Comparison of Classification Models for Early Prediction of Breast Cancer

Ghani, Muhammad Usman; Alam, Talha Mahboob; Jaskani, Fawwad Hassan

doi:10.1109/icic48496.2019.8966691

Cited by 40 publications

(23 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is easy to visualize the decision tree through a 2-D image to understand the selection of significant features. Various researches also extract significant features for better performance [38,39]. However, it is more beneficial to calculate the importance of each feature to enhance the model's prediction.…”

Section: Feature Selection Through Gradient Boosted Decision Tree Modelmentioning

confidence: 99%

An Investigation of Credit Card Default Prediction in the Imbalanced Datasets

et al. 2020

Self Cite

View full text Add to dashboard Cite

Financial threats are displaying a trend about the credit risk of commercial banks as the incredible improvement in the financial industry has arisen. In this way, one of the biggest threats faces by commercial banks is the risk prediction of credit clients. Recent studies mostly focus on enhancing the classifier performance for credit card default prediction rather than an interpretable model. In classification problems, an imbalanced dataset is also crucial to improve the performance of the model because most of the cases lied in one class, and only a few examples are in other categories. Traditional statistical approaches are not suitable to deal with imbalanced data. In this study, a model is developed for credit default prediction by employing various credit-related datasets. There is often a significant difference between the minimum and maximum values in different features, so Min-Max normalization is used to scale the features within one range. Data level resampling techniques are employed to overcome the problem of the data imbalance. Various undersampling and oversampling methods are used to resolve the issue of class imbalance. Different machine learning models are also employed to obtain efficient results. We developed the hypothesis of whether developed models using different machine learning techniques are significantly the same or different and whether resampling techniques significantly improves the performance of the proposed models. Oneway Analysis of Variance is a hypothesis-testing technique, used to test the significance of the results. The split method is utilized to validate the results in which data has split into training and test sets. The results on imbalanced datasets show the accuracy of 66.9% on Taiwan clients credit dataset, 70.7% on South German clients credit dataset, and 65% on Belgium clients credit dataset. Conversely, the results using our proposed methods significantly improve the accuracy of 89% on Taiwan clients credit dataset, 84.6% on South German clients credit dataset, and 87.1% on Belgium clients credit dataset. The results show that the performance of classifiers is better on the balanced dataset as compared to the imbalanced dataset. It is also observed that the performance of data oversampling techniques are better than undersampling techniques. Overall, the Gradient Boosted Decision Tree method performs better than other traditional machine learning classifiers. The Gradient Boosted Decision Tree method gives the best results while utilizing the K-means SMOTE oversampling method. Using one-way ANOVA, the null hypothesis was rejected by a p-value <0.001, hence confirming that the proposed model improved performance is statistical significance. The interpretable model is also deployed on the web to ease the different stakeholders. This model will help commercial banks, financial organizations, loan institutes, and other decision-makers to predict the loan defaulter earlier.

show abstract

Section: Feature Selection Through Gradient Boosted Decision Tree Modelmentioning

confidence: 99%

An Investigation of Credit Card Default Prediction in the Imbalanced Datasets

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Based on literature [ 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 ] in the biomedical area, a comparison was made between the different models implemented according to evaluation metrics described in this area of knowledge, taking into consideration that the values of these metrics are considered better as closer they are to unit.…”

Section: Resultsmentioning

confidence: 99%

Automatic Evaluation of Heart Condition According to the Sounds Emitted and Implementing Six Classification Methods

