An efficient approach to kNN algorithm for IoT devices

Gawri, Bhavesh; Kasturi, Anirudh; Neti, Lalita Bhanu Murthy; Hota, Chittaranjan

doi:10.1109/comsnets53615.2022.9668540

Cited by 7 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As well the testing (prediction) time complexity of the KNN method is O(k 1 dN), the space complexity is O(dN), where the N is the number of samples in the dataset (dataset size), d is the dimension of the dataset, and k 1 is the number of Nearest Neighbors [29].…”

Section: Analysis Of Computational Complexitymentioning

confidence: 99%

Arrhythmia Detection Based on New Multi-Model Technique for ECG Inter-Patient Classification

Oleiwi,

AlShemmary,

Al-Augby

2023

Int. J. Onl. Eng.

View full text Add to dashboard Cite

This paper presents a novel model for arrhythmia detection based on a cascading technique that utilizes a combination of the One-Sided Selection (OSS) method, Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) algorithms, this model denoted by (OWSK) model to classify four types of electrocardiogram (ECG) heartbeats following inter-patient scheme. The OWSK model consists of three stages. The first stage involves resampling using the One-Sided Selection (OSS) method to solve the imbalance problem and reduce data by removing noisy, borderline, and redundant samples. The second stage involves using Wavelet Transformation (WT) and Power Spectral Density (PSD) to extract the most relevant frequency domain features. The third stage involves a cascading process by constructing the classifier from SVM trained on the whole dataset to classify normal and abnormal beats. Then, KNN (K-Nearest Neighbors) is trained on only the three irregular minority classes to classify the three types of arrhythmias for the detection of ventricular ectopic beats, supraventricular ectopic beats, and fusion beats (V, S, and F). The performance of the proposed model is evaluated in terms of different metrics, including accuracy, recall, precision, and F1 score. The results show the superiority of the proposed model in medical diagnosis compared to the latest works, where it achieves 90%, 90%, 93%, and 91% for accuracy, recall, precision, and F1 score under the inter-patient paradigm and 98%, 98%, 98%, and 98% under the intra-patient paradigm.

show abstract

Section: Analysis Of Computational Complexitymentioning

confidence: 99%

Arrhythmia Detection Based on New Multi-Model Technique for ECG Inter-Patient Classification

Oleiwi,

AlShemmary,

Al-Augby

2023

Int. J. Onl. Eng.

View full text Add to dashboard Cite

show abstract

“…Given a pattern x ω for classification and a training set E, the k-NN algorithm works by calculating the distance between x ω and each of the elements in E, and assigning the majority class among the nearest k patterns. Due to its simplicity and good performance, the k-NN algorithm is widely used with all sorts of datasets [33][34][35], modified versions of it have been presented [36,37], and even an efficient approach for IoT devices has been proposed [38].…”

Section: Classical K-nn Algorithmmentioning

confidence: 99%

Quantum version of the k-NN classifier based on a quantum sorting algorithm

Quezada,

Sun,

Dong

2022

Preprint

View full text Add to dashboard Cite

In this work we introduce a quantum sorting algorithm with adaptable requirements of memory and circuit depth, and then use it to develop a new quantum version of the classical machine learning algorithm known as k-nearest neighbors (k-NN). Both the efficiency and performance of this new quantum version of the k-NN algorithm are compared to those of the classical k-NN and another quantum version proposed by Schuld et al. [20]. Results show that the efficiency of both quantum algorithms is similar to each other and superior to that of the classical algorithm. On the other hand, the performance of our proposed quantum k-NN algorithm is superior to the one proposed by Schuld et al. and similar to that of the classical k-NN.

show abstract

“…It is employed in the categorization and regression of data. The important feature of any k-NN technique, whether it's for classification or regression, is to locate the k-NN, which let us estimate the value or class for a given point [48]. The vanishing gradient problem is a significant drawback of RNNs [49].…”

Section: A Attacker Modelmentioning

confidence: 99%

Ensuring IoT Privacy using Padding Strategies against Machine Learning Approaches

Ergun,

Can

2022

IJMSIT

View full text Add to dashboard Cite

The widespread usage of Internet of Things (IoT) devices is increasing by the recent advances in embedded systems, cloud computing, artificial intelligence, and wireless communications. Besides, a huge amount of data is transmitted between IoT devices over insecure networks. The transferred data can be sensitive and confidential. On the other hand, these transmitted data may not appear to be sensitive or confidential data. However, machine learning techniques are used on these non-confidential data (such as packet length) to obtain data such as the type of the IoT device. An observer can monitor traffic to infer sensitive data by using machine learning techniques to analyze the generated encrypted traffic. For this purpose, padding can be added to the packets to ensure traffic privacy. This paper presents privacy problems that are caused by the traffic generated during the communication of IoT devices. Also, security and privacy measures that should be taken against the related privacy problems are explained. For this purpose, the current studies are examined by considering the attacker and the defender models.

show abstract

An efficient approach to kNN algorithm for IoT devices

Cited by 7 publications

References 14 publications

Arrhythmia Detection Based on New Multi-Model Technique for ECG Inter-Patient Classification

Arrhythmia Detection Based on New Multi-Model Technique for ECG Inter-Patient Classification

Quantum version of the k-NN classifier based on a quantum sorting algorithm

Ensuring IoT Privacy using Padding Strategies against Machine Learning Approaches

Contact Info

Product

Resources

About