Search

Classifying e-mails into distinct labels can have a great impact on customer support. By using machine learning to label e-mails, the system can set up queues containing e-mails of a specific category. This enables support personnel to handle request quicker and more easily by selecting a queue that match their expertise. This study aims to improve a manually defined rule-based algorithm, currently implemented at a large telecom company, by using machine learning. The proposed model should have higher$$F_1$$F1-score and classification rate. Integrating or migrating from a manually defined rule-based model to a machine learning model should also reduce the administrative and maintenance work. It should also make the model more flexible. By using the frameworks, TensorFlow, Scikit-learn and Gensim, the authors conduct a number of experiments to test the performance of several common machine learning algorithms, text-representations, word embeddings to investigate how they work together. A long short-term memory network showed best classification performance with an$$F_1$$F1-score of 0.91. The authors conclude that long short-term memory networks outperform other non-sequential models such as support vector machines and AdaBoost when predicting labels for e-mails. Further, the study also presents a Web-based interface that were implemented around the LSTM network, which can classify e-mails into 33 different labels.

The paper presents the development of a computeraided diagnostic (CAD) system for the early detection of endometrial cancer. The proposed CAD system supports reproducibility through texture feature standardization, standardized multifeature selection, and provides physicians with comparative distributions of the extracted texture features. The CAD system was validated using 516 regions of interest (ROIs) extracted from 52 subjects. The ROIs were equally distributed among normal and abnormal cases. To support reproducibility, the RGB images were first gamma corrected and then converted into HSV and YCrCb. From each channel of the gamma-corrected YCrCb, HSV, and RGB color systems, we extracted the following texture features: 1) statistical features (SFs), 2) spatial gray-level dependence matrices (SGLDM), and 3) gray-level difference statistics (GLDS). The texture features were then used as inputs with support vector machines (SVMs) and the probabilistic neural network (PNN) classifiers. After accounting for multiple comparisons, texture features extracted from abnormal ROIs were found to be significantly different than texture features extracted from normal ROIs. Compared to texture features extracted from normal ROIs, abnormal ROIs were characterized by lower image intensity, while variance, entropy, and contrast gave higher values. In terms of ROI classification, the best results were achieved by using SF and GLDS features with an SVM classifier. For this combination, the proposed CAD system achieved an 81% correct classification rate. 2168-2194 © 2014 IEEE. 19 3 1129 1136 Cited By :2

Working conditions and their interaction on Quality of Working Life (QWL) perception in health sector employees from Barranquilla-Colombia were studied and modeled in a sample of 333 people aged between 16 and 73 years old. Reliability analysis, exploratory factor and principal components analysis with VARIMAX rotation, and binary logistical regressions was performed. Results show acceptable internal reliability of the scale (α = 0.996) and satisfactory adequacy of factorial matrix data (KMO = 0.934). Sixty-three items grouped into ten elements presented the factorial structure: four belong to working conditions and six to QWL. The results showed that occupational safety and psychosocial aspects positively affects the Quality of Working Life, with an overall classification rate of 92.2%. QWL in the health sector is a function of environmental variables, which can have an impact on the quality of the service and, therefore, on social welfare. © 2017 International Business Information Management Association IBIMA. All Rights Reserved.

Network anomaly detection is currently a challenge due to the number of different attacks and the number of potential attackers. Intrusion detection systems aim to detect misuses or network anomalies in order to block ports or connections, whereas firewalls act according to a predefined set of rules. However, detecting the specific anomaly provides valuable information about the attacker that may be used to further protect the system, or to react accordingly. This way, detecting network intrusions is a current challenge due to growth of the Internet and the number of potential intruders. In this paper we present an intrusion detection technique using an ensemble of support vector classifiers and dimensionality reduction techniques to generate a set of discriminant features. The results obtained using the NSL-KDD dataset outperforms previously obtained classification rates La detección de anomalías en la red es actualmente un desafío debido a la cantidad de ataques diferentes y el número de posibles atacantes. Los sistemas de detección de intrusos apuntan a detectar los abusos. o anomalías de la red para bloquear puertos o conexiones, mientras que los firewalls actúan de acuerdo a un conjunto predefinido de reglas. Sin embargo, la detección de la anomalía específica proporciona información valiosa sobre el atacante que puede usarse para proteger aún más el sistema, o para reaccionar en consecuencia. De esta manera, la detección de intrusiones en la red es un desafío actual debido a el crecimiento de Internet y el número de posibles intrusos. En este trabajo presentamos una técnica de detección de intrusos utilizando un conjunto de clasificadores de vectores de soporte y técnicas de reducción de dimensionalidad para generar un conjunto de características discriminantes. Los resultados obtenido utilizando el conjunto de datos NSL-KDD supera las tasas de clasificación obtenidas previamente

This study introduces the use of multiscale amplitude modulation-frequency modulation (AM-FM) texture analysis of multiple sclerosis (MS) using magnetic resonance (MR) images from brain. Clinically, there is interest in identifying potential associations between lesion texture and disease progression, and in relating texture features with relevant clinical indexes, such as the expanded disability status scale (EDSS). This longitudinal study explores the application of 2-D AM-FM analysis of brain white matter MS lesions to quantify and monitor disease load. To this end, MS lesions and normal-appearing white matter (NAWM) from MS patients, as well as normal white matter (NWM) from healthy volunteers, were segmented on transverse T2-weighted images obtained from serial brain MR imaging (MRI) scans (0 and 6-12 months). The instantaneous amplitude (IA), the magnitude of the instantaneous frequency (IF), and the IF angle were extracted from each segmented region at different scales. The findings suggest that AM-FM characteristics succeed in differentiating 1) between NWM and lesions 2) between NAWM and lesions and 3) between NWM and NAWM. A support vector machine (SVM) classifier succeeded in differentiating between patients that, two years after the initial MRI scan, acquired an EDSS ≤ 2 from those with EDSS > 2 (correct classification rate = 86%). The best classification results were obtained from including the combination of the low-scale IA and IF magnitude with the medium-scale IA. The AM-FM features provide complementary information to classical texture analysis features like the gray-scale median, contrast, and coarseness. The findings of this study provide evidence that AM-FM features may have a potential role as surrogate markers of lesion load in MS. © 2006 IEEE. 15 1 119 129 Cited By :31

