal

Abstract:

Heart rate (HR) is a predictor of cardiovascular, cerebrovascular, and all-cause mortality in the general population, as well as in patients with cardio and cerebrovascular diseases. Machine learning (ML) significantly improves the accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment while avoiding unnecessary treatment of others. This research examines relationship between the individual's various heart health inputs like age, sex, cp, trestbps, thalach, oldpeaketc, and the likelihood of developing heart disease. Machine learning techniques like logistic regression and decision tree, and Python are used. The results of testing and evaluating the model using the Heart Failure Prediction Dataset show the chance of a person having a heart disease with variable accuracy. Logistic regression has yielded an accuracy of 80.48% without data handling. With data handling (normalization, standardscaler), the logistic regression resulted in improved accuracy of 87.80%, decision tree 100%, random forest 100%, and SVM 100%.

Keywords: heart rate, machine learning, SVM, decision tree, logistic regression, random forest

Procedia PDF Downloads 118

Authors: D.Mendes, J. Henriques, P. Carvalho, T. Rocha, S. Paredes, R. Cabiddu, R. Trimer, R. Mendes, A. Borghi-Silva, L. Kaminsky, E. Ashley, R. Arena, J. Myers

Abstract:

The continued exploration of clinically relevant predictive models continues to be an important pursuit. Cardiorespiratory fitness (CRF) portends clinical vital information and as such its accurate prediction is of high importance. Therefore, the aim of the current study was to develop a data-driven model, based on computational intelligence techniques and, in particular, clustering approaches, to predict CRF. Two prediction models were implemented and compared: 1) the traditional Wasserman/Hansen Equations; and 2) an interpretable clustering approach. Data used for this analysis were from the 'FRIEND - Fitness Registry and the Importance of Exercise: The National Data Base'; in the present study a subset of 10690 apparently healthy individuals were utilized. The accuracy of the models was performed through the computation of sensitivity, specificity, and geometric mean values. The results show the superiority of the clustering approach in the accurate estimation of CRF (i.e., maximal oxygen consumption).

Keywords: cardiorespiratory fitness, data-driven models, knowledge extraction, machine learning

Procedia PDF Downloads 263

Authors: S. Bouharati, F. Allag, M. Belmahdi, M. Bounechada

Abstract:

In terms of ecology forecast effects of desertification, the purpose of this study is to develop a predictive model of growth and adaptation of species in arid environment and bioclimatic conditions. The impact of climate change and the desertification phenomena is the result of combined effects in magnitude and frequency of these phenomena. Like the data involved in the phytopathogenic process and bacteria growth in arid soil occur in an uncertain environment because of their complexity, it becomes necessary to have a suitable methodology for the analysis of these variables. The basic principles of fuzzy logic those are perfectly suited to this process. As input variables, we consider the physical parameters, soil type, bacteria nature, and plant species concerned. The result output variable is the adaptability of the species expressed by the growth rate or extinction. As a conclusion, we prevent the possible strategies for adaptation, with or without shifting areas of plantation and nature adequate vegetation.

Keywords: climate changes, dry soil, phytopathogenicity, predictive model, fuzzy logic

Procedia PDF Downloads 297

Authors: Ermias A. Tegegn, Million Meshesha

Abstract:

Human Immunodeficiency Virus and Acquired Immunodeficiency Syndrome (HIV/AIDS) is a major cause of death for most African countries. Ethiopia is one of the seriously affected countries in sub Saharan Africa. Previously in Ethiopia, having HIV/AIDS was almost equivalent to a death sentence. With the introduction of Antiretroviral Therapy (ART), HIV/AIDS has become chronic, but manageable disease. The study focused on a data mining technique to predict future living status of HIV/AIDS patients at the time of drug regimen change when the patients become toxic to the currently taking ART drug combination. The data is taken from University of Gondar Hospital ART program database. Hybrid methodology is followed to explore the application of data mining on ART program dataset. Data cleaning, handling missing values and data transformation were used for preprocessing the data. WEKA 3.7.9 data mining tools, classification algorithms, and expertise are utilized as means to address the research problem. By using four different classification algorithms, (i.e., J48 Classifier, PART rule induction, Naïve Bayes and Neural network) and by adjusting their parameters thirty-two models were built on the pre-processed University of Gondar ART program dataset. The performances of the models were evaluated using the standard metrics of accuracy, precision, recall, and F-measure. The most effective model to predict the status of HIV patients with drug regimen substitution is pruned J48 decision tree with a classification accuracy of 98.01%. This study extracts interesting attributes such as Ever taking Cotrim, Ever taking TbRx, CD4 count, Age, Weight, and Gender so as to predict the status of drug regimen substitution. The outcome of this study can be used as an assistant tool for the clinician to help them make more appropriate drug regimen substitution. Future research directions are forwarded to come up with an applicable system in the area of the study.

Keywords: HIV drug regimen, data mining, hybrid methodology, predictive model

Procedia PDF Downloads 116

Authors: Palwinder Singh, Munish Sandhir, Tejinder Singh

Abstract:

The ordinary differential equations (ODE) represent a natural framework for mathematical modeling of many real-life situations in the field of engineering, control systems, physics, chemistry and astronomy etc. Such type of differential equations can be solved by analytical methods or by numerical methods. If the solution is calculated using analytical methods, it is done through calculus theories, and thus requires a longer time to solve. In this paper, we compare the numerical accuracy of the solutions given by the three main types of one-step initial value solvers: Taylor’s Series Method, Euler’s Method and Runge-Kutta Fourth Order Method (RK4). The comparison of accuracy is obtained through comparing the solutions of ordinary differential equation given by these three methods. Furthermore, to verify the accuracy; we compare these numerical solutions with the exact solutions.

Keywords: Ordinary differential equations (ODE), Taylor’s Series Method, Euler’s Method, Runge-Kutta Fourth Order Method

Procedia PDF Downloads 329

Authors: Mohammad Hadi Mahmoodi, Elahe Saedpanah

Abstract:

Since pronunciation is among those factors that can have strong effects on EFL learners’ successful communication, instructional programs with accurate pronunciation purposes seem to be a necessity in any L2 teaching context. The widespread use of smart mobile phones brings with itself various educational applications, which can assist foreign language learners in learning and speaking another language other than their L1. In line with this supportive innovation, the present study investigated the role of LDOCE (Longman Dictionary of Contemporary English), a mobile application, on improving Iranian EFL learners’ pronunciation accuracy. To this aim, 40 EFL learners studying English at the intermediate level participated in the current study. This was an experimental research with two groups of 20 students in an experimental and a control group. The data were collected through the administration of a pronunciation pretest before the instruction and a post-test after the treatment. In addition, the assessment was based on the pupils’ recorded voices while reading the selected words. The results of the independent samples t-test indicated that using LDOCE significantly affected Iranian EFL learners' pronunciation accuracy with those in the experimental group outperforming their control group counterparts.

Keywords: LDOCE, EFL learners, pronunciation accuracy, CALL, MALL

Procedia PDF Downloads 524