Proper tuning of hyper-parameters is essential to the successful application of SVM-classifiers. Several methods have been used for this problem: grid search, random search, estimation of distribution Algorithms (EDAs), bio-inspired metaheuristics, among others. The objective of this paper is to determine the optimal method among those that recently reported good results: Bat algorithm, Firefly algorithm, Fruit-fly optimization algorithm, particle Swarm optimization, Univariate Marginal Distribution Algorithm (UMDA), and Boltzmann-UMDA. The criteria for optimality include measures of effectiveness, generalization, efficiency, and complexity. Experimental results on 15 medical diagnosis problems reveal that EDAs are the optimal strategy under such criteria. Finally, a novel performance index to guide the optimization process, that improves the generalization of the solutions while maintaining their effectiveness, is presented.

Optimal Hyper-Parameter Tuning of SVM Classifiers With Application to Medical Diagnosis

Since indoor location information can be utilized in many aspects, indoor localization using existing wireless network has received much attentions. In this paper, a novel indoor localization algorithm using received signal strength indicator (RSSI) measurements and statistical feature of the channel state information (CSI) is proposed. In the off-line phase, with the obtained CSI measurement, the principal component analysis (PCA) pre-processing is utilized for dimension reduction at first. Then, the statistical features of the CSI measurements are extracted. Next, the RSSI measurements, the extracted feature of CSI measurements and the reference positions form the training data set. At last, the support vector machine (SVM) technique is proposed for regression learning and obtain the position based regression function. In the on-line phase, after PCA pre-processing and feature extraction of CSI measurements, the final position can estimated straightly with the regression function. The experiment results are shown that the proposed algorithm can offer more accurate localization result than other existing algorithms.

A New Indoor Localization Algorithm Using Received Signal Strength Indicator Measurements and Statistical Feature of the Channel State Information

Purpose The aim of this study was to verify improved, ensemble-based strategy for inferencing with use of our solution for quantitative assessment of tendons and ligaments healing process and to show possible applications of the method. Methods We chose the problem of the Achilles tendon rupture as an example representing a group of common sport traumas. We derived our dataset from 90 individuals and divided it into two subsets: healthy individuals and patients with complete Achilles tendon ruptures. We computed approx. 160 000 2D axial cross-sections from 3D MRI studies and preprocessed them to create a suitable input for artificial intelligence methods. Finally, we compared different training methods for chosen approaches for quantitative assessment of tendon tissue healing with the use of statistical analysis. Results We showed improvement in inferencing with use of the ensemble technique that results from achieving comparable accuracy of 99% for our previously published method trained on 500 000 samples and for the new ensemble technique trained on 160 000 samples. We also showed real-life applications of our approach that address several clinical problems: (1) automatic classification of healthy and injured tendons, (2) assessment of the healing process, (3) a pathologic tissue localization. Conclusions The presented method enables acquiring comparable accuracy with less training samples. The applications of the method presented in the paper as case studies can facilitate evaluation of the healing process and comparing with previous examination of the same patient as well as with other patients. This approach might be probably transferred to other musculoskeletal tissues and joints.

/pdf/monitoring-of-the-achilles-tendon-healing-process-can-532q6kgi9y.pdf

Monitoring of the Achilles tendon healing process: can artificial intelligence be helpful?

Abstract Recently, classification systems have received significant attention among researchers due to the important characteristics and behaviors of analysis required in real-time databases. Among the various classification-based methods suitable for real-time databases, fuzzy rule-based classification is effectively used by different researchers in various fields. An important issue in the design of fuzzy rule-based classification is the automatic generation of fuzzy if-then rules and the membership functions. The literature presents different techniques for automatic fuzzy design. Among the different techniques available in the literature, choosing the type, the number of membership functions, and defining parameters of membership function are still challenging tasks. In order to handle these challenges in the fuzzy rule-based classification system, this paper proposes a brain genetic fuzzy system (BGFS) for data classification by newly devising the exponential genetic brain storm optimization. Here, membership functions are optimally devised using exponential genetic brain storm optimization algorithm and rules are derived using the exponential brain storm optimization algorithm. The designed membership function and fuzzy rules are then effectively utilized for data classification. The proposed BGFS is analyzed with four datasets, using sensitivity, specificity, and accuracy. The outcome ensures that the proposed BGFS obtained the maximum accuracy of 88.8%, which is high as compared with the existing adaptive genetic fuzzy system.

BGFS: Design and Development of Brain Genetic Fuzzy System for Data Classification

Despite the development of two and three dimensional (2D&3D) technology, it has attracted the attention of researchers in recent years. This research is done to reveal the detailed effects of 2D in comparison with 3D technology on the human brain waves. The impact of 2D&3D video watching using electroencephalography (EEG) brain signals is studied. A group of eight healthy volunteers with the average age of 31+-3.06 years old participated in this three-stage test. EEG signal recording consisted of three stages: After a bit of relaxation (a), a 2D video was displayed (b), the recording of the signal continued for a short period of time as rest (c), and finally the trial ended. Exactly the same steps were repeated for the 3D video. Power spectrum density (PSD) based on short time Fourier transform (STFT) was used to analyze the brain signals of 2D&3D video viewers. After testing all the EEG frequency bands, delta and theta were extracted as the features. Partial least squares regression (PLSR) and Support vector machine (SVM) classification algorithms were considered in order to classify EEG signals obtained as the result of 2D&3D video watching. Successful classification results were obtained by selecting the correct combinations of effective channels representing the brain regions.

A Comprehensive Analysis of 2D&3D Video Watching of EEG Signals by Increasing PLSR and SVM Classification Results

Identifying performance of classifier is a challenging task. SVM plays an important role in classification. Here different kernel parameters are used as a tuning parameter to improve the classification accuracy. There are mainly four different types of kernels (Linear, Polynomial, RBF, and Sigmoid) that are popular in SVM classifier. The paper presents SVM classification results with above mentioned kernels on two different datasets (Diabetic Retinopathy dataset and Lung Cancer dataset). To evaluate the performance of the classifier we have used True positive rate, False Positive rate, Precision, Recall, F-measure and accuracy as performance measures of SVM. Finally we evaluated that SVM with linear kernel performs best among all.

Role of Kernel Parameters in Performance Evaluation of SVM

Data mining can help researchers to gain novel and deep insights for understanding of large datasets. Nowadays, people are using data mining algorithms in different contexts like banking, hospitals, marketing etc. Classification algorithm plays a vital role. In the study, we found that SVM is the best classifier amongst all the classifiers. Here we used learning algorithms with the historical dataset to train the classifier and the test samples are used to validate the correctness of the classifier. We might have structured semistructured and unstructured datasets which are used for classification. We have performed the study of reputed literatures that belong to classification area to identify some new enhancements in the classifiers. A few most important classifiers are SVM, decision tree, neural network, Naive Bayes. We found most of the literatures were concentrated on SVM classifier so we targeted SVM classifier for the performance enhancement. SVM are important tool in datamining to classify data. The aim of this review is to identify the effectiveness of kernel parameters for classification of data using Support Vector.

/pdf/study-of-data-mining-algorithms-in-the-context-of-4ugdr11ehv.pdf

Aditi Goel

Papers

Role of Kernel Parameters in Performance Evaluation of SVM

Study of Data Mining Algorithms in the Context of Performance Enhancement of Classification