A Non Invasive, Wearable Sensor Platform for Multi-parametric Remote Monitoring in CHF Patients

Mining Personal Data Using Smartphones and Wearable Devices: A Survey

“…For example, in case of tree-based algorithms, C4.5 [ 71 ] performed best with more than 90% accuracy compared with NB, but in another study [ 59 ], NB performed best with 86% ± 3.9% as compared with 85.9% ± 2.5% of C4.5; furthermore, NB consumes 71 KB less memory than C4.5. Similarly, in some other studies [ 65 , 73 ], SVM outperformed NB in all cases.…”

“…Similarly, neural network-based models, such as Artificial Neural Networks (ANNs) and Multi-Layer Perception (MLP)-based NNs, are used for fall detection, energy-efficient activity recognition, WSN-based activity recognition, and simple and complex activity recognition [ 21 , 65 , 71 , 72 ]. Statistical classifiers, including Support Vector Machines (SVM), Quadratic Discriminant Analysis (QDA), Linear Discriminant Analysis (LDA) and k Nearest Neighbors (kNN), are implemented for activity recognition, injury rehabilitation, physiological data analysis, optimized energy consumption, stress profiling, physical activity recognition, discrimination between stress and cognitive load, real-time activity recognition, and application usage prediction in mobile phones [ 65 , 73 – 79 ].…”

“…Moreover, several regression models have been used in numerous studies, including Localized Logistic Regression (LLR), Nonlinear Logistic Regression (NLR), and Maximum Entropy (Max Ent) for energy efficiency and intelligent context-based personalized services; Bayesian Network and instance-based algorithms like K-star (K*) for simple and complex activity recognition; and rule-based models like Decision Table (DT) for activity recognition [ 26 , 65 , 71 , 77 , 80 ]. Finally, probabilistic models, such as Naive Bayes (NB), are widely adopted for distance estimation among Wi-Fi users, distributed classification, “callee” recommendation using personal and social contexts, physiological data analysis, personalization, privacy and adaption, and activity recognition [ 26 , 59 , 65 , 71 , 73 , 81 – 84 ]. Such a wide-scale adoption of classification algorithms has encouraged researchers to explore new opportunities for PerDM in PSDs.…”

Frequent pattern mining in mobile devices: A feasibility study

“…The set C k contains the items of the form <Tid, [4]> where each I k represents a large k-itemset with transaction identifier (Tid).…”

“…Numerous data mining algorithms are successfully exploited in smartphones to comply with computation and resource constraints and give the optimal performance. For example classification algorithms are used for activity recognition [2], energy efficiency [3], physiological data analysis [4], personalization, privacy and adaption [5], intelligent distributed classification [6], fall detection [7], injury rehabilitation [8], discrimination between stress and cognitive load [9], and application usage prediction in mobile phones [10] amongst other. Such a wide-scale adoption of data mining algorithms has stimulated the researchers to explore new opportunities for mobile data mining inside smartphones.…”