Search results for: probabilistic classification vector machines

Authors: Dechen Tshering Vogel, Johannes T. Heverhagen, Bernard Kiss, Spyridon Arampatzis

Abstract:

In addition to computed tomography (CT), contrast enhanced ultrasound (CEUS), and magnetic resonance imaging (MRI) are being increasingly used for imaging of renal lesions. The aim of this prospective study was to compare the classification of complex cystic renal lesions using the Bosniak classification with CEUS and MRI to CT. Forty-eight patients with 65 cystic renal lesions were included in this study. All participants signed written informed consent. The agreement between the Bosniak classifications of complex renal lesions ( ≥ BII-F) on CEUS and MRI were compared to that of CT and were tested using Cohen’s Kappa. Sensitivity, specificity, positive and negative predictive values (PPV/NPV) and the accuracy of CEUS and MRI compared to CT in the detection of complex renal lesions were calculated. Twenty-nine (45%) out of 65 cystic renal lesions were classified as complex using CT. The agreement between CEUS and CT in the classification of complex cysts was fair (agreement 50.8%, Kappa 0.31), and was excellent between MRI and CT (agreement 93.9%, Kappa 0.88). Compared to CT, MRI had a sensitivity of 96.6%, specificity of 91.7%, a PPV of 54.7%, and an NPV of 54.7% with an accuracy of 63.1%. The corresponding values for CEUS were sensitivity 100.0%, specificity 33.3%, PPV 90.3%, and NPV 97.1% with an accuracy 93.8%. The classification of complex renal cysts based on MRI and CT scans correlated well, and MRI can be used instead of CT for this purpose. CEUS can exclude complex lesions, but due to higher sensitivity, cystic lesions tend to be upgraded. However, it is useful for initial imaging, for follow up of lesions and in those patients with contraindications to CT and MRI.

Keywords: Bosniak classification, computed tomography, contrast enhanced ultrasound, cystic renal lesions, magnetic resonance imaging

Procedia PDF Downloads 132

Authors: Zhang Shuqi, Liu Dan

Abstract:

For the problems in intelligent network anomaly traffic detection models, such as low detection accuracy caused by the lack of training samples, poor effect with small sample attack detection, a classification model enhancement method, F-ACGAN(Flow Auxiliary Classifier Generative Adversarial Network) which introduces generative adversarial network and adversarial training, is proposed to solve these problems. Generating adversarial data with category labels could enhance the training effect and improve classification accuracy and model robustness. FACGAN consists of three steps: feature preprocess, which includes data type conversion, dimensionality reduction and normalization, etc.; A generative adversarial network model with feature learning ability is designed, and the sample generation effect of the model is improved through adversarial iterations between generator and discriminator. The adversarial disturbance factor of the gradient direction of the classification model is added to improve the diversity and antagonism of generated data and to promote the model to learn from adversarial classification features. The experiment of constructing a classification model with the UNSW-NB15 dataset shows that with the enhancement of FACGAN on the basic model, the classification accuracy has improved by 8.09%, and the score of F1 has improved by 6.94%.

Keywords: data imbalance, GAN, ACGAN, anomaly detection, adversarial training, data augmentation

Procedia PDF Downloads 89

Authors: Yunwei Tang, Linhai Jing, Hui Li, Qingjie Liu, Xiuxia Li, Qi Yan, Haifeng Ding

Abstract:

The giant panda (Ailuropoda melanoleuca) is an endangered species, mainly live in central China, where bamboos act as the main food source of wild giant pandas. Knowledge of spatial distribution of bamboos therefore becomes important for identifying the habitat of giant pandas. There have been ongoing studies for mapping bamboos and other tree species using remote sensing. WorldView-2 (WV-2) is the first high resolution commercial satellite with eight Multi-Spectral (MS) bands. Recent studies demonstrated that WV-2 imagery has a high potential in classification of tree species. The advanced classification techniques are important for utilising high spatial resolution imagery. It is generally agreed that object-based image analysis is a more desirable method than pixel-based analysis in processing high spatial resolution remotely sensed data. Classifiers that use spatial information combined with spectral information are known as contextual classifiers. It is suggested that contextual classifiers can achieve greater accuracy than non-contextual classifiers. Thus, spatial correlation can be incorporated into classifiers to improve classification results. The study area is located at Wuyipeng area in Wolong, Sichuan Province. The complex environment makes it difficult for information extraction since bamboos are sparsely distributed, mixed with brushes, and covered by other trees. Extensive fieldworks in Wuyingpeng were carried out twice. The first one was on 11th June, 2014, aiming at sampling feature locations for geometric correction and collecting training samples for classification. The second fieldwork was on 11th September, 2014, for the purposes of testing the classification results. In this study, spectral separability analysis was first performed to select appropriate MS bands for classification. Also, the reflectance analysis provided information for expanding sample points under the circumstance of knowing only a few. Then, a spatially weighted object-based k-nearest neighbour (k-NN) classifier was applied to the selected MS bands to identify seven land cover types (bamboo, conifer, broadleaf, mixed forest, brush, bare land, and shadow), accounting for spatial correlation within classes using geostatistical modelling. The spatially weighted k-NN method was compared with three alternatives: the traditional k-NN classifier, the Support Vector Machine (SVM) method and the Classification and Regression Tree (CART). Through field validation, it was proved that the classification result obtained using the spatially weighted k-NN method has the highest overall classification accuracy (77.61%) and Kappa coefficient (0.729); the producer’s accuracy and user’s accuracy achieve 81.25% and 95.12% for the bamboo class, respectively, also higher than the other methods. Photos of tree crowns were taken at sample locations using a fisheye camera, so the canopy density could be estimated. It is found that it is difficult to identify bamboo in the areas with a large canopy density (over 0.70); it is possible to extract bamboos in the areas with a median canopy density (from 0.2 to 0.7) and in a sparse forest (canopy density is less than 0.2). In summary, this study explores the ability of WV-2 imagery for bamboo extraction in a mountainous region in Sichuan. The study successfully identified the bamboo distribution, providing supporting knowledge for assessing the habitats of giant pandas.

