An Improved Algorithm based on KNN and Random Forest

Liang, Jun; Liu, Qin; Nie, Nuihua; Zeng, Biqing; Zhang, Zan-Bo

doi:10.1145/3331453.3360963

Cited by 7 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These scores not only highlight which features are instrumental in the classification process but also provide valuable insights into the hierarchical importance of these features. The RF algorithm offers two ways to compute feature importance [25,26]: 1. Gini importance is calculated from the RF's structure, where each decision tree selects features based on criteria like Gini impurity or information gain for classification tasks and variance reduction for regression.…”

Section: Resultsmentioning

confidence: 99%

Comparison of Electrocardiographic Parameters by Gender in Heart Failure Patients with Preserved Ejection Fraction via Artificial Intelligence

Yilmaz,

Öz

2023

Diagnostics

View full text Add to dashboard Cite

Background: Heart failure (HF) causes high morbidity and mortality worldwide. The prevalence of HF with preserved ejection fraction (HFpEF) is increasing compared with HF with reduced ejection fraction (HFrEF). Patients with HFpEF are a patient group with a high rate of hospitalization despite medical treatment. Early diagnosis is very important in this group of patients, and early treatment can improve their prognosis. Although electrocardiographic (ECG) findings have been adequately studied in patients with HFrEF, there are not enough studies on these parameters in patients with HFpEF. There are very few studies in the literature, especially on gender-specific changes. The current research aims to compare gender-specific ECG parameters in patients with HFpEF based on the implications of artificial intelligence (AI). Methods: A total of 118 patients participated in the study, of which 66 (56%) were women with HFpEF and 52 (44%) were men with HFpEF. Demographic, echocardiographic, and electrocardiographic characteristics of the patients were analyzed to compare gender-specific ECG parameters in patients with HFpEF. The AI approach combined with machine learning approaches (gradient boosting machine, k-nearest neighbors, logistic regression, random forest, and support vector machines) was applied for distinguishing male patients with HFpEF from female patients with HFpEF. Results: After determining the parameters (demographic, echocardiographic, and electrocardiographic) to distinguish male patients with HFpEF from female patients with HFpEF, machine learning methods were applied, and among these methods, the random forest model achieved an average accuracy of 84.7%. The random forest algorithm results showed that smoking, P-wave dispersion, P-wave amplitude, T-end P/(PQ*Age), Cornell product, and P-wave duration were the most influential parameters for distinguishing male patients with HFpEF from female patients with HFpEF. Conclusions: The proposed model serves as a valuable tool for physicians, facilitating the diagnosis, treatment, and follow-up for distinguishing male patients with HFpEF from female patients with HFpEF. Analyzing readily accessible electrocardiographic parameters empowers medical professionals to make informed decisions and provide enhanced care to a wide range of individuals.

show abstract

Section: Resultsmentioning

confidence: 99%

Comparison of Electrocardiographic Parameters by Gender in Heart Failure Patients with Preserved Ejection Fraction via Artificial Intelligence

Yilmaz,

Öz

2023

Diagnostics

View full text Add to dashboard Cite

show abstract

“…Furthermore, we can see that LR is a linear classifier, its performance is poor when the feature space is large and there is correlation between features. Although RF classifier can discover the correlation of features, it has been proved by Liang et al 34 that there are overfitting problems in some noisy classification tasks. However, NB classifier has a simple…”

Section: Resultsmentioning

confidence: 99%

Feature transfer learning by reinforcement learning for detecting software defect

Guo

Wang

et al. 2022

Softw Pract Exp

View full text Add to dashboard Cite

Software defects, produced inevitably in software projects, seriously affect the efficiency of software testing and maintenance. An appealing solution is the software defect prediction (SDP) that has achieved good performance in many software projects. However, the difference between features and the difference of the same feature between training data and test data may degrade defect prediction performance if such differences violate the model's assumption. To address this issue, we propose a SDP method based on feature transfer learning (FTL), which performs a transformation sequence for each feature in order to map the original features to another feature space. Specifically, FTL first uses the reinforcement learning scheme that automatically learns a strategy for transferring the potential feature knowledge from the training data. Then, we use the learned feature knowledge to inspire the transformation of the test data. The classifier is trained by the transformed training data and predicts defects for transformed test data. We evaluate the validity of FTL on 43 projects from PROMISE and NASA MDP using three classifiers, logistic regression, random forest, and Naive Bayes (NB). Experimental results indicate that FTL is better than the original classifiers and has the best performance on the NB classifier. For PROMISE, after using FTL, the average results of F1-score, AUC, MCC are 0.601, 0.757, and 0.350 respectively, which are 24.9%, 2.6%, and 16.7% higher than the original NB classifier results. The number of projects with improved performance accounts for 83.87%, 83.87%, and 64.52%. Similarly, FTL performs well on NASA MDP. Besides, compared with four feature engineering (FE) methods, FTL achieves an excellent improvement on most projects and the average performance is also better than or close to the FE methods.

show abstract

“…Random Forest (RF) is a new and promising classifier by Breiman in 2001 [17]. Used as handling input variables that number in the thousands without having to delete variables and in their classification can estimate which variables are important [18].…”

Section: Introductionmentioning

confidence: 99%

Gender Classification for Anime Character Face Image Using Random Forest Classifier Method and GLCM Feature Extraction

Mulyana¹,

Pramansah²

2022

JUITA

View full text Add to dashboard Cite

Japan has many entertaining and unique artworks, especially its signature animation, called anime. Anime is an animation art that is unique in that the characterizations, characters, and storylines are made to resemble human life. The characters have 2 genders called male and female with unique visuals and are the characteristics of each anime character to entertain the audience. Training large-scale data and complex textures because not all of the anime images owned are of high quality, making classification by Machine Learning Algorithms low in accuracy. This study will describe an experiment using an anime face image dataset to classify the gender, namely male or female. From this problem, this research implements feature extraction to produce unique features of anime images with Gray-Level Cooccurrence Matrix (GLCM) and uses the Random Forest Classifier which is a classification algorithm in Machine Learning to classify gender. The results of this study get a good accuracy value of 95%, using 3,612 images where the test data used is 723 images and Homogeneity5 feature being the most relevant feature in increasing the accuracy value with a value of 0.06378389.

show abstract

An Improved Algorithm based on KNN and Random Forest

Cited by 7 publications

References 14 publications

Comparison of Electrocardiographic Parameters by Gender in Heart Failure Patients with Preserved Ejection Fraction via Artificial Intelligence

Comparison of Electrocardiographic Parameters by Gender in Heart Failure Patients with Preserved Ejection Fraction via Artificial Intelligence

Feature transfer learning by reinforcement learning for detecting software defect

Gender Classification for Anime Character Face Image Using Random Forest Classifier Method and GLCM Feature Extraction

Contact Info

Product

Resources

About