scholarly journals Determining Threshold Value on Information Gain Feature Selection to Increase Speed and Prediction Accuracy of Random Forest

2020 ◽  
Author(s):  
Maria Irmina Prasetiyowati ◽  
Nur Ulfa Maulidevi ◽  
Kridanto Surendro
Keyword(s):  
Feature Selection ◽  
Random Forest ◽  
Information Gain ◽  
Threshold Value ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Random Forest Classification ◽  
Average Value ◽  
Time Required

Abstract Feature selection is a preprocessing technique aims to remove the unnecessary features and speed up the algorithm's work process. One of the feature selection techniques is by calculating the information gain value of each feature in a dataset. From the information gain value obtained, then the determined threshold value will be used to make feature selection. Generally, the threshold value is used freely, or using a value of 0.05. This study proposed the determination of the threshold value using the standard deviation of the information gain value generated by each feature in the dataset. The determination of this threshold value was tested on ten original datasets and datasets that had been transformed by FFT and IFFT, then classified using Random Forest. The results of the average value of accuracy and the average time required from the Random Forest classification using the proposed threshold value are better compared to the results of feature selection with a threshold value of 0.05 and the Correlation-Base Feature Selection algorithm. Likewise, the result of the average accuracy value of the proposed threshold using a transformed dataset in terms are better than the threshold value of 0.05 and the Correlation-Base Feature Selection algorithm. However, the calculation results for the average time required are higher (slower).

A hybrid feature selection algorithm combining ReliefF and Particle swarm optimization for high-dimensional medical data

2021 ◽  
pp. 1-15
Author(s):  
Zhaozhao Xu ◽  
Derong Shen ◽  
Yue Kou ◽  
Tiezheng Nie
Keyword(s):  
Feature Selection ◽  
Particle Swarm Optimization ◽  
Random Forest ◽  
Classification Accuracy ◽  
Particle Swarm ◽  
Medical Data ◽  
High Dimensional ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Swarm Optimization

Due to high-dimensional feature and strong correlation of features, the classification accuracy of medical data is not as good enough as expected. feature selection is a common algorithm to solve this problem, and selects effective features by reducing the dimensionality of high-dimensional data. However, traditional feature selection algorithms have the blindness of threshold setting and the search algorithms are liable to fall into a local optimal solution. Based on it, this paper proposes a hybrid feature selection algorithm combining ReliefF and Particle swarm optimization. The algorithm is mainly divided into three parts: Firstly, the ReliefF is used to calculate the feature weight, and the features are ranked by the weight. Then ranking feature is grouped according to the density equalization, where the density of features in each group is the same. Finally, the Particle Swarm Optimization algorithm is used to search the ranking feature groups, and the feature selection is performed according to a new fitness function. Experimental results show that the random forest has the highest classification accuracy on the features selected. More importantly, it has the least number of features. In addition, experimental results on 2 medical datasets show that the average accuracy of random forest reaches 90.20%, which proves that the hybrid algorithm has a certain application value.

scholarly journals A feature selection algorithm combining information gain and multi-objective genetic search for intrusion detection system

2021 ◽  
Vol 336 ◽  
pp. 08008
Author(s):  
Tao Xie
Keyword(s):  
Feature Selection ◽  
Intrusion Detection ◽  
Intrusion Detection System ◽  
Detection Rate ◽  
Information Gain ◽  
Detection System ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Genetic Search ◽  
Multi Objective

In order to improve the detection rate and speed of intrusion detection system, this paper proposes a feature selection algorithm. The algorithm uses information gain to rank the features in descending order, and then uses a multi-objective genetic algorithm to gradually search the ranking features to find the optimal feature combination. We classified the Kddcup98 dataset into five classes, DOS, PROBE, R2L, and U2R, and conducted numerous experiments on each class. Experimental results show that for each class of attack, the proposed algorithm can not only speed up the feature selection, but also significantly improve the detection rate of the algorithm.

Feature Selection Algorithm Application in Near-Infrared Spectroscopy Classification Based on Binary Search Combined with Random Forest Pruning

2017 ◽  
Vol 54 (10) ◽  
pp. 103001
Author(s):  
刘 明 Liu Ming ◽  
李忠任 Li Zhongren ◽  
张海涛 Zhang Haitao ◽  
于春霞 Yu Chunxia ◽  
唐兴宏 Tang Xinghong ◽  
...  
Keyword(s):  
Feature Selection ◽  
Infrared Spectroscopy ◽  
Random Forest ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Binary Search ◽  
Selection Algorithm ◽  
Feature Selection Algorithm

Efficient Classification of DDoS Attacks Using an Ensemble Feature Selection Algorithm

2017 ◽  
Vol 29 (1) ◽  
pp. 71-83 ◽  
Author(s):  
Khundrakpam Johnson Singh ◽  
Tanmay De
Keyword(s):  
Feature Selection ◽  
Denial Of Service ◽  
Computation Time ◽  
Threshold Value ◽  
Ddos Attacks ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Cyber Threats ◽  
The Given