Keywords: bamboo mapping, classification, geostatistics, k-NN, worldview-2

Procedia PDF Downloads 300

Authors: Souhir Chelly, Chahida Harizi, Aicha Hechaichi, Sihem Aissaoui, Leila Ben Ayed, Maha Bergaoui, Mohamed Kouni Chahed

Abstract:

Introduction: The International Classification of Primary Care (ICPC) is part of the morbidity classification system. It had 17 chapters, and each is coded by an alphanumeric code: the letter corresponds to the chapter, the number to a paragraph in the chapter. The objective of this study is to show the utility of this classification in the coding of the reasons for demand for care in Primary health care (PHC), its advantages and limits. Methods: This is a cross-sectional descriptive study conducted in 4 PHC in Ariana district. Data on the demand for care during 2 days in the same week were collected. The coding of the information was done according to the CISP. The data was entered and analyzed by the EPI Info 7 software. Results: A total of 523 demands for care were investigated. The patients who came for the consultation are predominantly female (62.72%). Most of the consultants are young with an average age of 35 ± 26 years. In the ICPC, there are 7 rubrics: 'infections' is the most common reason with 49.9%, 'other diagnoses' with 40.2%, 'symptoms and complaints' with 5.5%, 'trauma' with 2.1%, 'procedures' with 2.1% and 'neoplasm' with 0.3%. The main advantage of the ICPC is the fact of being a standardized tool. It is very suitable for classification of the reasons for demand for care in PHC according to their specificity, capacity to be used in a computerized medical file of the PHC. Its current limitations are related to the difficulty of classification of some reasons for demand for care. Conclusion: The ICPC has been developed to provide healthcare with a coding reference that takes into account their specificity. The CIM is in its 10th revision; it would gain from revision to revision to be more efficient to be generalized and used by the teams of PHC.

Keywords: international classification of primary care, medical file, primary health care, Tunisia

Procedia PDF Downloads 251

Authors: Chothmal, Basant Agarwal

Abstract:

Sentiment analysis means to classify a given review document into positive or negative polar document. Sentiment analysis research has been increased tremendously in recent times due to its large number of applications in the industry and academia. Sentiment analysis models can be used to determine the opinion of the user towards any entity or product. E-commerce companies can use sentiment analysis model to improve their products on the basis of users’ opinion. In this paper, we propose a new One-class Support Vector Machine (One-class SVM) based sentiment analysis model for movie review documents. In the proposed approach, we initially extract features from one class of documents, and further test the given documents with the one-class SVM model if a given new test document lies in the model or it is an outlier. Experimental results show the effectiveness of the proposed sentiment analysis model.

Keywords: feature selection methods, machine learning, NB, one-class SVM, sentiment analysis, support vector machine

Procedia PDF Downloads 500

Authors: B. S. Harish, M. B. Revanasiddappa

Abstract:

Due to rapid growth of text documents in digital form, automated text classification has become an important research in the last two decades. The major challenge of text document representations are high dimension, sparsity, volume and semantics. Since the terms are only features that can be found in documents, selection of good terms (features) plays an very important role. In text classification, feature selection is a strategy that can be used to improve classification effectiveness, computational efficiency and accuracy. In this paper, we present a quantitative analysis of most widely used feature selection (FS) methods, viz. Term Frequency-Inverse Document Frequency (tf
idf ), Mutual Information (MI), Information Gain (IG), CHISquare (x2), Term Frequency-Relevance Frequency (tf
rf ), Term Strength (TS), Ambiguity Measure (AM) and Symbolic Feature Selection (SFS) to classify text documents. We evaluated all the feature selection methods on standard datasets like 20 Newsgroups, 4 University dataset and Reuters-21578.

