Search results for: data preprocessing

Authors: Leila Keshavarz Afshar, Hedieh Sajedi

Abstract:

Estimation of biological brain age from MR images is a topic that has been much addressed in recent years due to the importance it attaches to early diagnosis of diseases such as Alzheimer's. In this paper, we use a 3D Convolutional Neural Network (CNN) to provide a method for estimating the biological age of the brain. The 3D-CNN model is trained by MRI data that has been normalized. In addition, to reduce computation while saving overall performance, some effectual slices are selected for age estimation. By this method, the biological age of individuals using selected normalized data was estimated with Mean Absolute Error (MAE) of 4.82 years.

Keywords: brain age estimation, biological age, 3D-CNN, deep learning, T1-weighted image, SPM, preprocessing, MRI, canny, gray matter

Procedia PDF Downloads 142

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 492

Authors: Fuad Noman, Sh-Hussain Salleh, Chee-Ming Ting, Hadri Hussain, Syed Rasul

Abstract:

In this paper, a data-driven dictionary approach is proposed for the automatic detection and classification of cardiovascular abnormalities. Electrocardiography (ECG) signal is represented by the trained complete dictionaries that contain prototypes or atoms to avoid the limitations of pre-defined dictionaries. The data-driven trained dictionaries simply take the ECG signal as input rather than extracting features to study the set of parameters that yield the most descriptive dictionary. The approach inherently learns the complicated morphological changes in ECG waveform, which is then used to improve the classification. The classification performance was evaluated with ECG data under two different preprocessing environments. In the first category, QT-database is baseline drift corrected with notch filter and it filters the 60 Hz power line noise. In the second category, the data are further filtered using fast moving average smoother. The experimental results on QT database confirm that our proposed algorithm shows a classification accuracy of 92%.

Keywords: electrocardiogram, dictionary learning, sparse coding, classification

Procedia PDF Downloads 378

Authors: Teklu Urgessa, Wookjae Maeng, Joong Seek Lee

Abstract:

Application of five implementations of three data mining classification techniques was experimented for extracting important insights from tourism data. The aim was to find out the best performing algorithm among the compared ones for tourism knowledge discovery. Knowledge discovery process from data was used as a process model. 10-fold cross validation method is used for testing purpose. Various data preprocessing activities were performed to get the final dataset for model building. Classification models of the selected algorithms were built with different scenarios on the preprocessed dataset. The outperformed algorithm tourism dataset was Random Forest (76%) before applying information gain based attribute selection and J48 (C4.5) (75%) after selection of top relevant attributes to the class (target) attribute. In terms of time for model building, attribute selection improves the efficiency of all algorithms. Artificial Neural Network (multilayer perceptron) showed the highest improvement (90%). The rules extracted from the decision tree model are presented, which showed intricate, non-trivial knowledge/insight that would otherwise not be discovered by simple statistical analysis with mediocre accuracy of the machine using classification algorithms.

Keywords: classification algorithms, data mining, knowledge discovery, tourism

Procedia PDF Downloads 291

Authors: Ole-Christian Galbo Engstrøm, Erik Schou Dreier, Birthe Møller Jespersen, Kim Steenstrup Pedersen

Abstract:

Grain quality analysis is a multi-parameterized problem that includes a variety of qualitative and quantitative parameters such as grain type classification, damage type classification, and nutrient regression. Currently, these parameters require human inspection, a multitude of instruments employing a variety of sensor technologies, and predictive model types or destructive and slow chemical analysis. This paper investigates the feasibility of applying near-infrared hyperspectral imaging (NIR-HSI) to grain quality analysis. For this study two datasets of NIR hyperspectral images in the wavelength range of 900 nm - 1700 nm have been used. Both datasets contain images of sparsely and densely packed grain kernels. The first dataset contains ~87,000 image crops of bulk wheat samples from 63 harvests where protein value has been determined by the FOSS Infratec NOVA which is the golden industry standard for protein content estimation in bulk samples of cereal grain. The second dataset consists of ~28,000 image crops of bulk grain kernels from seven different wheat varieties and a single rye variety. In the first dataset, protein regression analysis is the problem to solve while variety classification analysis is the problem to solve in the second dataset. Deep convolutional neural networks (CNNs) have the potential to utilize spatio-spectral correlations within a hyperspectral image to simultaneously estimate the qualitative and quantitative parameters. CNNs can autonomously derive meaningful representations of the input data reducing the need for advanced preprocessing techniques required for classical chemometric model types such as artificial neural networks (ANNs) and partial least-squares regression (PLS-R). A comparison between different CNN architectures utilizing 2D and 3D convolution is conducted. These results are compared to the performance of ANNs and PLS-R. Additionally, a variety of preprocessing techniques from image analysis and chemometrics are tested. These include centering, scaling, standard normal variate (SNV), Savitzky-Golay (SG) filtering, and detrending. The results indicate that the combination of NIR-HSI and CNNs has the potential to be the foundation for an automatic system unifying qualitative and quantitative grain quality analysis within a single sensor technology and predictive model type.

