Impact of Data Normalization on Classification Model Accuracy

Borkin, Dmitrii; Nemethova, Andrea; Michalconok, German; Maiorov, Konstantin

doi:10.2478/rput-2019-0029

Cited by 42 publications

(19 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Patient allergic history is commonly associated cause of adverse effects to many types of drugs and vaccines 13 , and in the case of COVID-19 this has also been reported 16,24 ; allergic-related reactions are found to a significant degree in every data group used in our study. Patient age is another important aspect, where the mortality rate in advanced old-age persons is comparatively higher than in other cases, and previous findings also supported this evidence 15 .…”

Section: Discussionsupporting

confidence: 72%

“…Firstly, we calculated the feature importance scores for each distinct feature, for individual machine learning classifiers, and then we normalized the values to render the data with the same scale, i.e. between 0 and 1, by using the min-max normalization technique 13 . This was followed by the mean aggregation of those values as shown in Figure 6.…”

Section: Feature Importance Analysis For Finding Significant Features Using Machine Learning Classifiersmentioning

confidence: 99%

See 1 more Smart Citation

Adverse effects of COVID-19 vaccination: machine learning and statistical approach to identify and classify incidences of morbidity and post-vaccination reactogenicity

Ahamad

Aktar

Uddin

et al. 2021

Preprint

View full text Add to dashboard Cite

Good vaccine safety and reliability are essential to prevent infectious disease spread. A small but significant number of apparent adverse reactions to the new COVID-19 vaccines have been reported. Here, we aim to identify possible common causes for such adverse reactions with a view to enabling strategies that reduce patient risk by using patient data to classify and characterise patients those at risk of such reactions. We examined patient medical histories and data documenting post-vaccination effects and outcomes. The data analyses were conducted by different statistical approaches followed by a set of machine learning classification algorithms. In most cases, similar features were significantly associated with poor patient reactions. These included patient prior illnesses, admission to hospitals and SARS-CoV-2 reinfection. The analyses indicated that patient age, gender, allergic history, taking other medications, type-2 diabetes, hypertension and heart disease are the most significant pre-existing factors associated with risk of poor outcome and long duration of hospital treatments, pyrexia, headache, dyspnoea, chills, fatigue, various kind of pain and dizziness are the most significant clinical predictors. The machine learning classifiers using medical history were also able to predict patients most likely to have complication-free vaccination with an accuracy score above 85%. Our study identifies profiles of individuals that may need extra monitoring and care (e.g., vaccination at a location with access to comprehensive clinical support) to reduce negative outcomes through classification approaches. Important classifiers achieving these reactions notably included allergic susceptibility and incidence of heart disease or type-2 diabetes.

show abstract

Section: Discussionsupporting

confidence: 72%

Section: Feature Importance Analysis For Finding Significant Features Using Machine Learning Classifiersmentioning

confidence: 99%

Adverse effects of COVID-19 vaccination: machine learning and statistical approach to identify and classify incidences of morbidity and post-vaccination reactogenicity

Ahamad

Aktar

Uddin

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The present study focused on the importance of other highly influential factors along with student academic results, such as the students per teacher ratio, the number of schools in a region, whether schools were located in rural or urban areas, the availability or lack of classrooms, electrical facilities in schools, availability or lack of furniture for students, open-air classes, computer lab facilities, science labs, and playgrounds in schools. Previous research [44][45][46][47][48] suggested that data pre-processing (normalisation, discretisation) techniques enhanced classifier performance, as these techniques reduce the biases among features. Furthermore, related studies showed that the min-max normalisation method performed better than other data normalisation methods [49][50][51].…”

Section: Discussionmentioning

confidence: 99%

A Novel Method for Performance Measurement of Public Educational Institutions Using Machine Learning Models

et al. 2021

View full text Add to dashboard Cite

Lack of education is a major concern in underdeveloped countries because it leads to poor human and economic development. The level of education in public institutions varies across all regions around the globe. Current disparities in access to education worldwide are mostly due to systemic regional differences and the distribution of resources. Previous research focused on evaluating students’ academic performance, but less has been done to measure the performance of educational institutions. Key performance indicators for the evaluation of institutional performance differ from student performance indicators. There is a dire need to evaluate educational institutions’ performance based on their disparities and academic results on a large scale. This study proposes a model to measure institutional performance based on key performance indicators through data mining techniques. Various feature selection methods were used to extract the key performance indicators. Several machine learning models, namely, J48 decision tree, support vector machines, random forest, rotation forest, and artificial neural networks were employed to build an efficient model. The results of the study were based on different factors, i.e., the number of schools in a specific region, teachers, school locations, enrolment, and availability of necessary facilities that contribute to school performance. It was also observed that urban regions performed well compared to rural regions due to the improved availability of educational facilities and resources. The results showed that artificial neural networks outperformed other models and achieved an accuracy of 82.9% when the relief-F based feature selection method was used. This study will help support efforts in governance for performance monitoring, policy formulation, target-setting, evaluation, and reform to address the issues and challenges in education worldwide.

show abstract

“…Python software is used to generate the targets, namely Fc, e c , and T The full factorial design for the four factors and 8 levels for each factor gives 4096 training data. The normalization is used to make the data with the same scale, the training data are normalized in the interval [0,1] using equations (3), this improves the regression and classification model's performance and training stability [14], [15].…”

Section: A Data Preparationmentioning

confidence: 99%

Cutting Forces and Temperature Optimization in Turning using a Predictive Machining Theory, ANN, and MOGA

Outemsaa¹,

Farissi²,

Hamouti³

et al. 2022

IJETAE

View full text Add to dashboard Cite

To minimise stresses on the tool and workpiece, such as wear, thermal effect, workpiece stresses, cutting power, etc., the cutting force and the heat in the cutting area should be minimised. This work aims to introduce an artificial intelligence tool, more precisely the neural network, to achieve optimized cutting conditions. Oxley cutting modelling in conjunction with Johnson-Cook of an AISI 1045 material is converted to an artificial neural network model which will be used to determine a fitness function to be optimized. The Artificial Neural Network is constructed based on the training data collected from the predictive model of Oxley and JC, the choice of the most accurate ANN of minimal MSE= 0.001108 is based on a specific method of tuning the hyperparameters which result in an architecture; two hidden layers, 25 neurons for each hidden layer, a sigmoid activation function, a trainlm learning algorithm, and a learning rate of 0.01. A multi-objective optimization is performed using the MATLAB tool to obtain the optimum values for cutting velocity Vc, advance f, penetration depth ap, and cutting angle of the tool. It is found that the neural network is a more rapid calculation of cutting conditions such as shear zone forces, shear zone temperatures, and others. contrary to the Oxley and JC mathematical model which will require a lot of calculations. The optimum values for cutting conditions are 208 mm/min for cutting speed, 0.06mm/rev for f, 0.38 for ap, and 10° for clearance angle.

show abstract

Impact of Data Normalization on Classification Model Accuracy

Cited by 42 publications

References 6 publications

Adverse effects of COVID-19 vaccination: machine learning and statistical approach to identify and classify incidences of morbidity and post-vaccination reactogenicity

Adverse effects of COVID-19 vaccination: machine learning and statistical approach to identify and classify incidences of morbidity and post-vaccination reactogenicity

A Novel Method for Performance Measurement of Public Educational Institutions Using Machine Learning Models

Cutting Forces and Temperature Optimization in Turning using a Predictive Machining Theory, ANN, and MOGA

Contact Info

Product

Resources

About