Keywords: classifiers, feature selection, text classification

Procedia PDF Downloads 447

Authors: Varun Agarwal

Abstract:

Introduction: With the advent of personalized medicine, histopathological review of whole slide images (WSIs) for cancer diagnosis presents an exceedingly time-consuming, complex task. Specifically, detecting metastatic regions in WSIs of sentinel lymph node biopsies necessitates a full-scanned, holistic evaluation of the image. Thus, digital pathology, low-level image manipulation algorithms, and machine learning provide significant advancements in improving the efficiency and accuracy of WSI analysis. Using Camelyon16 data, this paper proposes a deep learning pipeline to automate and ameliorate breast cancer metastasis localization and WSI classification. Methodology: The model broadly follows five stages -region of interest detection, WSI partitioning into image tiles, convolutional neural network (CNN) image-segment classifications, probabilistic mapping of tumor localizations, and further processing for whole WSI classification. Transfer learning is applied to the task, with the implementation of Inception-ResNetV2 - an effective CNN classifier that uses residual connections to enhance feature representation, adding convolved outputs in the inception unit to the proceeding input data. Moreover, in order to augment the performance of the transfer learning CNN, a stack of convolutional denoising autoencoders (CDAE) is applied to produce embeddings that enrich image representation. Through a saliency-detection algorithm, visual training segments are generated, which are then processed through a denoising autoencoder -primarily consisting of convolutional, leaky rectified linear unit, and batch normalization layers- and subsequently a contrast-normalization function. A spatial pyramid pooling algorithm extracts the key features from the processed image, creating a viable feature map for the CNN that minimizes spatial resolution and noise. Results and Conclusion: The simplified and effective architecture of the fine-tuned transfer learning Inception-ResNetV2 network enhanced with the CDAE stack yields state of the art performance in WSI classification and tumor localization, achieving AUC scores of 0.947 and 0.753, respectively. The convolutional feature retention and compilation with the residual connections to inception units synergized with the input denoising algorithm enable the pipeline to serve as an effective, efficient tool in the histopathological review of WSIs.

Keywords: breast cancer, convolutional neural networks, metastasis mapping, whole slide images

Procedia PDF Downloads 119

Authors: Tianyi Wang, Hieyong Jeong, An Guo, Yuko Ohno

Abstract:

Healthcare robot for assisting sit-to-stand (STS) performance had aroused numerous research interests. To author’s best knowledge, knowledge about how evaluating healthcare robot is still unknown. Robot should be labeled as fault if users feel demanding during STS when they are assisted by robot. In this research, we aim to propose a method to evaluate sit-to-stand assist robot through center of pressure (CoP), then classify different STS performance. Experiments were executed five times with ten healthy subjects under four conditions: two self-performed STSs with chair heights of 62 cm and 43 cm, and two robot-assisted STSs with chair heights of 43 cm and robot end-effect speed of 2 s and 5 s. CoP was measured using a Wii Balance Board (WBB). Bayesian classification was utilized to classify STS performance. The results showed that faults occurred when decreased the chair height and slowed robot assist speed. Proposed method for fault classification showed high probability of classifying fault classes form others. It was concluded that faults for STS assist robot could be detected by inspecting center of pressure and be classified through proposed classification algorithm.

Keywords: center of pressure, fault classification, healthcare robot, sit-to-stand movement

Procedia PDF Downloads 188

Authors: Chirawat Wattanapanich, Hong Wei

Abstract:

This study presents new gait representations for improving gait recognition accuracy on cross gait appearances, such as normal walking, wearing a coat and carrying a bag. Based on the Gait Energy Image (GEI), two ideas are implemented to generate new gait representations. One is to append lower knee regions to the original GEI, and the other is to apply convolutional operations to the GEI and its variants. A set of new gait representations are created and used for training multi-class Support Vector Machines (SVMs). Tests are conducted on the CASIA dataset B. Various combinations of the gait representations with different convolutional kernel size and different numbers of kernels used in the convolutional processes are examined. Both the entire images as features and reduced dimensional features by Principal Component Analysis (PCA) are tested in gait recognition. Interestingly, both new techniques, appending the lower knee regions to the original GEI and convolutional GEI, can significantly contribute to the performance improvement in the gait recognition. The experimental results have shown that the average recognition rate can be improved from 75.65% to 87.50%.

Keywords: convolutional image, lower knee, gait

Procedia PDF Downloads 193

Authors: Omid Shahmirzadi, Adam Lugowski, Kenneth Younge

Abstract:

Automatic measurement of semantic text similarity is an important task in natural language processing. In this paper, we evaluate the performance of different vector space models to perform this task. We address the real-world problem of modeling patent-to-patent similarity and compare TFIDF (and related extensions), topic models (e.g., latent semantic indexing), and neural models (e.g., paragraph vectors). Contrary to expectations, the added computational cost of text embedding methods is justified only when: 1) the target text is condensed; and 2) the similarity comparison is trivial. Otherwise, TFIDF performs surprisingly well in other cases: in particular for longer and more technical texts or for making finer-grained distinctions between nearest neighbors. Unexpectedly, extensions to the TFIDF method, such as adding noun phrases or calculating term weights incrementally, were not helpful in our context.