Keywords: deep learning, grain analysis, hyperspectral imaging, preprocessing techniques

Procedia PDF Downloads 96

Authors: Tawfik Thelaidjia, Salah Chenikher

Abstract:

Bearing fault diagnosis has attracted significant attention over the past few decades. It consists of two major parts: vibration signal feature extraction and condition classification for the extracted features. In this paper, feature extraction from faulty bearing vibration signals is performed by a combination of the signal’s Kurtosis and features obtained through the preprocessing of the vibration signal samples using Db2 discrete wavelet transform at the fifth level of decomposition. In this way, a 7-dimensional vector of the vibration signal feature is obtained. After feature extraction from vibration signal, the support vector machine (SVM) was applied to automate the fault diagnosis procedure. To improve the classification accuracy for bearing fault prediction, particle swarm optimization (PSO) is employed to simultaneously optimize the SVM kernel function parameter and the penalty parameter. The results have shown feasibility and effectiveness of the proposed approach

Keywords: condition monitoring, discrete wavelet transform, fault diagnosis, kurtosis, machine learning, particle swarm optimization, roller bearing, rotating machines, support vector machine, vibration measurement

Procedia PDF Downloads 434

Authors: Kemal Efe Eseller, Göktuğ Yazici

Abstract:

Laser-induced breakdown spectroscopy (LIBS) is a rapid optical atomic emission spectroscopy which is used for material identification and analysis with the advantages of in-situ analysis, elimination of intensive sample preparation, and micro-destructive properties for the material to be tested. LIBS delivers short pulses of laser beams onto the material in order to create plasma by excitation of the material to a certain threshold. The plasma characteristics, which consist of wavelength value and intensity amplitude, depends on the material and the experiment’s environment. In the present work, medicine samples’ spectrum profiles were obtained via LIBS. Medicine samples’ datasets include two different concentrations for both paracetamol based medicines, namely Aferin and Parafon. The spectrum data of the samples were preprocessed via filling outliers based on quartiles, smoothing spectra to eliminate noise and normalizing both wavelength and intensity axis. Statistical information was obtained and principal component analysis (PCA) was incorporated to both the preprocessed and raw datasets. The machine learning models were set based on two different train-test splits, which were 70% training – 30% test and 80% training – 20% test. Cross-validation was preferred to protect the models against overfitting; thus the sample amount is small. The machine learning results of preprocessed and raw datasets were subjected to comparison for both splits. This is the first time that all supervised machine learning classification algorithms; consisting of Decision Trees, Discriminant, naïve Bayes, Support Vector Machines (SVM), k-NN(k-Nearest Neighbor) Ensemble Learning and Neural Network algorithms; were incorporated to LIBS data of paracetamol based pharmaceutical samples, and their different concentrations on preprocessed and raw dataset in order to observe the effect of preprocessing.

Keywords: machine learning, laser-induced breakdown spectroscopy, medicines, principal component analysis, preprocessing

Procedia PDF Downloads 85

Authors: Man-Yun Liu, Emily Chia-Yu Su

Abstract:

Alzheimer's disease (AD) is the public health crisis of the 21st century. AD is a degenerative brain disease and the most common cause of dementia, a costly disease on the healthcare system. Unfortunately, the cause of AD is poorly understood, furthermore; the treatments of AD so far can only alleviate symptoms rather cure or stop the progress of the disease. Currently, there are several ways to diagnose AD; medical imaging can be used to distinguish between AD, other dementias, and early onset AD, and cerebrospinal fluid (CSF). Compared with other diagnostic tools, blood (plasma) test has advantages as an approach to population-based disease screening because it is simpler, less invasive also cost effective. In our study, we used blood biomarkers dataset of The Alzheimer’s disease Neuroimaging Initiative (ADNI) which was funded by National Institutes of Health (NIH) to do data analysis and develop a prediction model. We used independent analysis of datasets to identify plasma protein biomarkers predicting early onset AD. Firstly, to compare the basic demographic statistics between the cohorts, we used SAS Enterprise Guide to do data preprocessing and statistical analysis. Secondly, we used logistic regression, neural network, decision tree to validate biomarkers by SAS Enterprise Miner. This study generated data from ADNI, contained 146 blood biomarkers from 566 participants. Participants include cognitive normal (healthy), mild cognitive impairment (MCI), and patient suffered Alzheimer’s disease (AD). Participants’ samples were separated into two groups, healthy and MCI, healthy and AD, respectively. We used the two groups to compare important biomarkers of AD and MCI. In preprocessing, we used a t-test to filter 41/47 features between the two groups (healthy and AD, healthy and MCI) before using machine learning algorithms. Then we have built model with 4 machine learning methods, the best AUC of two groups separately are 0.991/0.709. We want to stress the importance that the simple, less invasive, common blood (plasma) test may also early diagnose AD. As our opinion, the result will provide evidence that blood-based biomarkers might be an alternative diagnostics tool before further examination with CSF and medical imaging. A comprehensive study on the differences in blood-based biomarkers between AD patients and healthy subjects is warranted. Early detection of AD progression will allow physicians the opportunity for early intervention and treatment.

Keywords: Alzheimer's disease, blood-based biomarkers, diagnostics, early detection, machine learning

Procedia PDF Downloads 318

Authors: Jazim Sohail, Filipe Teixeira-Dias

Abstract:

Mild traumatic brain injury (mTBI) within non-helmeted contact sports is a growing concern due to the serious risk of potential injury. Extensive research is being conducted looking into head kinematics in non-helmeted contact sports utilizing instrumented mouthguards that allow researchers to record accelerations and velocities of the head during and after an impact. This does not, however, allow the location of the impact on the head, and its magnitude and orientation, to be determined. This research proposes and validates two methods to quantify impact locations from instrumented mouthguard kinematic data, one using rigid body dynamics, the other utilizing machine learning. The rigid body dynamics technique focuses on establishing and matching moments from Euler’s and torque equations in order to find the impact location on the head. The methodology is validated with impact data collected from a lab test with the dummy head fitted with an instrumented mouthguard. Additionally, a Hybrid III Dummy head finite element model was utilized to create synthetic kinematic data sets for impacts from varying locations to validate the impact location algorithm. The algorithm calculates accurate impact locations; however, it will require preprocessing of live data, which is currently being done by cross-referencing data timestamps to video footage. The machine learning technique focuses on eliminating the preprocessing aspect by establishing trends within time-series signals from instrumented mouthguards to determine the impact location on the head. An unsupervised learning technique is used to cluster together impacts within similar regions from an entire time-series signal. The kinematic signals established from mouthguards are converted to the frequency domain before using a clustering algorithm to cluster together similar signals within a time series that may span the length of a game. Impacts are clustered within predetermined location bins. The same Hybrid III Dummy finite element model is used to create impacts that closely replicate on-field impacts in order to create synthetic time-series datasets consisting of impacts in varying locations. These time-series data sets are used to validate the machine learning technique. The rigid body dynamics technique provides a good method to establish accurate impact location of impact signals that have already been labeled as true impacts and filtered out of the entire time series. However, the machine learning technique provides a method that can be implemented with long time series signal data but will provide impact location within predetermined regions on the head. Additionally, the machine learning technique can be used to eliminate false impacts captured by sensors saving additional time for data scientists using instrumented mouthguard kinematic data as validating true impacts with video footage would not be required.

Keywords: head impacts, impact location, instrumented mouthguard, machine learning, mTBI

Procedia PDF Downloads 214

Authors: Soha A. Bahanshal, Byung G. Kim

Abstract:

Identification of patients at high risk for hospital readmission is of crucial importance for quality health care and cost reduction. Predicting hospital readmissions among diabetic patients has been of great interest to many researchers and health decision makers. We build a prediction model to predict hospital readmission for diabetic patients within 30 days of discharge. The core of the prediction model is a modified k Nearest Neighbor called Hybrid Fuzzy Weighted k Nearest Neighbor algorithm. The prediction is performed on a patient dataset which consists of more than 70,000 patients with 50 attributes. We applied data preprocessing using different techniques in order to handle data imbalance and to fuzzify the data to suit the prediction algorithm. The model so far achieved classification accuracy of 80% compared to other models that only use k Nearest Neighbor.

Keywords: machine learning, prediction, classification, hybrid fuzzy weighted k-nearest neighbor, diabetic hospital readmission

Procedia PDF Downloads 181

Authors: Hyun-Woo Cho

Abstract:

It is quite important to utilize right time and intelligent production monitoring and diagnosis of industrial processes in terms of quality and safety issues. When compared with monitoring task, fault diagnosis represents the task of finding process variables responsible causing a specific fault in the process. It can be helpful to process operators who should investigate and eliminate root causes more effectively and efficiently. This work focused on the active use of combining a nonlinear statistical technique with a preprocessing method in order to implement practical real-time fault identification schemes for data-rich cases. To compare its performance to existing identification schemes, a case study on a benchmark process was performed in several scenarios. The results showed that the proposed fault identification scheme produced more reliable diagnosis results than linear methods. In addition, the use of the filtering step improved the identification results for the complicated processes with massive data sets.

Keywords: diagnosis, filtering, nonlinear statistical techniques, process monitoring

Procedia PDF Downloads 239

Authors: Ayad Dalloo, Sulaf Dalloo

Abstract:

Electrocardiogram (ECG) recording is prone to complications, on analysis by physicians, due to noise and artifacts, thus creating ambiguity leading to possible error of diagnosis. Such drawbacks may be overcome with the advent of high resolution Methods, such as Discrete Wavelet Analysis and Digital Signal Processing (DSP) techniques. This ECG signal analysis is implemented in three stages: ECG preprocessing, features extraction and classification with the aim of realizing high resolution ECG diagnosis and improved detection of abnormal conditions in the heart. The preprocessing stage involves removing spurious artifacts (noise), due to such factors as muscle contraction, motion, respiration, etc. ECG features are extracted by applying DSP and suggested sloping method techniques. These measured features represent peak amplitude values and intervals of P, Q, R, S, R’, and T waves on ECG, and other features such as ST elevation, QRS width, heart rate, electrical axis, QR and QT intervals. The classification is preformed using these extracted features and the criteria for cardiovascular diseases. The ECG diagnostic system is successfully applied to 12-lead ECG recordings for 12 cases. The system is provided with information to enable it diagnoses 15 different diseases. Physician’s and computer’s diagnoses are compared with 90% agreement, with respect to physician diagnosis, and the time taken for diagnosis is 2 seconds. All of these operations are programmed in Matlab environment.

Keywords: ECG diagnostic system, QRS detection, ECG baseline removal, cardiovascular diseases

Procedia PDF Downloads 293

Authors: Tomas Premoli, Sareh Rowlands

Abstract:

In this study, we investigate the impact of various convolutional neural network (CNN) architectures on the accuracy of diagnosing Alzheimer’s disease (AD) using patient MRI scans. Alzheimer’s disease is a debilitating neurodegenerative disorder that affects millions worldwide. Early, accurate, and non-invasive diagnostic methods are required for providing optimal care and symptom management. Deep learning techniques, particularly CNNs, have shown great promise in enhancing this diagnostic process. We aim to contribute to the ongoing research in this field by comparing the effectiveness of different CNN architectures and providing insights for future studies. Our methodology involved preprocessing MRI data, implementing multiple CNN architectures, and evaluating the performance of each model. We employed intensity normalization, linear registration, and skull stripping for our preprocessing. The selected architectures included VGG, ResNet, and DenseNet models, all implemented using the Keras library. We employed transfer learning and trained models from scratch to compare their effectiveness. Our findings demonstrated significant differences in performance among the tested architectures, with DenseNet201 achieving the highest accuracy of 86.4%. Transfer learning proved to be helpful in improving model performance. We also identified potential areas for future research, such as experimenting with other architectures, optimizing hyperparameters, and employing fine-tuning strategies. By providing a comprehensive analysis of the selected CNN architectures, we offer a solid foundation for future research in Alzheimer’s disease diagnosis using deep learning techniques. Our study highlights the potential of CNNs as a valuable diagnostic tool and emphasizes the importance of ongoing research to develop more accurate and effective models.

