Search results for: parallel data mining

Authors: Flora Babongo, Valerie Chavez

Abstract:

Inferring causality from observational data is one of the fundamental subjects, especially in quantitative finance. So far most of the papers analyze additive noise models with either linearity, nonlinearity or Gaussian noise. We fill in the gap by providing a nonlinear and non-gaussian causal multiplicative noise model that aims to distinguish the cause from the effect using a two steps method based on Bayesian additive models for location, scale and shape (BAMLSS) and on causal additive models (CAM). We have tested our method on simulated and real data and we reached an accuracy of 0.86 on average. As real data, we considered the causality between financial indices such as S&P 500, Nasdaq, CAC 40 and Nikkei, and companies' log-returns. Our results can be useful in inferring causality when the data is heteroskedastic or non-injective.

Keywords: causal inference, DAGs, BAMLSS, financial index

Procedia PDF Downloads 139

Authors: Madiha Liaqat, Rehan Ahmed Khan, Shahid Kamal

Abstract:

Multiple imputation with delta adjustment is a versatile and transparent technique for addressing univariate missing data in the presence of various missing mechanisms. This approach allows for the exploration of sensitivity to the missing-at-random (MAR) assumption. In this review, we outline the delta-adjustment procedure and illustrate its application for assessing the sensitivity to deviations from the MAR assumption. By examining diverse missingness scenarios and conducting sensitivity analyses, we gain valuable insights into the implications of missing data on our analyses, enhancing the reliability of our study's conclusions. In our study, we focused on assessing logPSA, a continuous biomarker in incomplete prostate cancer data, to examine the robustness of conclusions against plausible departures from the MAR assumption. We introduced several approaches for conducting sensitivity analyses, illustrating their application within the pattern mixture model (PMM) under the delta adjustment framework. This proposed approach effectively handles missing data, particularly loss to follow-up.

Keywords: loss to follow-up, incomplete response, multiple imputation, sensitivity analysis, prostate cancer

Procedia PDF Downloads 71

Authors: Guru Prakash, Revanth Dugalam

Abstract:

A road’s condition often deteriorates due to harsh loading such as overload due to trucks, and severe environmental conditions such as heavy rain, snow load, and cyclic loading. In absence of proper maintenance planning, this results in potholes, wide cracks, bumps, and increased roughness of roads. In this paper, a data-driven model will be developed to detect these damages using vibration and image signals. The key idea of the proposed methodology is that the road anomaly manifests in these signals, which can be detected by training a machine learning algorithm. The use of various machine learning techniques such as the support vector machine and Radom Forest method will be investigated. The proposed model will first be trained and tested with artificially simulated data, and the model architecture will be finalized by comparing the accuracies of various models. Once a model is fixed, the field study will be performed, and data will be collected. The field data will be used to validate the proposed model and to predict the future road’s health condition. The proposed will help to automate the road condition monitoring process, repair cost estimation, and maintenance planning process.

Keywords: SVM, data-driven, road health monitoring, pot-hole

Procedia PDF Downloads 68

Authors: U. Borasi, Amit Kr. Jain, Rakesh, Piyush Jain

Abstract:

Data collection, data preparation, model training, model evaluation, and deployment are all processes in a typical machine learning workflow. Training data needs to be gathered and organised. This often entails collecting a sizable dataset and cleaning it to remove or correct any inaccurate or missing information. Preparing the data for use in the machine learning model requires pre-processing it after it has been acquired. This often entails actions like scaling or normalising the data, handling outliers, selecting appropriate features, reducing dimensionality, etc. This pre-processed data is then used to train a model on some machine learning algorithm. After the model has been trained, it needs to be assessed by determining metrics like accuracy, precision, and recall, utilising a test dataset. Every time a new model is built, both data pre-processing and model training—two crucial processes in the Machine learning (ML) workflow—must be carried out. Thus, there are various Machine Learning algorithms that can be employed for every single approach to data pre-processing, generating a large set of combinations to choose from. Example: for every method to handle missing values (dropping records, replacing with mean, etc.), for every scaling technique, and for every combination of features selected, a different algorithm can be used. As a result, in order to get the optimum outcomes, these tasks are frequently repeated in different combinations. This paper suggests a simple architecture for organizing this largely produced “combination set of pre-processing steps and algorithms” into an automated workflow which simplifies the task of carrying out all possibilities.