Keywords: big data, patent, text embedding, text similarity, vector space model

Procedia PDF Downloads 159

Authors: Jamshaid ul Rahman, Akhter Ali, Adnan Manzoor

Abstract:

Learning-based systems have received increasing interest in recent years; recognition structures, including end-to-end speak recognition, are one of the hot topics in this area. A famous work on end-to-end speaker verification by using Angular Softmax Loss gained significant importance and is considered useful to directly trains a discriminative model instead of the traditional adopted i-vector approach. The margin-based strategy in angular softmax is beneficial to learn discriminative speaker embeddings where the random selection of margin values is a big issue in additive angular margin and multiplicative angular margin. As a better solution in this matter, we present an alternative approach by introducing a bit similar form of an additive parameter that was originally introduced for face recognition, and it has a capacity to adjust automatically with the corresponding margin values and is applicable to learn more discriminative features than the Softmax. Experiments are conducted on the part of Fisher dataset, where it observed that the additive parameter with angular softmax to train the front-end and probabilistic linear discriminant analysis (PLDA) in the back-end boosts the performance of the structure.

Keywords: additive parameter, angular softmax, speaker verification, PLDA

Procedia PDF Downloads 85

Authors: Quznetsov Gunn

Abstract:

In the study of the logical foundations of probability theory, it was found that the terms and equations of the fundamental theoretical physics represent terms and theorems of the classical probability theory, more precisely, of that part of this theory, which considers the probability of dot events in the 3 + 1 space-time. In particular, the masses, moments, energies, spins, etc. turn out of parameters of probability distributions such events. The terms and the equations of the electroweak and of the quark-gluon theories turn out the theoretical-probabilistic terms and theorems. Here the relation of a neutrino to his lepton becomes clear, the W and Z bosons masses turn out dynamic ones, the cause of the asymmetry between particles and antiparticles is the impossibility of the birth of single antiparticles. In addition, phenomena such as confinement and asymptotic freedom receive their probabilistic explanation. And here we have the logical foundations of the gravity theory with phenomena dark energy and dark matter.

Keywords: classical theory of probability, logical foundation of fundamental theoretical physics, masses, moments, energies, spins

Procedia PDF Downloads 285

Authors: Jerzy Banski

Abstract:

The basic aim of this study is to discuss and assess different classifications and research approaches to small towns that take their social and economic functions into account, as well as relations with surrounding areas. The subject literature typically includes three types of approaches to the classification of small towns: 1) the structural, 2) the location-related, and 3) the mixed. The structural approach allows for the grouping of towns from the point of view of the social, cultural and economic functions they discharge. The location-related approach draws on the idea of there being a continuum between the center and the periphery. A mixed classification making simultaneous use of the different approaches to research brings the most information to bear in regard to categories of the urban locality. Bearing in mind the approaches to classification, it is possible to propose a synthetic method for classifying small towns that takes account of economic structure, location and the relationship between the towns and their surroundings. In the case of economic structure, the small centers may be divided into two basic groups – those featuring a multi-branch structure and those that are specialized economically. A second element of the classification reflects the locations of urban centers. Two basic types can be identified – the small town within the range of impact of a large agglomeration, or else the town outside such areas, which is to say located peripherally. The third component of the classification arises out of small towns’ relations with their surroundings. In consequence, it is possible to indicate 8 types of small-town: from local centers enjoying good accessibility and a multi-branch economic structure to peripheral supra-local centers characterised by a specialized economic structure.

Keywords: small towns, classification, functional structure, localization

Procedia PDF Downloads 170

Authors: Syed Basit Ali Shah Bukhari, Yan Qiang, Saad Abdul Rauf, Syed Saqlaina Bukhari

Abstract:

Text Classification is the methodology to classify any given text into the respective category from a given set of categories. It is highly important and vital to use proper set of pre-processing , feature selection and classification techniques to achieve this purpose. In this paper we have used different ensemble techniques along with variance in feature selection parameters to see the change in overall accuracy of the result and also on some other individual class based features which include precision value of each individual category of the text. After subjecting our data through pre-processing and feature selection techniques , different individual classifiers were tested first and after that classifiers were combined to form ensembles to increase their accuracy. Later we also studied the impact of decreasing the classification categories on over all accuracy of data. Text classification is highly used in sentiment analysis on social media sites such as twitter for realizing people’s opinions about any cause or it is also used to analyze customer’s reviews about certain products or services. Opinion mining is a vital task in data mining and text categorization is a back-bone to opinion mining.