Keywords: Alzheimer’s disease, convolutional neural networks, deep learning, medical imaging, MRI

Procedia PDF Downloads 70

Authors: Mikiyas Adugna

Abstract:

Effort estimation is among the activities carried out in software development processes. An accurate model of estimation leads to project success. The method of agile effort estimation is a complex task because of the dynamic nature of software development. Researchers are still conducting studies on agile effort estimation to enhance prediction accuracy. Due to these reasons, we investigated and proposed a model on LASSO and Elastic Net regression to enhance estimation accuracy. The proposed model has major components: preprocessing, train-test split, training with default parameters, and cross-validation. During the preprocessing phase, the entire dataset is normalized. After normalization, a train-test split is performed on the dataset, setting training at 80% and testing set to 20%. We chose two different phases for training the two algorithms (Elastic Net and LASSO) regression following the train-test-split. In the first phase, the two algorithms are trained using their default parameters and evaluated on the testing data. In the second phase, the grid search technique (the grid is used to search for tuning and select optimum parameters) and 5-fold cross-validation to get the final trained model. Finally, the final trained model is evaluated using the testing set. The experimental work is applied to the agile story point dataset of 21 software projects collected from six firms. The results show that both Elastic Net and LASSO regression outperformed the compared ones. Compared to the proposed algorithms, LASSO regression achieved better predictive performance and has acquired PRED (8%) and PRED (25%) results of 100.0, MMRE of 0.0491, MMER of 0.0551, MdMRE of 0.0593, MdMER of 0.063, and MSE of 0.0007. The result implies LASSO regression algorithm trained model is the most acceptable, and higher estimation performance exists in the literature.

Keywords: agile software development, effort estimation, elastic net regression, LASSO

Procedia PDF Downloads 63

Authors: Sikiru Ademola Adewale, Joe Thomas, Bolanle Hafiz Matti, Tosin Ige

Abstract:

This project demonstrates the implementation and use of an encoder-decoder model to perform a many-to-many mapping of video data to text captions. The many-to-many mapping occurs via an input temporal sequence of video frames to an output sequence of words to form a caption sentence. Data preprocessing, model construction, and model training are discussed. Caption correctness is evaluated using 2-gram BLEU scores across the different splits of the dataset. Specific examples of output captions were shown to demonstrate model generality over the video temporal dimension. Predicted captions were shown to generalize over video action, even in instances where the video scene changed dramatically. Model architecture changes are discussed to improve sentence grammar and correctness.

Keywords: decoder, encoder, many-to-many mapping, video captioning, 2-gram BLEU

Procedia PDF Downloads 100

Authors: Xiaogang Li, Jieqiong Miao

Abstract:

As for the problem of the grey forecasting model prediction accuracy is low, an improved grey prediction model is put forward. Firstly, use trigonometric function transform the original data sequence in order to improve the smoothness of data , this model called SGM( smoothness of grey prediction model), then combine the improved grey model with support vector machine , and put forward the grey support vector machine model (SGM - SVM).Before the establishment of the model, we use trigonometric functions and accumulation generation operation preprocessing data in order to enhance the smoothness of the data and weaken the randomness of the data, then use support vector machine (SVM) to establish a prediction model for pre-processed data and select model parameters using genetic algorithms to obtain the optimum value of the global search. Finally, restore data through the "regressive generate" operation to get forecasting data. In order to prove that the SGM-SVM model is superior to other models, we select the battery life data from calce. The presented model is used to predict life of battery and the predicted result was compared with that of grey model and support vector machines．For a more intuitive comparison of the three models, this paper presents root mean square error of this three different models .The results show that the effect of grey support vector machine (SGM-SVM) to predict life is optimal, and the root mean square error is only 3.18%. Keywords: grey forecasting model, trigonometric function, support vector machine, genetic algorithms, root mean square error

Keywords: Grey prediction model, trigonometric functions, support vector machines, genetic algorithms, root mean square error

Procedia PDF Downloads 456

Authors: Yike Lamu, Jieyu Tang, Jialin Wu, Jianyun Huang

Abstract:

With the development of economy and science and technology, the urban heat island effect becomes more and more serious. Taking Beijing city as an example, this paper divides the value of each influence index of heat island intensity and establishes a mathematical model – neural network system based on the fuzzy comprehensive evaluation index of heat island effect. After data preprocessing, the algorithm of weight of each factor affecting heat island effect is generated, and the data of sex indexes affecting heat island intensity of Shenyang City and Shanghai City, Beijing, and Hangzhou City are input, and the result is automatically output by the neural network system. It is of practical significance to show the intensity of heat island effect by visual method, which is simple, intuitive and can be dynamically monitored.