Keywords: machine learning, automation, AUTOML, architecture, operator pool, configuration, scheduler

Procedia PDF Downloads 39

Authors: Abdelkader Soumaya, Aymen Arfaoui, Noureddine Ben Ayed, Ali Kadri

Abstract:

The Chotts belt represents the southernmost folded structure in the Tunisian Atlas domain. It is dominated by inherited deep extensional E-W trending fault zones, which are reactivated as strike-slip faults during the Cenozoic compression. By examining the geological maps at different scales and based on the fieldwork data, we propose new structural interpretations for the geometries and fault kinematics in the Chotts chain. A set of ENE-WSW right-lateral en echelon folds, with curved shapes and steeply inclined southern limbs, is visible in the map view of this belt. These asymmetric tight anticlines are affected by E-W trending fault segments linked by local bends and stepovers. The revealed kinematic indicators along one of these E-W striated faults (Tafferna segment), such as breccias and gently inclined slickenlines (N094, 80N, 15°W pitch angles), show direct evidence of dextral strike-slip movement. The calculated stress tensors from corresponding faults slip data reveal an overall strike-slip tectonic regime with reverse component and NW-trending sub-horizontal σ1 axis ranking between N130 to N150. From west to east, we distinguished several types of structures along the segmented dextral fault system of the Chotts Range. The NE-SW striking fold-thrust belt (~25 km-long) between two continuously linked E-W fault segments (NW of Tozeur town) has been suggested as a local restraining bend. The central part of the Chotts chain is occupied by the ENE-striking Ksar Asker anticlines (Taferna, Torrich, and Sif Laham), which are truncated by a set of E-W strike-slip fault segments. Further east, the fault segments of Hachichina and Sif Laham connected across the NW-verging asymmetric fold-thrust system of Bir Oum Ali, which can be interpreted as a left-stepping contractional bend (~20 km-long). The oriental part of the Chotts belt corresponds to an array of subparallel E-W oriented fault segments (i.e., Beidha, Bouloufa, El Haidoudi-Zemlet El Beidha) with similar lengths (around 10 km). Each of these individual separated segments is associated with curved ENE-trending en echelon right-stepping anticlines. These folds are affected by a set of conjugate R and R′ shear-type faults indicating a dextral strike-lip motion. In addition, the relay zones between these E-W overstepping fault segments define local releasing stepovers dominated by NW-SE subsidiary faults. Finally, the Chotts chain provides well-exposed examples of strike-slip tectonics along E-W distributed fault segments. Each fault zone shows a typical strike-slip architecture, including parallel fault segments connecting via local stepovers or bends. Our new structural interpretations for this region reveal a great influence of the E-W deep fault segments on regional tectonic deformations and stress field during the Cenozoic shortening.

Keywords: chotts belt, tunisian atlas, strike-slip fault, stepovers, fault segments

Procedia PDF Downloads 56

Authors: Shih-Ting Yang, Chih-Yun Lin, Ming-Yu Li, Jhong-Ting Syue, Wei-Ming Huang

Abstract:

Global positioning system (GPS) has become increasing precise in recent years, and the location-based service (LBS) has developed rapidly. Take the example of finding a parking lot (such as Parking apps). The location-based service can offer immediate information about a nearby parking lot, including the information about remaining parking spaces. However, it cannot provide expected search results according to the requirement situations of users. For that reason, this paper develops a “Location-based Search Approach according to Users’ Application Scenario” according to the location-based search and demand determination to help users obtain the information consistent with their requirements. The “Location-based Search Approach based on Users’ Application Scenario” of this paper consists of one mechanism and three kernel modules. First, in the Information Pre-processing Mechanism (IPM), this paper uses the cosine theorem to categorize the locations of users. Then, in the Information Category Evaluation Module (ICEM), the kNN (k-Nearest Neighbor) is employed to classify the browsing records of users. After that, in the Information Volume Level Determination Module (IVLDM), this paper makes a comparison between the number of users’ clicking the information at different locations and the average number of users’ clicking the information at a specific location, so as to evaluate the urgency of demand; then, the two-dimensional space is used to estimate the application situations of users. For the last step, in the Location-based Search Module (LBSM), this paper compares all search results and the average number of characters of the search results, categorizes the search results with the Manhattan Distance, and selects the results according to the application scenario of users. Additionally, this paper develops a Web-based system according to the methodology to demonstrate practical application of this paper. The application scenario-based estimate and the location-based search are used to evaluate the type and abundance of the information expected by the public at specific location, so that information demanders can obtain the information consistent with their application situations at specific location.