Keywords: Natural Language Processing, Ensemble Classifier, Bagging Classifier, AdaBoost

Procedia PDF Downloads 219

Authors: Hernández Eva Selene, Reyna Mary Carmen, Rivera Héctor, Barragán Irving

Abstract:

We propose an approach that jointly addresses the layout of a facility and the scheduling of a sequence of jobs. In real production, these two problems are interrelated. However, they are treated separately in the literature. Our approach is an extension of the job shop problem with transportation delay, where the location of the machines is selected among possible sites. The model minimizes the makespan, using the short processing times rule with two algorithms; the first one considers all the permutations for the location of machines, and the second only a heuristic to select some specific permutations that reduces computational time. Some instances are proved and compared with literature.

Keywords: layout problem, job shop scheduling problem, concurrent scheduling and layout problem, metaheuristic

Procedia PDF Downloads 594

Authors: Taner Ersoz, Filiz Ersoz, Seyma Ozbilge

Abstract:

In this study, the clients who applied to a bank branch for loan were analyzed through data mining. The study was composed of the information such as amounts of loans received by personal and SME clients working with the bank branch, installment numbers, number of delays in loan installments, payments available in other banks and number of banks to which they are in debt between 2010 and 2013. The client risk profile was examined through Classification and Regression Tree (CART) analysis, one of the decision tree classification methods. At the end of the study, 5 different types of customers have been determined on the decision tree. The classification of these types of customers has been created with the rating of those posing a risk for the bank branch and the customers have been classified according to the risk ratings.

Keywords: client classification, loan suitability, risk rating, CART analysis

Procedia PDF Downloads 326

Authors: M. M. M. Jaradat

Abstract:

For a given graph G, the set Ԑ of all subsets of E(G) forms an |E(G)| dimensional vector space over Z2 with vector addition X⊕Y = (X\Y ) [ (Y \X) and scalar multiplication 1.X = X and 0.X = Ø for all X, Yϵ Ԑ. The cycle space, C(G), of a graph G is the vector subspace of (E; ⊕; .) spanned by the cycles of G. Traditionally there have been two notions of minimality among bases of C(G). First, a basis B of G is called a d-fold if each edge of G occurs in at most d cycles of the basis B. The basis number, b(G), of G is the least non-negative integer d such that C(G) has a d-fold basis; a required basis of C(G) is a basis for which each edge of G belongs to at most b(G) elements of B. Second, a basis B is called a minimum cycle basis (MCB) if its total length Σ BϵB |B| is minimum among all bases of C(G). The lexicographic product GρH has the vertex set V (GρH) = V (G) x V (H) and the edge set E(GρH) = {(u1, v1)(u2, v2)|u1 = u2 and v1 v2 ϵ E(H); or u1u2 ϵ E(G) and there is α ϵ Aut(H) such that α (v1) = v2}. In this work, a construction of a minimum cycle basis for the wreath product of wheels with paths is presented. Also, the length of the longest cycle of a minimum cycle basis is determined. Moreover, the basis number for the wreath product of the same is investigated.

Keywords: cycle space, minimum cycle basis, basis number, wreath product

Procedia PDF Downloads 267

Authors: Soumya Sajjan

Abstract:

Liver cancer is the most common cancer disease worldwide in men and women, and is one of the few cancers still on the rise. Liver disease is the 4th leading cause of death. According to new NHS (National Health Service) figures, deaths from liver diseases have reached record levels, rising by 25% in less than a decade; heavy drinking, obesity, and hepatitis are believed to be behind the rise. In this study, we focus on Development of Diagnostic Classifier for Ultrasound liver lesion. Ultrasound (US) Sonography is an easy-to-use and widely popular imaging modality because of its ability to visualize many human soft tissues/organs without any harmful effect. This paper will provide an overview of underlying concepts, along with algorithms for processing of liver ultrasound images Naturaly, Ultrasound liver lesion images are having more spackle noise. Developing classifier for ultrasound liver lesion image is a challenging task. We approach fully automatic machine learning system for developing this classifier. First, we segment the liver image by calculating the textural features from co-occurrence matrix and run length method. For classification, Support Vector Machine is used based on the risk bounds of statistical learning theory. The textural features for different features methods are given as input to the SVM individually. Performance analysis train and test datasets carried out separately using SVM Model. Whenever an ultrasonic liver lesion image is given to the SVM classifier system, the features are calculated, classified, as normal and diseased liver lesion. We hope the result will be helpful to the physician to identify the liver cancer in non-invasive method.