Keywords: heat island effect, neural network, comprehensive evaluation, visualization

Procedia PDF Downloads 132

Authors: Zeki Oralhan, Mahmut Tokmakçı

Abstract:

SSVEP based brain computer interface (BCI) systems have been preferred, because of high information transfer rate (ITR) and practical use. ITR is the parameter of BCI overall performance. For high ITR value, one of specification BCI system is that has high accuracy. In this study, we investigated to recognize SSVEP with shorter time and lower error rate. In the experiment, there were 8 flickers on light crystal display (LCD). Participants gazed to flicker which had 12 Hz frequency and 50% duty cycle ratio on the LCD during 10 seconds. During the experiment, EEG signals were acquired via EEG device. The EEG data was filtered in preprocessing session. After that Canonical Correlation Analysis (CCA), Multiset CCA (MsetCCA), phase constrained CCA (PCCA), and Multiway CCA (MwayCCA) methods were applied on data. The highest average accuracy value was reached when MsetCCA was applied.

Keywords: brain computer interface, canonical correlation analysis, human computer interaction, SSVEP

Procedia PDF Downloads 261

Authors: Ayten Ekin Meşe, Selahattin Şentürk, Melike Duvanoğlu

Abstract:

Petroleum refineries are a highly complex process industry with continuous production and high operating costs. Physical separation of crude oil starts with the crude oil distillation unit, continues with various conversion and purification units, and passes through many stages until obtaining the final product. To meet the desired product specification, process parameters are strictly followed. To be able to ensure the quality of distillates, routine analyses are performed in quality control laboratories based on appropriate international standards such as American Society for Testing and Materials (ASTM) standard methods and European Standard (EN) methods. The cut point of distillates in the crude distillation unit is very crucial for the efficiency of the upcoming processes. In order to maximize the process efficiency, the determination of the quality of distillates should be as fast as possible, reliable, and cost-effective. In this sense, an alternative study was carried out on the crude oil distillation unit that serves the entire refinery process. In this work, studies were conducted with three different crude oil distillates which are Light Straight Run Naphtha (LSRN), Heavy Straight Run Naphtha (HSRN), and Kerosene. These products are named after separation by the number of carbons it contains. LSRN consists of five to six carbon-containing hydrocarbons, HSRN consist of six to ten, and kerosene consists of sixteen to twenty-two carbon-containing hydrocarbons. Physical properties of three different crude distillation unit products (LSRN, HSRN, and Kerosene) were determined using Near-Infrared Spectroscopy with multivariate calibration. The absorbance spectra of the petroleum samples were obtained in the range from 10000 cm⁻¹ to 4000 cm⁻¹, employing a quartz transmittance flow through cell with a 2 mm light path and a resolution of 2 cm⁻¹. A total of 400 samples were collected for each petroleum sample for almost four years. Several different crude oil grades were processed during sample collection times. Extended Multiplicative Signal Correction (EMSC) and Savitzky-Golay (SG) preprocessing techniques were applied to FT-NIR spectra of samples to eliminate baseline shifts and suppress unwanted variation. Two different multivariate calibration approaches (Partial Least Squares Regression, PLS and Genetic Inverse Least Squares, GILS) and an ensemble model were applied to preprocessed FT-NIR spectra. Predictive performance of each multivariate calibration technique and preprocessing techniques were compared, and the best models were chosen according to the reproducibility of ASTM reference methods. This work demonstrates the developed models can be used for routine analysis instead of conventional analytical methods with over 90% accuracy.

Keywords: crude distillation unit, multivariate calibration, near infrared spectroscopy, data preprocessing, refinery

Procedia PDF Downloads 120

Authors: Enver Kucukkulahli, Pakize Erdogmus, Kemal Polat

Abstract:

The 2D image segmentation is a significant process in finding a suitable region in medical images such as MRI, PET, CT etc. In this study, we have focused on 2D MRI images for image segmentation process. We have designed a GUI (graphical user interface) written in MATLABTM for 2D MRI images. In this program, there are two different interfaces including data pre-processing and image clustering or segmentation. In the data pre-processing section, there are median filter, average filter, unsharp mask filter, Wiener filter, and custom filter (a filter that is designed by user in MATLAB). As for the image clustering, there are seven different image segmentations for 2D MR images. These image segmentation algorithms are as follows: PSO (particle swarm optimization), GA (genetic algorithm), Lloyds algorithm, k-means, the combination of Lloyds and k-means, mean shift clustering, and finally BBO (Biogeography Based Optimization). To find the suitable cluster number in 2D MRI, we have designed the histogram based cluster estimation method and then applied to these numbers to image segmentation algorithms to cluster an image automatically. Also, we have selected the best hybrid method for each 2D MR images thanks to this GUI software.

Keywords: image segmentation, clustering, GUI, 2D MRI

Procedia PDF Downloads 373

Authors: Amine Ganibardi, Cherif Arab Ali

Abstract:

