8. Subset Selection in Regression (Monographs on Statistics and Applied Probability, no. 40). By A. J. Miller. ISBN 0 412 35380 6. Chapman and Hall, London, 1990. 240 pp. £25.00.

Subset Selection in Regression

Parkinson's Disease (PD) is a progressive neurodegenerative disease with multiple motor and non-motor characteristics. PD patients commonly face vocal impairments during the early stages of the disease. So, diagnosis systems based on vocal disorders are at the forefront on recent PD detection studies. Our study proposes two frameworks based on Convolutional Neural Networks to classify Parkinson's Disease (PD) using sets of vocal (speech) features. Although, both frameworks are employed for the combination of various feature sets, they have difference in terms of combining feature sets. While the first framework combines different feature sets before given to 9-layered CNN as inputs, whereas the second framework passes feature sets to the parallel input layers which are directly connected to convolution layers. Thus, deep features from each parallel branch are extracted simultaneously before combining in the merge layer. Proposed models are trained with dataset taken from UCI Machine Learning repository and their performances are validated with Leave-One-Person-Out Cross Validation (LOPO CV). Due to imbalanced class distribution in our data, F-Measure and Matthews Correlation Coefficient metrics are used for the assessment along with accuracy. Experimental results show that the second framework seems to be very promising, since it is able to learn deep features from each feature set via parallel convolution layers. Extracted deep features are not only successful at distinguishing PD patients from healthy individuals but also effective in boosting up the discriminative power of the classifiers.

/pdf/deep-learning-based-parkinson-s-disease-classification-using-q8tvwb0xn2.pdf

Deep Learning-Based Parkinson’s Disease Classification Using Vocal Feature Sets

Imbalance-XGBoost: leveraging weighted and focal losses for binary label-imbalanced classification with XGBoost

Horizontal Federated learning (FL) handles multi-client data that share the same set of features, and vertical FL trains a better predictor that combine all the features from different clients. This paper targets solving vertical FL in an asynchronous fashion, and develops a simple FL method. The new method allows each client to run stochastic gradient algorithms without coordination with other clients, so it is suitable for intermittent connectivity of clients. This method further uses a new technique of perturbed local embedding to ensure data privacy and improve communication efficiency. Theoretically, we present the convergence rate and privacy level of our method for strongly convex, nonconvex and even nonsmooth objectives separately. Empirically, we apply our method to FL on various image and healthcare datasets. The results compare favorably to centralized and synchronous FL methods.

VAFL: a Method of Vertical Asynchronous Federated Learning

Most real-world data are scattered across different companies or government organizations, and cannot be easily integrated under data privacy and related regulations such as the European Union's General Data Protection Regulation (GDPR) and China' Cyber Security Law. Such data islands situation and data privacy & security are two major challenges for applications of artificial intelligence. In this paper, we tackle these challenges and propose a privacy-preserving machine learning model, called Federated Forest, which is a lossless learning model of the traditional random forest method, i.e., achieving the same level of accuracy as the non-privacy-preserving approach. Based on it, we developed a secure cross-regional machine learning system that allows a learning process to be jointly trained over different regions' clients with the same user samples but different attribute sets, processing the data stored in each of them without exchanging their raw data. A novel prediction algorithm was also proposed which could largely reduce the communication overhead. Experiments on both real-world and UCI data sets demonstrate the performance of the Federated Forest is as accurate as the non-federated version. The efficiency and robustness of our proposed system had been verified. Overall, our model is practical, scalable and extensible for real-life tasks.

Federated Forest

A comparative analysis of speech signal processing algorithms for Parkinson’s disease classification and the use of the tunable Q-factor wavelet transform

In recent years, there is an increasing interest in building e-health systems. The systems built to deliver the health services with the use of internet and communication technologies aim to reduce the costs arising from outpatient visits of patients. Some of the related recent studies propose machine learning–based telediagnosis and telemonitoring systems for Parkinson’s disease (PD). Motivated from the studies showing the potential of speech disorders in PD telemonitoring systems, in this study, we aim to estimate the severity of PD from voice recordings of the patients using motor Unified Parkinson’s Disease Rating Scale (UPDRS) as the evaluation metric. For this purpose, we apply various speech processing algorithms to the voice signals of the patients and then use these features as input to a two-stage estimation model. The first step is to apply a wrapper-based feature selection algorithm, called Boruta, and select the most informative speech features. The second step is to feed the selected set of features to a decision tree–based boosting algorithm, extreme gradient boosting, which has been recently applied successfully in many machine learning tasks due to its generalization ability and speed. The feature selection analysis showed that the vibration pattern of the vocal fold is an important indicator of PD severity. Besides, we also investigate the effectiveness of using age and years passed since diagnosis as covariates together with speech features. The lowest mean absolute error with 3.87 was obtained by combining these covariates and speech features with prediction level fusion.

Estimation of Parkinson’s disease severity using speech features and extreme gradient boosting

As the e-commerce sales grow in global retail sector year by year, detecting anomalies that occur in the most important key performance indicators (KPI) in real-time has become a critical requirement for e-commerce companies. Such anomalies that may arise from software updates, server failures, or incorrect price entries cause substantial revenue loss in the meantime until they are detected with their root-causes. In this paper, we present a comparative analysis of various anomaly detection methods in detecting e-commerce anomalies. For this purpose, we first present the univariate analysis of six commonly used anomaly detection methods on two important KPIs of an e-commerce website. The highest F1 Scores and recall values on the test sets of both KPIs are obtained using Long-Short Term Memory (LSTM) network, showing that LSTM fits better to the dynamics of e-commerce KPIs than time-series based prediction methods. Then, in addition to the univariate analysis of the methods, we feed the campaign information into LSTM network considering that campaigns have significant effects on the values of KPIs in e-commerce domain and this information can be helpful to prevent false positives that may occur in the campaign periods. The results also show that constructing a multivariate LSTM by feeding the campaign information as an additional input improves the adaptability of the model to sudden changes occurring in campaign periods.

Hunkar Can Tunc

Papers

A comparative analysis of speech signal processing algorithms for Parkinson’s disease classification and the use of the tunable Q-factor wavelet transform

Estimation of Parkinson’s disease severity using speech features and extreme gradient boosting

Detection of e-Commerce Anomalies using LSTM-recurrent Neural Networks

Explaining safety failures in NetKAT

A Language-Based Causal Model for Safety