Keywords: data mining, knowledge management, location-based service, user application scenario

Procedia PDF Downloads 109

Authors: Sara Mosallanejad, Bogdan Z. Dlugogorski, Jeff Gore, Mohammednoor Altarawneh

Abstract:

Ammonium nitrate (NH­₄NO₃, AN) is commonly used as the main component of AN emulsion and fuel oil (ANFO) explosives, that use extensively in civilian and mining operations for underground development and tunneling applications. The emulsion formulation and wettability of AN prills, which affect the physical stability and detonation of ANFO, highly depend on the surface tension, density, viscosity of the used liquid. Therefore, for engineering applications of this material, the determination of density and surface tension of concentrated aqueous solutions of AN is essential. The molecular dynamics (MD) simulation method have been used to investigate the density and the surface tension of high concentrated ammonium nitrate solutions; up to its solubility limit in water. Non-polarisable models for water and ions have carried out the simulations, and the electronic continuum correction model (ECC) uses a scaling of the charges of the ions to apply the polarisation implicitly into the non-polarisable model. The results of calculated density and the surface tension of the solutions have been compared to available experimental values. Our MD simulations show that the non-polarisable model with full-charge ions overestimates the experimental results while the reduce-charge model for the ions fits very well with the experimental data. Ions in the solutions show repulsion from the interface using the non-polarisable force fields. However, when charges of the ions in the original model are scaled in line with the scaling factor of the ECC model, the ions create a double ionic layer near the interface by the migration of anions toward the interface while cations stay in the bulk of the solutions. Similar ions orientations near the interface were observed when polarisable models were used in simulations. In conclusion, applying the ECC model to the non-polarisable force field yields the density and surface tension of the AN solutions with high accuracy in comparison to the experimental measurements.

Keywords: ammonium nitrate, electronic continuum correction, non-polarisable force field, surface tension

Procedia PDF Downloads 208

Authors: Rifky Cahya Putra

Abstract:

At this time of corona makes some countries affected difficult. As a result, many traders or companies are difficult to work in this pandemic era. So human activities in some fields must implement a new lifestyle or known as new normal. The transition from the one activity to another certainly requires high adaptation. So that almost in all sectors experience the impact of this phase, on of which is the wedding organizer. This research aims to find out what strategies are used so that the company can run in this pandemic. Techniques in data collection in the form interview to the owner of the wedding organizer and his team. Data analysis qualitative descriptive use interactive model analysis consisting of three main things, namely data reduction, data presentaion, and conclusion. For the result of the interview, the conclusion is that there are three strategies consisting of social media, sponsorship, and promotion.

Keywords: strategy, wedding organizer, pandemic, indonesia

Procedia PDF Downloads 120

Authors: Azadeh Askarinejad, Parham Hayati, Reza Parchami, Parisa Hayati

Abstract:

One of the most important industries and building operations causing carbon dioxide emission is the cement and concrete related industries so that cement production (including direct fuel for mining and transporting raw material) consumes approximately 6 million Btus per metric-ton, and releases about 1 metric-ton of CO2. Reducing the consumption of cement with simultaneous utilizing waste materials as cement replacement is preferred for reasons of environmental protection. Blended cements consist of different supplementary cementitious materials (SCM), such as fly ash, silica fume, Ground Granulated Blast Furnace Slag (GGBFS), limestone, natural pozzolans, etc. these materials should be chemically activated to show effective cementitious properties. The present review article reports three different methods of nanofabrication that were used for activation of two types of SCMs.

Keywords: nanofabrication, cement replacement materials, activation, concrete

Procedia PDF Downloads 597

Authors: Zhou Mo, Dennis Chow

Abstract:

In the dynamic positioning system (DPS) for vessels, the reliable information transmission between each note basically relies on the wireless protocols. From the perspective of cluster-based routing pro-tocols for wireless sensor networks, the data fusion technology based on the sleep scheduling mechanism and remaining energy in network layer is proposed, which applies the sleep scheduling mechanism to the routing protocols, considering the remaining energy of node and location information when selecting cluster-head. The problem of uneven distribution of nodes in each cluster is solved by the Equilibrium. At the same time, Classified Forwarding Mechanism as well as Redelivery Policy strategy is adopted to avoid congestion in the transmission of huge amount of data, reduce the delay in data delivery and enhance the real-time response. In this paper, a simulation test is conducted to improve the routing protocols, which turns out to reduce the energy consumption of nodes and increase the efficiency of data delivery.