This paper addresses the data cleaning issue as a part of web usage data preprocessing within the scope of Web Usage Mining. Weblog data recorded by web servers within log files reflect usage activity, i.e., End-users’ clicks and underlying user-agents’ hits. As Web Usage Mining is interested in End-users’ behavior, user-agents’ hits are referred to as noise to be cleaned-off before mining. Filtering hits from clicks is not trivial for two reasons, i.e., a server records requests interlaced in sequential order regardless of their source or type, website resources may be set up as requestable interchangeably by end-users and user-agents. The current methods are content-centric based on filtering heuristics of relevant/irrelevant items in terms of some cleaning attributes, i.e., website’s resources filetype extensions, website’s resources pointed by hyperlinks/URIs, http methods, user-agents, etc. These methods need exhaustive extra-weblog data and prior knowledge on the relevant and/or irrelevant items to be assumed as clicks or hits within the filtering heuristics. Such methods are not appropriate for dynamic/responsive Web for three reasons, i.e., resources may be set up to as clickable by end-users regardless of their type, website’s resources are indexed by frame names without filetype extensions, web contents are generated and cancelled differently from an end-user to another. In order to overcome these constraints, a clustering-based cleaning method centered on the logging structure is proposed. This method focuses on the statistical properties of the logging structure at the requested and referring resources attributes levels. It is insensitive to logging content and does not need extra-weblog data. The used statistical property takes on the structure of the generated logging feature by webpage requests in terms of clicks and hits. Since a webpage consists of its single URI and several components, these feature results in a single click to multiple hits ratio in terms of the requested and referring resources. Thus, the clustering-based method is meant to identify two clusters based on the application of the appropriate distance to the frequency matrix of the requested and referring resources levels. As the ratio clicks to hits is single to multiple, the clicks’ cluster is the smallest one in requests number. Hierarchical Agglomerative Clustering based on a pairwise distance (Gower) and average linkage has been applied to four logfiles of dynamic/responsive websites whose click to hits ratio range from 1/2 to 1/15. The optimal clustering set on the basis of average linkage and maximum inter-cluster inertia results always in two clusters. The evaluation of the smallest cluster referred to as clicks cluster under the terms of confusion matrix indicators results in 97% of true positive rate. The content-centric cleaning methods, i.e., conventional and advanced cleaning, resulted in a lower rate 91%. Thus, the proposed clustering-based cleaning outperforms the content-centric methods within dynamic and responsive web design without the need of any extra-weblog. Such an improvement in cleaning quality is likely to refine dependent analysis.

Keywords: clustering approach, data cleaning, data preprocessing, weblog data, web usage data

Procedia PDF Downloads 167

Authors: Farnaz Farhangi

Abstract:

Many people in New Zealand believe that the access to water is inexhaustible, and it comes from a history of virtually unrestricted access to it. For the region like Hamilton which is one of New Zealand’s fastest growing cities, it is crucial for policy makers to know about the future water consumption and implementation of rules and regulation such as universal water metering. Hamilton residents use water freely and they do not have any idea about how much water they use. Hence, one of proposed objectives of this research is focusing on forecasting water consumption using different methods. Residential water consumption time series exhibits seasonal and trend variations. Seasonality is the pattern caused by repeating events such as weather conditions in summer and winter, public holidays, etc. The problem with this seasonal fluctuation is that, it dominates other time series components and makes difficulties in determining other variations (such as educational campaign’s effect, regulation, etc.) in time series. Apart from seasonality, a stochastic trend is also combined with seasonality and makes different effects on results of forecasting. According to the forecasting literature, preprocessing (de-trending and de-seasonalization) is essential to have more performed forecasting results, while some other researchers mention that seasonally non-adjusted data should be used. Hence, I answer the question that is pre-processing essential? A wide range of forecasting methods exists with different pros and cons. In this research, I apply double seasonal ARIMA and Artificial Neural Network (ANN), considering diverse elements such as seasonality and calendar effects (public and school holidays) and combine their results to find the best predicted values. My hypothesis is the examination the results of combined method (hybrid model) and individual methods and comparing the accuracy and robustness. In order to use ARIMA, the data should be stationary. Also, ANN has successful forecasting applications in terms of forecasting seasonal and trend time series. Using a hybrid model is a way to improve the accuracy of the methods. Due to the fact that water demand is dominated by different seasonality, in order to find their sensitivity to weather conditions or calendar effects or other seasonal patterns, I combine different methods. The advantage of this combination is reduction of errors by averaging of each individual model. It is also useful when we are not sure about the accuracy of each forecasting model and it can ease the problem of model selection. Using daily residential water consumption data from January 2000 to July 2015 in Hamilton, I indicate how prediction by different methods varies. ANN has more accurate forecasting results than other method and preprocessing is essential when we use seasonal time series. Using hybrid model reduces forecasting average errors and increases the performance.

Keywords: artificial neural network (ANN), double seasonal ARIMA, forecasting, hybrid model

Procedia PDF Downloads 334

Authors: Samina Khalid, Shamila Nasreen

Abstract:

Dimensionality reduction as a preprocessing step to machine learning is effective in removing irrelevant and redundant data, increasing learning accuracy, and improving result comprehensibility. However, the recent increase of dimensionality of data poses a severe challenge to many existing feature selection and feature extraction methods with respect to efficiency and effectiveness. In the field of machine learning and pattern recognition, dimensionality reduction is important area, where many approaches have been proposed. In this paper, some widely used feature selection and feature extraction techniques have analyzed with the purpose of how effectively these techniques can be used to achieve high performance of learning algorithms that ultimately improves predictive accuracy of classifier. An endeavor to analyze dimensionality reduction techniques briefly with the purpose to investigate strengths and weaknesses of some widely used dimensionality reduction methods is presented.

