Feature Engineering and Selection: A Practical Approach for Predictive Models

Butcher, Brandon; Smith, Brian J.

doi:10.1080/00031305.2020.1790217

Cited by 38 publications

(21 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea is to include only those features that significantly contribute to the ability of the algorithm to distinguish between different classes, effectively limiting the number of computations necessary to obtain an accurate prediction. Thus, the exact number of selected features may vary from one model to the next, depending on the type of algorithm used, structure and volume of raw data, and the main tasks that the machine learning classifier will be expected to complete [92].…”

Section: ) Feature Selectionmentioning

confidence: 99%

“…In terms of the principles used to rank the features, we can talk about Filter methods (such as variance threshold, correlation coefficient, or Chi-square test) which capture some of the native properties of each feature, and Wrapper methods (i.e. forward feature selection or backward feature elimination), which measure how a proposed set of features works with a particular algorithm [92]. There are also Embedded (LASSO regularization or random forest importance) and Hybrid approaches, which combine some of the main strengths of both Filter and Wrapper methods.…”

Section: ) Feature Selectionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning for Sign Language Recognition: Current Techniques, Benchmarks, and Open Issues

Al‐Qurishi¹,

Khalid²,

Souissi³

2021

IEEE Access

View full text Add to dashboard Cite

People with hearing impairments are found worldwide; therefore, the development of effective local level sign language recognition (SLR) tools is essential. We conducted a comprehensive review of automated sign language recognition based on machine/deep learning methods and techniques published between 2014 and 2021 and concluded that the current methods require conceptual classification to interpret all available data correctly. Thus, we turned our attention to elements that are common to almost all sign language recognition methodologies. This paper discusses their relative strengths and weaknesses, and we propose a general framework for researchers. This study also indicates that input modalities bear great significance in this field; it appears that recognition based on a combination of data sources, including vision-based and sensor-based channels, is superior to a unimodal analysis. In addition, recent advances have allowed researchers to move from simple recognition of sign language characters and words towards the capacity to translate continuous sign language communication with minimal delay. Many of the presented models are relatively effective for a range of tasks, but none currently possess the necessary generalization potential for commercial deployment. However, the pace of research is encouraging, and further progress is expected if specific difficulties are resolved.

show abstract

Section: ) Feature Selectionmentioning

confidence: 99%

Section: ) Feature Selectionmentioning

confidence: 99%

Deep Learning for Sign Language Recognition: Current Techniques, Benchmarks, and Open Issues

Al‐Qurishi¹,

Khalid²,

Souissi³

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…They are a set of techniques that assign a score to input features based on how useful they are at predicting a target variable. Feature importance scores play an important role in a predictive modeling project, providing an insight into data, an insight into the model, and the basis for dimensionality reduction and feature selection which can improve the efficiency and effectiveness of a predictive model on the problem [36].…”

Section: E Feature Importancementioning

confidence: 99%

A Framework for IoT Based Appliance Recognition in Smart Homes

et al. 2021

View full text Add to dashboard Cite

Internet of Things (IoT) technologies will play an important role in enabling the smart grid achieving its goals in monitoring, protecting, and controlling by incorporating sensors, actuators, and metering devices while supporting various network functions and system automation. In this regard, home energy management systems (HEMS) enable households to better manage energy consumption by providing information about energy usage and allowing more precise control of major appliances. This work proposes a novel framework for IoT based appliance recognition in smart homes. It consists of two parts: training framework and inference framework. The proposed framework allows incorporating different loads in the monitoring system and enables selecting and testing specific parameters related to dataset configuration, feature extraction, and classifier model setting. The work contributes by developing an easy-to-use tool that allows customization of the training/prediction parameters according to the user criterion. Once the data and all its parameters are loaded, a novel feature extraction algorithm is used to obtain a total of ten statistical features. For the classification task, three machine learning models are included: a feed-forward neural network (FFNN), a long short-term memory (LSTM) and a support vector machine (SVM). In addition, the user can apply a set of techniques to overcome the class imbalance, and also measure the influence of the selected features in the classifiers' prediction by performing a feature importance analysis.INDEX TERMS Appliance recognition, frameworks, intrusive load monitoring, internet of things, smart grids, smart homes.

show abstract

“…In the first stage, four statistical measures based on correlation were used. According to Butcher and Smith (2020), inferential statistical approaches provide a better solution to appraising the contribution of a predictor to the underlying model or the dataset. These measures are Analysis of Variance (ANOVA), Pearson's Correlation Coefficient, Mutual Information, and Chi-Squared.…”

Section: Proposed Feature Engineering Approachmentioning

confidence: 99%

“…Feature selection and engineering are essential factors in machine learning as they can increase the predictive power of ML algorithms. Feature engineering involves understanding the domain knowledge of the dataset to create new features or combine existing features that make ML algorithms perform better (Butcher & Smith, 2020). Hence, the contribution of this study is the feature selection and engineering processes on the BoT-IoT dataset in an attempt to improve the performance of ML algorithms.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Methods for Detecting Internet-of-Things (IoT) Malware

Yaokumah

Appati

Kumah³

2021

International Journal of Cognitive Informatics and Natural Intelligence

View full text Add to dashboard Cite

This study aims to analyze the performance of machine learning models for detecting Internet of Things malware utilizing a recent IoT dataset. Experiments on the IoT dataset were conducted with nine well-known machine learning techniques, consisting of Logistic Regression (LR), Naive Bayes (NB), Decision Tree (DT), k-Nearest Neighbors (KNN), Support Vector Machines (SVM), Neural Networks (NN), Random Forest (RF), Bagging (BG), and Stacking (ST). The results show that the proposed model attains 100% accuracy in detecting IoT malware for DT, SVM, RF, BG; about 99.9% percent for LR, NB, KNN, NN; and only 28.16% for ST classifier. This study also shows higher performance than other proposed machine learning models evaluated on the same dataset. Therefore, the results of this study can help both the researchers and application developers in designing and building intelligent malware detection systems for IoT devices.

show abstract

Feature Engineering and Selection: A Practical Approach for Predictive Models

Cited by 38 publications

References 6 publications

Deep Learning for Sign Language Recognition: Current Techniques, Benchmarks, and Open Issues

Deep Learning for Sign Language Recognition: Current Techniques, Benchmarks, and Open Issues

A Framework for IoT Based Appliance Recognition in Smart Homes

Machine Learning Methods for Detecting Internet-of-Things (IoT) Malware

Contact Info

Product

Resources

About