Keywords: segmentation, Support Vector Machine, ultrasound liver lesion, co-occurance Matrix

Procedia PDF Downloads 395

Authors: F. Jiménez, R. Jódar, M. Martín, G. Sánchez, G. Sciavicco

Abstract:

Abstract—Attribute or feature selection is one of the basic strategies to improve the performances of data classification tasks, and, at the same time, to reduce the complexity of classifiers, and it is a particularly fundamental one when the number of attributes is relatively high. Its application to unsupervised classification is restricted to a limited number of experiments in the literature. Evolutionary computation has already proven itself to be a very effective choice to consistently reduce the number of attributes towards a better classification rate and a simpler semantic interpretation of the inferred classifiers. We present a feature selection wrapper model composed by a multi-objective evolutionary algorithm, the clustering method Expectation-Maximization (EM), and the classifier C4.5 for the unsupervised classification of data extracted from a psychological test named BASC-II (Behavior Assessment System for Children - II ed.) with two objectives: Maximizing the likelihood of the clustering model and maximizing the accuracy of the obtained classifier. We present a methodology to integrate feature selection for unsupervised classification, model evaluation, decision making (to choose the most satisfactory model according to a a posteriori process in a multi-objective context), and testing. We compare the performance of the classifier obtained by the multi-objective evolutionary algorithms ENORA and NSGA-II, and the best solution is then validated by the psychologists that collected the data.

Keywords: evolutionary computation, feature selection, classification, clustering

Procedia PDF Downloads 355

Authors: Vishwa Joshi

Abstract:

The study of mood recognition in the field of music has gained a lot of momentum in the recent years with machine learning and data mining techniques and many audio features contributing considerably to analyze and identify the relation of mood plus music. In this paper we consider the same idea forward and come up with making an effort to build a system for automatic recognition of mood underlying the audio song’s clips by mining their audio features and have evaluated several data classification algorithms in order to learn, train and test the model describing the moods of these audio songs and developed an open source framework. Before classification, Preprocessing and Feature Extraction phase is necessary for removing noise and gathering features respectively.

Keywords: music, mood, features, classification

Procedia PDF Downloads 486

Authors: Jorge A. Ruiz-Vanoye, Ocotlán Díaz-Parra, Alejandro Fuentes-Penna, Daniel Vélez-Díaz, Edith Olaco García

Abstract:

In this paper, we present the use of the discriminant analysis to select evolutionary algorithms that better solve instances of the vehicle routing problem with time windows. We use indicators as independent variables to obtain the classification criteria, and the best algorithm from the generic genetic algorithm (GA), random search (RS), steady-state genetic algorithm (SSGA), and sexual genetic algorithm (SXGA) as the dependent variable for the classification. The discriminant classification was trained with classic instances of the vehicle routing problem with time windows obtained from the Solomon benchmark. We obtained a classification of the discriminant analysis of 66.7%.

Keywords: Intelligent Transportation Systems, data-mining techniques, evolutionary algorithms, discriminant analysis, machine learning

Procedia PDF Downloads 453

Authors: Despoina Mouratidis, Katia Lida Kermanidis

Abstract:

Machine translation (MT) is a very important task in Natural Language Processing (NLP). MT evaluation is crucial in MT development, as it constitutes the means to assess the success of an MT system, and also helps improve its performance. Several methods have been proposed for the evaluation of (MT) systems. Some of the most popular ones in automatic MT evaluation are score-based, such as the BLEU score, and others are based on lexical similarity or syntactic similarity between the MT outputs and the reference involving higher-level information like part of speech tagging (POS). This paper presents a language-independent machine learning framework for classifying pairwise translations. This framework uses vector representations of two machine-produced translations, one from a statistical machine translation model (SMT) and one from a neural machine translation model (NMT). The vector representations consist of automatically extracted word embeddings and string-like language-independent features. These vector representations used as an input to a multi-layer neural network (NN) that models the similarity between each MT output and the reference, as well as between the two MT outputs. To evaluate the proposed approach, a professional translation and a "ground-truth" annotation are used. The parallel corpora used are English-Greek (EN-GR) and English-Italian (EN-IT), in the educational domain and of informal genres (video lecture subtitles, course forum text, etc.) that are difficult to be reliably translated. They have tested three basic deep learning (DL) architectures to this schema: (i) fully-connected dense, (ii) Convolutional Neural Network (CNN), and (iii) Long Short-Term Memory (LSTM). Experiments show that all tested architectures achieved better results when compared against those of some of the well-known basic approaches, such as Random Forest (RF) and Support Vector Machine (SVM). Better accuracy results are obtained when LSTM layers are used in our schema. In terms of a balance between the results, better accuracy results are obtained when dense layers are used. The reason for this is that the model correctly classifies more sentences of the minority class (SMT). For a more integrated analysis of the accuracy results, a qualitative linguistic analysis is carried out. In this context, problems have been identified about some figures of speech, as the metaphors, or about certain linguistic phenomena, such as per etymology: paronyms. It is quite interesting to find out why all the classifiers led to worse accuracy results in Italian as compared to Greek, taking into account that the linguistic features employed are language independent.