Keywords: age related macular degeneration, feature selection feature subset selection feature extraction/transformation, FSA’s, relief, correlation based method, PCA, ICA

Procedia PDF Downloads 490

Authors: Budoor Al Abid

Abstract:

Educational systems are increasingly provided as open online services, providing guidance and support for individual learners. To adapt the learning systems, a proper evaluation must be made. This paper builds the evaluation model Fuzzy C Means Adaptive System (FCMAS) based on data mining techniques to assess the difficulty of the questions. The following steps are implemented; first using a dataset from an online international learning system called (slepemapy.cz) the dataset contains over 1300000 records with 9 features for students, questions and answers information with feedback evaluation. Next, a normalization process as preprocessing step was applied. Then FCM clustering algorithms are used to adaptive the difficulty of the questions. The result is three cluster labeled data depending on the higher Wight (easy, Intermediate, difficult). The FCM algorithm gives a label to all the questions one by one. Then Random Forest (RF) Classifier model is constructed on the clustered dataset uses 70% of the dataset for training and 30% for testing; the result of the model is a 99.9% accuracy rate. This approach improves the Adaptive E-learning system because it depends on the student behavior and gives accurate results in the evaluation process more than the evaluation system that depends on feedback only.

Keywords: machine learning, adaptive, fuzzy logic, data mining

Procedia PDF Downloads 192

Authors: Mawloud Mosbah, Bachir Boucheham

Abstract:

Color Histogram is considered as the oldest method used by CBIR systems for indexing images. In turn, the global histograms do not include the spatial information; this is why the other techniques coming later have attempted to encounter this limitation by involving the segmentation task as a preprocessing step. The weak segmentation is employed by the local histograms while other methods as CCV (Color Coherent Vector) are based on strong segmentation. The indexation based on local histograms consists of splitting the image into N overlapping blocks or sub-regions, and then the histogram of each block is computed. The dissimilarity between two images is reduced, as consequence, to compute the distance between the N local histograms of the both images resulting then in N*N values; generally, the lowest value is taken into account to rank images, that means that the lowest value is that which helps to designate which sub-region utilized to index images of the collection being asked. In this paper, we make under light the local histogram indexation method in the hope to compare the results obtained against those given by the global histogram. We address also another noteworthy issue when Relying on local histograms namely which value, among N*N values, to trust on when comparing images, in other words, which sub-region among the N*N sub-regions on which we base to index images. Based on the results achieved here, it seems that relying on the local histograms, which needs to pose an extra overhead on the system by involving another preprocessing step naming segmentation, does not necessary mean that it produces better results. In addition to that, we have proposed here some ideas to select the local histogram on which we rely on to encode the image rather than relying on the local histogram having lowest distance with the query histograms.

Keywords: CBIR, color global histogram, color local histogram, weak segmentation, Euclidean distance

Procedia PDF Downloads 356

Authors: Rigas Kotsakis, Nikolaos Tsigilis, Vasilis Grammatikopoulos, Evridiki Zachopoulou

Abstract:

During the last decade, Greece has faced an unprecedented financial crisis, affecting various aspects and functionalities of Higher Education. Besides the restricted funding of academic institutions, the students and their families encountered economical difficulties that undoubtedly influenced the effective completion of their studies. In this context, a fairly large number of students in Alexander campus of International Hellenic University (IHU) delay, interrupt, or even abandon their studies, especially when they come from low-income families, belong to sensitive social or special needs groups, they have different cultural origins, etc. For this reason, a European project, named “Umbrella”, was initiated aiming at providing the necessary psychological support and counseling, especially to disadvantaged students, towards the completion of their studies. To this end, a network of various academic members (academic staff and students) from IHU, namely iMentor, were implicated in different roles. Specifically, experienced academic staff trained students to serve as intermediate links for the integration and educational support of students that fall into the aforementioned sensitive social groups and face problems for the completion of their studies. The main idea of the project is held upon its person-centered character, which facilitates direct student-to-student communication without the intervention of the teaching staff. The backbone of the iMentors network are senior students that face no problem in their academic life and volunteered for this project. It should be noted that there is a provision from the Umbrella structure for substantial and ethical rewards for their engagement. In this context, a well-defined, stringent methodology was implemented for the evaluation of the extent of the problem in IHU and the detection of the profile of the “candidate” disadvantaged students. The first phase included two steps, (a) data collection and (b) data cleansing/ preprocessing. The first step involved the data collection process from the Secretary Services of all Schools in IHU, from 1980 to 2019, which resulted in 96.418 records. The data set included the School name, the semester of studies, a student enrolling criteria, the nationality, the graduation year or the current, up-to-date academic state (still studying, delayed, dropped off, etc.). The second step of the employed methodology involved the data cleansing/preprocessing because of the existence of “noisy” data, missing and erroneous values, etc. Furthermore, several assumptions and grouping actions were imposed to achieve data homogeneity and an easy-to-interpret subsequent statistical analysis. Specifically, the duration of 40 years recording was limited to the last 15 years (2004-2019). In 2004 the Greek Technological Institutions were evolved into Higher Education Universities, leading into a stable and unified frame of graduate studies. In addition, the data concerning active students were excluded from the analysis since the initial processing effort was focused on the detection of factors/variables that differentiated graduate and deleted students. The final working dataset included 21.432 records with only two categories of students, those that have a degree and those who abandoned their studies. Findings of the first phase are presented across faculties and further discussed.