Abstract In the current cyber world, one of the most severe cyber threats are distributed denial of service (DDoS) attacks, which make websites and other online resources unavailable to legitimate clients. It is different from other cyber threats that breach security parameters; however, DDoS is a short-term attack that brings down the server temporarily. Appropriate selection of features plays a crucial role for effective detection of DDoS attacks. Too many irrelevant features not only produce unrelated class categories but also increase computation overhead. In this article, we propose an ensemble feature selection algorithm to determine which attribute in the given training datasets is efficient in categorizing the classes. The result of the ensemble algorithm when compared to a threshold value will enable us to decide the features. The selected features are deployed as training inputs for various classifiers to select a classifier that yields maximum accuracy. We use a multilayer perceptron classifier as the final classifier, as it provides better accuracy when compared to other conventional classification models. The proposed method classifies the new datasets into either attack or normal classes with an efficiency of 98.3% and also reduces the overall computation time. We use the CAIDA 2007 dataset to evaluate the performance of the proposed method using MATLAB and Weka 3.6 simulators.

scholarly journals A Systematic Algorithm for Data Cluster Using Map-Reduce Approach

2019 ◽  
pp. 564-569
Author(s):  
Kechika. S ◽  
Sapthika. B ◽  
Keerthana. B ◽  
Abinaya. S ◽  
Abdulfaiz. A
Keyword(s):  
Feature Selection ◽  
Map Reduce ◽  
Features Selection ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Main Concept ◽  
Selection Approach ◽  
Clustering Data ◽  
Data Objects ◽  
Time Required

We have been studying the problem clustering data objects as we have implemented a new algorithm called algorithm of clustering data using map reduce approach. In cluster, main part is feature selection which involves in recognition of set of features of a subset, since feature selection is considered as a important process. They also produces the approximate and according requests with the original set of features used in this type of approach. The main concept beyond this paper is to give the outcome of the clustering features. This paper which also gives the knowledge about cluster and it's own process. To processing of large datasets the nature of clustering where some more concepts are more helpful and important in a clustering process. In a clustering methodology where more concepts are very useful. The feature selection algorithm which affects, the entire process of clustering is the map-reduce concept. since, feature selection or extraction which is also used in map-reduce approach. The most desirable component is time complexity where efficiency concerns in this criterion. Here time required to find the effective features, where features of quality subsets is equal to effectiveness. The complexity to find based on this criteria based map-reduce features selection approach, which is proposed and evaluated in this paper.

scholarly journals Classification of Diabetes using Random Forest with Feature Selection Algorithm

2019 ◽  
Vol 9 (1) ◽  
pp. 1295-1300 ◽  
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Random Forest ◽  
Electronic Health Records ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Health Records

Diabetes has become a serious problem now a day. So there is a need to take serious precautions to eradicate this. To eradicate, we should know the level of occurrence. In this project we predict the level of occurrence of diabetes. We predict the level of occurrence of diabetes using Random Forest, a Machine Learning Algorithm. Using the patient’s Electronic Health Records (EHR) we can build accurate models that predict the presence of diabetes.

MRF-RFS: A Modified Random Forest Recursive Feature Selection Algorithm for Nasopharyngeal Carcinoma Segmentation

2020 ◽  
Vol 59 (04/05) ◽  
pp. 151-161
Author(s):  
Yuchen Fei ◽  
Fengyu Zhang ◽  
Chen Zu ◽  
Mei Hong ◽  
Xingchen Peng ◽  
...  
Keyword(s):  
Feature Selection ◽  
Random Forest ◽  
Nasopharyngeal Carcinoma ◽  
Soft Tissues ◽  
Feature Selection Method ◽  
Selection Method ◽  
Feature Subset ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Tumor Margins

Abstract Background An accurate and reproducible method to delineate tumor margins is of great importance in clinical diagnosis and treatment. In nasopharyngeal carcinoma (NPC), due to limitations such as high variability, low contrast, and discontinuous boundaries in presenting soft tissues, tumor margin can be extremely difficult to identify in magnetic resonance imaging (MRI), increasing the challenge of NPC segmentation task. Objectives The purpose of this work is to develop a semiautomatic algorithm for NPC image segmentation with minimal human intervention, while it is also capable of delineating tumor margins with high accuracy and reproducibility. Methods In this paper, we propose a novel feature selection algorithm for the identification of the margin of NPC image, named as modified random forest recursive feature selection (MRF-RFS). Specifically, to obtain a more discriminative feature subset for segmentation, a modified recursive feature selection method is applied to the original handcrafted feature set. Moreover, we combine the proposed feature selection method with the classical random forest (RF) in the training stage to take full advantage of its intrinsic property (i.e., feature importance measure). Results To evaluate the segmentation performance, we verify our method on the T1-weighted MRI images of 18 NPC patients. The experimental results demonstrate that the proposed MRF-RFS method outperforms the baseline methods and deep learning methods on the task of segmenting NPC images. Conclusion The proposed method could be effective in NPC diagnosis and useful for guiding radiation therapy.

Sign in / Sign up

Export Citation Format

Share Document