Feature Selection in Machine Learning: Methods and Comparison

Kaur, Amandeep; Guleria, Kalpna; Trivedi, Naresh Kumar

doi:10.1109/icacite51222.2021.9404623

Cited by 54 publications

(12 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The first method known as filter which typically rely on some filter index like statistical measures or scoring techniques to evaluate the relevance of variables, independent of any specific predictive model. Common filter techniques include correlation analysis, mutual information, chi-square test, and information gain [13]. The variables are ranked or assigned a score based on these measures, and a subset of top-ranking variables is selected.…”

Section: Variable Selection Methods In Machine Learningmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of different variable selection methods for predicting the occurrence of Metisa Plana in oil palm plantation using machine learning

Wang,

Idris,

Muharam

et al. 2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

Monitoring and predicting the spatio-temporal distribution of crop pests and assessing related risks are crucial for effective pest management strategies. Machine learning techniques have shown potential in analysing agricultural data and providing accurate predictions. Variable selection plays a critical role in crop pest analysis by identifying the most informative and influential features that contribute to pest distribution and risk prediction. The current practice of choosing variable selection methods is mostly based on previous experience and may involve a certain degree of subjectivity. This paper aims to provide empirical comparisons of different variable selection methods for machine learning applications in crop pest spatio-temporal distribution and risk prediction. This study conducted various variable selection methods, including filter methods (information gain, chi-square test, mutual information), wrapper methods (RFE), and embedded methods (Random Forest), using worms pest (Metisa plana) in oil palm trees as the experimental subject. The initial set of variables included bioclimatic, vegetation indices, and terrain variables. The experimental results indicated that there was some overlap in the selected variables across different methods, bioclimatic variables (rainfall (RF), relative humidity (RH)) were selected as important variables by different methods; non-important variables like NDVI and elevation when added to the ANN modelling can clearly contribute to the improvement in prediction accuracy. These empirical findings can provide guidance for relevant data monitoring in the prediction of crop pest and disease outbreaks. Additionally, the results can serve as a reference for variable selection in spatiotemporal prediction of pests and diseases in other agricultural and forestry crops.

show abstract

Section: Variable Selection Methods In Machine Learningmentioning

confidence: 99%

“…Among various variable selection methods, different types of methods have different scopes of application and conditions for use [13,14]. Filter methods are independent of specific machine learning algorithms and are applied during the pre-processing stage.…”

Section: Introductionmentioning

confidence: 99%

Comparison of different variable selection methods for predicting the occurrence of Metisa Plana in oil palm plantation using machine learning

Wang,

Idris,

Muharam

et al. 2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

show abstract

“…Nilai chi-square yang tinggi menunjukkan suatu fitur memiliki hubungan yang signifikan dengan kelas target [50], [54]- [57]. Chi-square umumnya diterapkan pada data kategori ataupun campuran [37], [49], [58]. Berikut adalah persamaan dari chi-square:…”

Section: Chi-squareunclassified

Seleksi Fitur pada Supervised Learning: Klasifikasi Prestasi Belajar Mahasiswa Saat dan Pasca Pandemi COVID-19

Rahmadeyan

Mustakim

2023

TEKNOSI

View full text Add to dashboard Cite

Dampak pandemi COVID-19 membuat lembaga pendidikan mengubah metode belajar menjadi pembelajaran jarak jauh secara daring. Banyak perguruan tinggi menyatakan keprihatinannya pada prestasi akademik mahasiswanya selama selama periode tersebut, namun disisi lain terdapat mahasiswa yang merasa puas dan nyaman. Di masa pasca pandemi terjadi transisi bertahap untuk kembali ke pembelajaran tatap muka. Ini dilakukan karena pembelajaran tatap muka dianggap lebih efektif dibandingkan pembelajaran daring. Untuk meningkatkan dan memantau kemajuan prestasi akademik mahasiswa demi menghasilkan lulusan yang berkualitas, maka diperlukan analisis terkait perilaku dan prestasi belajar mahaiswa, salah satunya dengan menggunakan teknik data mining. Penelitian ini bertujuan untuk menemukan pola dan faktor yang mempengaruhi prestasi akademik mahasiswa saat dan pasca pandemi COVID-19. Chi-Square dan Mutual Information diterapkan sebagai seleksi fitur untuk menentukan fitur paling berpengaruh pada data. Data dengan fitur optimal akan dilakukan klasifikasi dengan algoritma NBC, CART, RF, dan SVM. Berdasarkan hasil dan analisis yang dilakukan, dapat disimpulkan seleksi fitur efektif dalam meningkatkan kemampuan model dan mempercepat waktu komputasi. Pemodelan dengan 4 algoritma dan 2 metode seleksi fitur menghasilkan CART dengan Chi-Square pada 2 fitur sebagai model terbaik dengan akurasi 89,00%, precision 87,72%, recall 93,57% dan waktu komputasi 0,01814s. Dibandingkan tanpa seleksi fitur, performa CART dengan Chi-Square mengalami peningkatan akurasi sebesar 3% dan waktu komputasi 0,00629s. Chi-Square menjadi seleksi fitur yang efektif pada penelitian ini, yang mana Chi-Square unggul pada rara-rata recall dan waktu komputasi dibandingkan Mutual Information.

show abstract

“…According to the relationship with classifier, feature selection methods can be divided into four types: filter, wrapper, embedded and hybrid [3]. The filter method mainly defines the contribution of each feature and sorts them according to the correlation between features and classification tags.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Feature Selection Method Based on Feature Subset and Factor Analysis

et al. 2022

View full text Add to dashboard Cite

With the advent of big data era and the rapid improvement of raw data scale, feature selection is the basic and critical technologies for data mining. However, in most of the studies on feature selection methods before, mainly directed to treat the single feature or overall feature subset, while the influence of the correlation and redundancy of features in the feature subset on the classification results is ignored. In this paper, combination of feature subset grouping and factor analysis (FA), a hybrid feature selection method based on feature subsets generation through factor analysis (FAFS_HFS) is proposed. Firstly, generate feature subsets based on the maximum load (maximum explanatory power) of each feature through factor analysis. Then, minimum redundancy maximum relevance (mRMR) and sequential forward selection (SFS) is used to remove the redundancy of each feature subset. Finally, fisher score (F-score) and feature subset SFS (FS_SFS) was utilized to evaluate and select feature subset, and obtain the optimal feature subset. Experiments are conducted on 14 datasets, the results show that FAFS_HFS has high classification accuracy and dimension reduction on almost all datasets, especially in high-dimensional datasets, and it has excellent efficiency and competitive classification performance compared with other contrastive methods.INDEX TERMS hybrid feature selection, feature subset, factor analysis (FA), mRMR, F-score, sequential forward selection (SFS)

show abstract

Feature Selection in Machine Learning: Methods and Comparison

Cited by 54 publications

References 23 publications

Comparison of different variable selection methods for predicting the occurrence of Metisa Plana in oil palm plantation using machine learning

Comparison of different variable selection methods for predicting the occurrence of Metisa Plana in oil palm plantation using machine learning

Seleksi Fitur pada Supervised Learning: Klasifikasi Prestasi Belajar Mahasiswa Saat dan Pasca Pandemi COVID-19

Hybrid Feature Selection Method Based on Feature Subset and Factor Analysis

Contact Info

Product

Resources

About