Keywords: higher education, students support, economic crisis, mentoring

Procedia PDF Downloads 109

Authors: Ermias A. Tegegn, Million Meshesha

Abstract:

Human Immunodeficiency Virus and Acquired Immunodeficiency Syndrome (HIV/AIDS) is a major cause of death for most African countries. Ethiopia is one of the seriously affected countries in sub Saharan Africa. Previously in Ethiopia, having HIV/AIDS was almost equivalent to a death sentence. With the introduction of Antiretroviral Therapy (ART), HIV/AIDS has become chronic, but manageable disease. The study focused on a data mining technique to predict future living status of HIV/AIDS patients at the time of drug regimen change when the patients become toxic to the currently taking ART drug combination. The data is taken from University of Gondar Hospital ART program database. Hybrid methodology is followed to explore the application of data mining on ART program dataset. Data cleaning, handling missing values and data transformation were used for preprocessing the data. WEKA 3.7.9 data mining tools, classification algorithms, and expertise are utilized as means to address the research problem. By using four different classification algorithms, (i.e., J48 Classifier, PART rule induction, Naïve Bayes and Neural network) and by adjusting their parameters thirty-two models were built on the pre-processed University of Gondar ART program dataset. The performances of the models were evaluated using the standard metrics of accuracy, precision, recall, and F-measure. The most effective model to predict the status of HIV patients with drug regimen substitution is pruned J48 decision tree with a classification accuracy of 98.01%. This study extracts interesting attributes such as Ever taking Cotrim, Ever taking TbRx, CD4 count, Age, Weight, and Gender so as to predict the status of drug regimen substitution. The outcome of this study can be used as an assistant tool for the clinician to help them make more appropriate drug regimen substitution. Future research directions are forwarded to come up with an applicable system in the area of the study.

Keywords: HIV drug regimen, data mining, hybrid methodology, predictive model

Procedia PDF Downloads 140

Authors: Thota Sai Prakash, B. Yaswanth, Jhade Bhuvaneswar, Marreddy Divakar Reddy, Shyam Ji Gupta

Abstract:

Across the globe, there are a lot of chronic diseases, and heart disease stands out as one of the most perilous. Sadly, many lives are lost to this condition, even though early intervention could prevent such tragedies. However, identifying heart disease in its initial stages is not easy. To address this challenge, we propose an automated system aimed at predicting the presence of heart disease using advanced techniques. By doing so, we hope to empower individuals with the knowledge needed to take proactive measures against this potentially fatal illness. Our approach towards this problem involves meticulous data preprocessing and the development of predictive models utilizing classification algorithms such as Support Vector Machines (SVM), Decision Tree, and Random Forest. We assess the efficiency of every model based on metrics like accuracy, ensuring that we select the most reliable option. Additionally, we conduct thorough data analysis to reveal the importance of different attributes. Among the models considered, Random Forest emerges as the standout performer with an accuracy rate of 96.04% in our study.

Keywords: support vector machines, decision tree, random forest

Procedia PDF Downloads 33

Authors: John Paul Q. Tomas, Mark Wilson L. de los Reyes, Kirsten Joyce P. Vasquez

Abstract:

The most frequent type of fracture is a wrist fracture, which often makes it difficult for medical professionals to find and locate. In this study, fractures in wrist x-ray pictures were located and identified using deep learning and computer vision. The researchers used image filtering, masking, morphological operations, and data augmentation for the image preprocessing and trained the RetinaNet and Faster R-CNN models with ResNet50 backbones and Adam optimizers separately for each image filtering technique and projection. The RetinaNet model with Anisotropic Diffusion Smoothing filter trained with 50 epochs has obtained the greatest accuracy of 99.14%, precision of 100%, sensitivity/recall of 98.41%, specificity of 100%, and an IoU score of 56.44% for the Posteroanterior projection utilizing augmented data. For the Lateral projection using augmented data, the RetinaNet model with an Anisotropic Diffusion filter trained with 50 epochs has produced the highest accuracy of 98.40%, precision of 98.36%, sensitivity/recall of 98.36%, specificity of 98.43%, and an IoU score of 58.69%. When comparing the test results of the different individual projections, models, and image filtering techniques, the Anisotropic Diffusion filter trained with 50 epochs has produced the best classification and regression scores for both projections.

Keywords: Artificial Intelligence, Computer Vision, Wrist Fracture, Deep Learning

Procedia PDF Downloads 69

Authors: Christian Neunzig, Simon Fahle, Jürgen Schulz, Matthias Möller, Bernd Kuhlenkötter

Abstract:

Manufacturing companies are facing global competition and enormous cost pressure. The use of machine learning applications can help reduce production costs and create added value. Predictive quality enables the securing of product quality through data-supported predictions using machine learning models as a basis for decisions on test results. Furthermore, machine learning methods are able to process large amounts of data, deal with unfavourable row-column ratios and detect dependencies between the covariates and the given target as well as assess the multidimensional influence of all input variables on the target. Real production data are often subject to highly fluctuating boundary conditions and unbalanced data sets. Changes in production data manifest themselves in trends, systematic shifts, and seasonal effects. Thus, Machine learning applications require intensive pre-processing and feature selection. Data preprocessing includes rule-based data cleaning, the application of dimensionality reduction techniques, and the identification of comparable data subsets. Within the used real data set of Bosch hydraulic valves, the comparability of the same production conditions in the production of hydraulic valves within certain time periods can be identified by applying the concept drift method. Furthermore, a classification model is developed to evaluate the feature importance in different subsets within the identified time periods. By selecting comparable and stable features, the number of features used can be significantly reduced without a strong decrease in predictive power. The use of cross-process production data along the value chain of hydraulic valves is a promising approach to predict the quality characteristics of workpieces. In this research, the ada boosting classifier is used to predict the leakage of hydraulic valves based on geometric gauge blocks from machining, mating data from the assembly, and hydraulic measurement data from end-of-line testing. In addition, the most suitable methods are selected and accurate quality predictions are achieved.

Keywords: classification, achine learning, predictive quality, feature selection

Procedia PDF Downloads 158