Keywords: machine learning, machine translation evaluation, neural network architecture, pairwise classification

Procedia PDF Downloads 116

Authors: C. Lanzerstorfer

Abstract:

The off-gas from the basic oxygen furnace (BOF), where pig iron is converted into steel, is treated in the primary ventilation system. This system is in full operation only during oxygen-blowing when the BOF converter vessel is in a vertical position. When pig iron and scrap are charged into the BOF and when slag or steel are tapped, the vessel is tilted. The generated emissions during charging and tapping cannot be captured by the primary off-gas system. To capture these emissions, a secondary ventilation system is usually installed. The emissions are captured by a canopy hood installed just above the converter mouth in tilted position. The aim of this study was to investigate the dependence of Zn and other components on the particle size of BOF secondary ventilation dust. Because of the high temperature of the BOF process it can be expected that Zn will be enriched in the fine dust fractions. If Zn is enriched in the fine fractions, classification could be applied to split the dust into two size fractions with a different content of Zn. For this air classification experiments with dust from the secondary ventilation system of a BOF were performed. The results show that Zn and Pb are highly enriched in the finest dust fraction. For Cd, Cu and Sb the enrichment is less. In contrast, the non-volatile metals Al, Fe, Mn and Ti were depleted in the fine fractions. Thus, air classification could be considered for the treatment of dust from secondary BOF off-gas cleaning.

Keywords: air classification, converter dust, recycling, zinc

Procedia PDF Downloads 418

Authors: Jiahe Liu, Hongyang Yu, Feng Qian, Miao Luo

Abstract:

SMPL-X was a powerful parametric human body model that included male, neutral, and female models, with significant gender differences between these three models. During the process of 3D human body reconstruction, the correct selection of standard templates was crucial for obtaining accurate results. To address this issue, we developed an efficient gender classification algorithm to automatically select the appropriate template for 3D human body reconstruction. The key to this gender classification algorithm was the precise analysis of human body features. By using the SMPL-X model, the algorithm could detect and identify gender features of the human body, thereby determining which standard template should be used. The accuracy of this algorithm made the 3D reconstruction process more accurate and reliable, as it could adjust model parameters based on individual gender differences. SMPL-X and the related gender classification algorithm have brought important advancements to the field of 3D human body reconstruction. By accurately selecting standard templates, they have improved the accuracy of reconstruction and have broad potential in various application fields. These technologies continue to drive the development of the 3D reconstruction field, providing us with more realistic and accurate human body models.

Keywords: gender classification, joint detection, SMPL-X, 3D reconstruction

Procedia PDF Downloads 57

Authors: Zhong Ma, Zhuping Wang, Congxin Liu, Xiangzeng Liu

Abstract:

Satellite imagery classification is a challenging problem with many practical applications. In this paper, we designed a deep convolution neural network (DCNN) to classify the satellite imagery. The contributions of this paper are twofold — First, to cope with the large-scale variance in the satellite image, we introduced the inception module, which has multiple filters with different size at the same level, as the building block to build our DCNN model. Second, we proposed a genetic algorithm based method to efficiently search the best hyper-parameters of the DCNN in a large search space. The proposed method is evaluated on the benchmark database. The results of the proposed hyper-parameters search method show it will guide the search towards better regions of the parameter space. Based on the found hyper-parameters, we built our DCNN models, and evaluated its performance on satellite imagery classification, the results show the classification accuracy of proposed models outperform the state of the art method.

Keywords: satellite imagery classification, deep convolution network, genetic algorithm, hyper-parameter optimization

Procedia PDF Downloads 284

Authors: Omid Khodabakhshi, Amir Rozdel

Abstract:

The subject of cloud computing security research has allocated a number of challenges and competitions because the data center is comprised of complex private information and are always faced various risks of information disclosure by hacker attacks or internal enemies. Accordingly, the security of virtual machines in the cloud computing infrastructure layer is very important. So far, there are many software solutions to develop security in virtual machines. But using software alone is not enough to solve security problems. The purpose of this article is to examine the challenges and security requirements for accessing and storing data in an insecure cloud environment. In other words, in this article, a structure is proposed for the implementation of highly isolated security-sensitive codes using secure computing hardware in virtual environments. It also allows remote code validation with inputs and outputs. We provide these security features even in situations where the BIOS, the operating system, and even the super-supervisor are infected. To achieve these goals, we will use the hardware support provided by the new Intel and AMD processors, as well as the TPM security chip. In conclusion, the use of these technologies ultimately creates a root of dynamic trust and reduces TCB to security-sensitive codes.

Keywords: code, cloud computing, security, virtual machines

Procedia PDF Downloads 179

Authors: Qamar Iqbal

Abstract:

Companies strive to improve their forecasting methods to predict the fluctuations in customer demand. These fluctuation and variation in demand affect the manufacturing operations and can limit a company’s ability to fulfill customer demand on time. Companies keep the inventory buffer and maintain the stocking levels to reduce the impact of demand variation. A mid-size company deals with thousands of stock keeping units (skus). It is neither easy and nor efficient to control and manage each sku. Inventory classification provides a tool to the management to increase their ability to support customer demand. The paper presents a framework that shows how inventory classification can play a role to increase supply chain responsiveness. A case study will be presented to further elaborate the method both for build-to-order and build-to-forecast manufacturing environments. Results will be compared that will show which manufacturing setting has advantage over another under different circumstances. The outcome of this study is very useful to the management because this will give them an insight on how inventory classification can be used to increase their ability to respond to changing customer needs.