Conference name: First International Workshop, HBU 2010 Date of Conference: August 22, 2010 This paper provides a comparative study on the different techniques of classifying human activities that are performed using body-worn miniature inertial and magnetic sensors. The classification techniques implemented and compared in this study are: Bayesian decision making (BDM), the least-squares method (LSM), the k-nearest neighbor algorithm (k-NN), dynamic time warping (DTW), support vector machines (SVM), and artificial neural networks (ANN). Daily and sports activities are classified using five sensor units worn by eight subjects on the chest, the arms, and the legs. Each sensor unit comprises a triaxial gyroscope, a triaxial accelerometer, and a triaxial magnetometer. Principal component analysis (PCA) and sequential forward feature selection (SFFS) methods are employed for feature reduction. For a small number of features, SFFS demonstrates better performance and should be preferable especially in real-time applications. The classifiers are validated using different cross-validation techniques. Among the different classifiers we have considered, BDM results in the highest correct classification rate with relatively small computational cost. © 2010 Springer-Verlag Berlin Heidelberg.

Date of Conference: 27-29 Oct. 2008 Conference name: 23rd International Symposium on Computer and Information Sciences, ISCIS 2008 This study investigates the use of low-cost infrared (IR) sensors for the determination of geometry and surface properties of commonly encountered features or targets in indoor environments, such as planes, corners, edges, and cylinders using artificial neural networks (ANNs). The intensity measurements obtained from such sensors are highly dependent on the location, geometry, and surface properties of the reflecting target in a way which cannot be represented by a simple analytical relationship, therefore complicating the localization and classification process. We propose the use of angular intensity scans and feature vectors obtained by modeling of angular intensity scans and present two different neural network based approaches in order to classify the geometry and/or the surface type of the targets. In the first case, where planes, 90° corners, and 90° edges covered with aluminum, white cloth, and Styrofoam packaging material are differentiated, an average correct classification rate of 78% of both geometry and surface over all target types is achieved. In the second case, where planes, 90° edges, and cylinders covered with different surface materials are differentiated, an average correct classification rate of 99.5% is achieved. The method demonstrated shows that ANNs can be used to extract substantially more information than IR sensors are commonly employed for. © 2008 IEEE.

This paper introduces the surface electromyogram (EMG) classification system based on statistical and entropy metrics. The system is intended for diagnostic use and enables classification of examined subject as normal, myopathic or neuropathic, regarding to the acquired EMG signals. 39 subjects in total participated in the experiment, 19 normal, 11 myopathic and 9 neuropathic. Surface EMG was recorded using 4-channel surface electrodes on the biceps brachii muscle at isometric voluntary contractions. The recording time was only 5 seconds long to avoid muscle fatigue, and contractions at five force levels were performed, i.e. 10, 30, 50, 70 and 100 % of maximal voluntary contraction. The feature extraction routine deployed the wavelet transform and calculation of the Shannon entropy across all the scales in order to obtain a feature set for each subject. Subjects were classified regarding the extracted features using three machine learning techniques, i.e. decision trees, support vector machines and ensembles of support vector machines. Four 2-class classifications and a 3-class classification were performed. The scored classification rates were the following: 64±11% for normal/ abnormal, 74±7% for normal/myopathic, 79±8% for normal /neuropathic, 49±20% for myopathic/neuropathic, and 63±8% for normal/myopathic/neuropathic. 4 2 28 35

Date of Conference: 4-8 Sept. 2005 Conference name: 13th European Signal Processing Conference, 2005 We propose the use of angular intensity signals obtained with low-cost infrared (IR) sensors and present an algorithm to simultaneously extract the geometry and surface properties of commonly encountered features or targets in indoor environments. The method is verified experimentally with planes, 90° corners, and 90° edges covered with aluminum, white cloth, and Styrofoam packaging material. An average correct classification rate of 80% of both geometry and surface over all target types is achieved and targets are localized within absolute range and azimuth errors of 1.5 cm and 1.1°, respectively. Taken separately, the geometry and surface type of targets can be correctly classified with rates of 99% and 81%, respectively, which shows that the geometrical properties of the targets are more distinctive than their surface properties, and surface determination is the limiting factor. The method demonstrated shows that simple IR sensors, when coupled with appropriate signal processing, can be used to extract substantially more information than such devices are commonly employed for.

A new supervised learning algorithm using naïve Bayesian classifier is presented in this paper, which calculates the prior and conditional probabilities from a given training data and classifies the training examples using these probabilities. If any training example is misclassified then the algorithm calculates the information gain of attributes of the training data and chooses one attribute from training data with maximum information gain value. After the algorithm splits the training data into sub-datasets depending on the attribute values of the selected attribute, and again calculates the prior and conditional probabilities for each sub-dataset and classifies the examples of the each sub-dataset using their respective probabilities. The process will continue until all the training examples are correctly classified. Finally, the algorithm preserves the probabilities of each dataset for the future classification of unknown examples, whose attributes value are known but class value is unknown. The proposed algorithm addresses the problem of classifying the large dataset and it has been successfully tested on a number of benchmark problems, which achieved high classification rates using limited computational resources.