Soto-Murillo

Galván-Tejada

et al. 2021

Healthcare

View full text Add to dashboard Cite

The main cause of death in Mexico and the world is heart disease, and it will continue to lead the death rate in the next decade according to data from the World Health Organization (WHO) and the National Institute of Statistics and Geography (INEGI). Therefore, the objective of this work is to implement, compare and evaluate machine learning algorithms that are capable of classifying normal and abnormal heart sounds. Three different sounds were analyzed in this study; normal heart sounds, heart murmur sounds and extra systolic sounds, which were labeled as healthy sounds (normal sounds) and unhealthy sounds (murmur and extra systolic sounds). From these sounds, fifty-two features were calculated to create a numerical dataset; thirty-six statistical features, eight Linear Predictive Coding (LPC) coefficients and eight Cepstral Frequency-Mel Coefficients (MFCC). From this dataset two more were created; one normalized and one standardized. These datasets were analyzed with six classifiers: k-Nearest Neighbors, Naive Bayes, Decision Trees, Logistic Regression, Support Vector Machine and Artificial Neural Networks, all of them were evaluated with six metrics: accuracy, specificity, sensitivity, ROC curve, precision and F1-score, respectively. The performances of all the models were statistically significant, but the models that performed best for this problem were logistic regression for the standardized data set, with a specificity of 0.7500 and a ROC curve of 0.8405, logistic regression for the normalized data set, with a specificity of 0.7083 and a ROC curve of 0.8407, and Support Vector Machine with a lineal kernel for the non-normalized data; with a specificity of 0.6842 and a ROC curve of 0.7703. Both of these metrics are of utmost importance in evaluating the performance of computer-assisted diagnostic systems.

show abstract

“…In previous years, a lot of similar works has been done using data mining techniques in healthcare sector, the result of this work demonstrate the ability of applied algorithms that confirm the results those reported in [5], as the original creators of the used dataset, obtaining 95% confidence interval for sensitivity, specificity and AUC was (82%, 88%), (84%, 90%) and (0.87, 0.91) respectively (using SVM), also by Alickovic and Subasi [11] who used the Random Forest and Genetic Algorithm, obtaining the highest accuracy of 99.48 to diagnosis of breast cancer. In addition the authors in [12] applied different classification model to find the best biomarker for breast cancer. The result of artificial neural network comparatively better than the KNN algorithms results and get an accuracy of 80%.…”

Section: Discussionmentioning

confidence: 99%

Association Analysis of Obesity/Overweight and Breast Cancer Using Data Mining Techniques

Soufi

Ferdousi

2021

Front Health Inform

View full text Add to dashboard Cite

Introduction: Growing evidence has shown that some overweight factors could be implicated in tumor genesis, higher recurrence and mortality. In addition, association of various overweight factors and breast cancer has not been extensively explored. The goal of this research was to explore and evaluate the association of various overweight/obesity factors and breast cancer, based on obesity breast cancer data set.Material and Methods: Several studies show that a significantly stronger association is obvious between overweight and higher breast cancer incidence, but the role of some overweight factors such as BMI, insulin-resistance, Homeostasis Model Assessment (HOMA), Leptin, adiponectin, glucose and MCP.1 is still debatable, So for experiment of research work several clinical and biochemical overweight factors, including age, Body Mass Index (BMI), Glucose, Insulin, Homeostatic Model Assessment (HOMA), Leptin, Adiponectin, Resistin and Monocyte chemo attractant protein-1(MCP-1) were analyzed. Data mining algorithms including k-means, Apriori, Hierarchical clustering algorithm (HCM) were applied using orange version 3.22 as an open source data mining tool.Results: The Apriori algorithm generated a list of frequent item sets and some strong rules from dataset and found that insulin, HOMA and leptin are two items often simultaneously were seen for BC patients that leads to cancer progression. K-means algorithm applied and it divided samples on three clusters and its results showed that the pair of andlt;Adiponectin, MCP.1andgt; has the highest effect on seperation of clusters. In addition HCM was carried out and classified BC patients into 1-32 clusters to So this research apply HCM algorithm. We carried out hierarchical clustering with average linkage without purning and classified BC patients into 1–32 clusters in order to identify BC patients with similar charestrictics.Conclusion: These finding provide the employed algorithms in this study can be helpful to our aim.

show abstract

Comparison of Classification Models for Early Prediction of Breast Cancer

Cited by 40 publications

References 21 publications

An Investigation of Credit Card Default Prediction in the Imbalanced Datasets

An Investigation of Credit Card Default Prediction in the Imbalanced Datasets

Automatic Evaluation of Heart Condition According to the Sounds Emitted and Implementing Six Classification Methods

Association Analysis of Obesity/Overweight and Breast Cancer Using Data Mining Techniques

Contact Info

Product

Resources

About