Keywords: inventory classification, supply chain responsiveness, forecast, manufacturing environment

Procedia PDF Downloads 588

Authors: Ying Yi Wu

Abstract:

The study of finite groups is a central topic in algebraic structures, and one of the most fundamental questions in this field is the classification of finite groups up to isomorphism. In this paper, we investigate the cyclic property of groups of prime order, which is a crucial result in the classification of finite abelian groups. We prove the following statement: If p is a prime, then every group G of order p is cyclic. Our proof utilizes the properties of group actions and the class equation, which provide a powerful tool for studying the structure of finite groups. In particular, we first show that any non-identity element of G generates a cyclic subgroup of G. Then, we establish the existence of an element of order p, which implies that G is generated by a single element. Finally, we demonstrate that any two generators of G are conjugate, which shows that G is a cyclic group. Our result has significant implications in the classification of finite groups, as it implies that any group of prime order is isomorphic to the cyclic group of the same order. Moreover, it provides a useful tool for understanding the structure of more complicated finite groups, as any finite abelian group can be decomposed into a direct product of cyclic groups. Our proof technique can also be extended to other areas of group theory, such as the classification of finite p-groups, where p is a prime. Therefore, our work has implications beyond the specific result we prove and can contribute to further research in algebraic structures.

Keywords: group theory, finite groups, cyclic groups, prime order, classification.

Procedia PDF Downloads 75

Authors: K. A. Chinmaya, M. Udaya Bhaskar

Abstract:

Multiphase motor drives are now a day considered for numerous applications due to the advantages that they offer when compared to their three-phase counterparts. Proper modeling of inverters and motors are important in devising an appropriate control algorithm. This paper develops a complete modeling of a five-phase inverter and five-phase space vector modulation schemes which can be used for five-phase motor drives. A novel modified algorithm is introduced which enables the sinusoidal output voltages up to certain voltage value. The waveforms of phase to neutral voltage are compared with the different modulation techniques and also different modulation indexes in terms of Low-order Harmonic (LH) voltage of 3rd and 7th present. A detailed performance evolution of existing and newly modified schemes is done in terms of Total Harmonic Distortion (THD).

Keywords: multi-phase drives, space vector modulation, voltage source inverter, low order harmonic voltages, total harmonic distortion

Procedia PDF Downloads 390

Authors: Farid Khosravikia, Patricia Clayton

Abstract:

The main objective of this work is to evaluate the advantages and disadvantages of various machine learning techniques in two key steps of seismic hazard and risk assessment of different types of structures. The first step is the development of ground-motion models, which are used for forecasting ground-motion intensity measures (IM) given source characteristics, source-to-site distance, and local site condition for future events. IMs such as peak ground acceleration and velocity (PGA and PGV, respectively) as well as 5% damped elastic pseudospectral accelerations at different periods (PSA), are indicators of the strength of shaking at the ground surface. Typically, linear regression-based models, with pre-defined equations and coefficients, are used in ground motion prediction. However, due to the restrictions of the linear regression methods, such models may not capture more complex nonlinear behaviors that exist in the data. Thus, this study comparatively investigates potential benefits from employing other machine learning techniques as statistical method in ground motion prediction such as Artificial Neural Network, Random Forest, and Support Vector Machine. The results indicate the algorithms satisfy some physically sound characteristics such as magnitude scaling distance dependency without requiring pre-defined equations or coefficients. Moreover, it is shown that, when sufficient data is available, all the alternative algorithms tend to provide more accurate estimates compared to the conventional linear regression-based method, and particularly, Random Forest outperforms the other algorithms. However, the conventional method is a better tool when limited data is available. Second, it is investigated how machine learning techniques could be beneficial for developing probabilistic seismic demand models (PSDMs), which provide the relationship between the structural demand responses (e.g., component deformations, accelerations, internal forces, etc.) and the ground motion IMs. In the risk framework, such models are used to develop fragility curves estimating exceeding probability of damage for pre-defined limit states, and therefore, control the reliability of the predictions in the risk assessment. In this study, machine learning algorithms like artificial neural network, random forest, and support vector machine are adopted and trained on the demand parameters to derive PSDMs for them. It is observed that such models can provide more accurate estimates of prediction in relatively shorter about of time compared to conventional methods. Moreover, they can be used for sensitivity analysis of fragility curves with respect to many modeling parameters without necessarily requiring more intense numerical response-history analysis.

Keywords: artificial neural network, machine learning, random forest, seismic risk analysis, seismic hazard analysis, support vector machine

Procedia PDF Downloads 92