Keywords: DPS for vessel, wireless sensor network, data fusion, routing protocols

Procedia PDF Downloads 433

Authors: Mehmet Yavuz Huseyinca

Abstract:

The clay beds are located at the south-southeast of Doğanhisar and northwest of Konya in the Central Anatolia. In the scope of preliminary study, three types of samples were investigated including basement phyllite (Bp) overlain by the clay beds, weathered phyllite (Wp) and Doğanhisar clay (Dc). The Chemical Index of Alteration (CIA) values of Dc range from 81 to 88 with an average of 85. This value is higher than that of Post Archean Australian Shale (PAAS) and defines very intense chemical weathering in the source-area. On the other hand, the A-CN-K diagram indicates that Bp underwent high degree post-depositional K-metasomatism. The average reconstructed CIA value of the Bp prior to the K-metasomatism is mainly 81 which overlaps the CIA values of the Wp (83) and Dc (85). Similar CIA values indicate parallel weathering trends. Also, extrapolation of the samples back to the plagioclase-alkali feldspar line in the A-CN-K diagram suggests an identical provenance close to granite in composition. Hereby the weathering background of Dc includes two steps. First one is intense weathering process of a granitic source to Bp with post-depositional K-metasomatism and the latter is progressively weathering of Bp to premetasomatised conditions (formation of Wp) ending with Dc deposition.

Keywords: clay beds, Doganhisar, provenance, weathering

Procedia PDF Downloads 296

Authors: Abdulhakeem Yusuf, Yomi Monday Aiyesimi, Mohammed Jiya

Abstract:

The failure in an ordinary heat transfer fluid to meet up with today’s industrial cooling rate has resulted in the development of high thermal conductivity fluid which nanofluids belongs. In this work, the problem of unsteady one dimensional laminar flow of an incompressible fluid within a parallel wall is considered with one wall assumed to be wavy. The model is presented in its rectangular coordinate system and incorporates the effects of thermophoresis and Brownian motion. The local similarity solutions were also obtained which depends on Soret number, Dufour number, Biot number, Lewis number, and heat generation parameter. The analytical solution is obtained in a closed form via the Adomian decomposition method. It was found that the method has a good agreement with the numerical method, and it is also established that the heat generation parameter has to be kept low so that heat energy are easily evacuated from the system.

Keywords: Adomian decomposition method, Biot number, Dufour number, nanofluid

Procedia PDF Downloads 313

Authors: Ahmed Fradi

Abstract:

In a digital world in perpetual evolution and acceleration, data more and more voluminous, rich and varied, the new software solutions emerged with the Big Data phenomenon offer new opportunities to the company enabling it not only to optimize its business and to evolve its production model, but also to reorganize itself to increase competitiveness and to identify new strategic axes. Design and manufacturing industrial companies, like the others, face these challenges, data represent a major asset, provided that they know how to capture, refine, combine and analyze them. The objective of our paper is to propose a solution allowing geometric and topological information extraction from 3D CAD model (precisely STEP files) databases, with specific algorithm based on the programming paradigm MapReduce. Our proposal is the first step of our future approach to 3D CAD object retrieval.

Keywords: Big Data, MapReduce, 3D object retrieval, CAD, STEP format

Procedia PDF Downloads 530

Authors: Elena B. Cherepetskaya, Alexander A.Karabutov, Vladimir A. Makarov, Elena A. Mironova, Ivan A. Shibaev

Abstract:

In our study, the object of laser ultrasonic testing was plane-parallel plate of shungit (length 41 mm, width 31 mm, height 15 mm, medium exchange density 2247 kg/m3). We used laser-ultrasonic defectoscope with wideband opto-acoustic transducer in our investigation of the velocities of longitudinal and shear elastic ultrasound waves. The duration of arising elastic pulses was less than 100 ns. Under known material thickness, the values of the velocities were determined by the time delay of the pulses reflected from the bottom surface of the sample with respect to reference pulses. The accuracy of measurement was 0.3% in the case of longitudinal wave velocity and 0.5% in the case of shear wave velocity (scanning pitch along the surface was 2 mm). On the base of found velocities of elastic waves, local elastic moduli of shungit (Young modulus, shear modulus and Poisson's ratio) were uniquely determined.

Keywords: laser ultrasonic testing , local elastic moduli, shear wave velocity, shungit

Procedia PDF Downloads 289

Authors: Duru Helin Ozaner

Abstract:

Due to the doctrine of equivalent, the words in the claims' sentences are insufficient for the protection area provided by the patent registration. While this situation widens the boundaries of the protection area, it also obscures the boundaries of the protected area of patents. In addition, it creates distrust for third parties. Therefore, the doctrine of equivalent aims to establish a balance between the rights of patent owners and the legal security of third parties. The current legal system of Turkey has been tried to be created as a parallel judicial system to the widely applied regulations. Therefore, the regulations regarding the protection provided by patents in the current Turkish legal system are similar to many countries. However, infringement through equivalent is common by third parties. This study, it is aimed to explain that the protection provided by the patent is not only limited to the words of the claims but also the wide-ranging protection provided by the claims for the doctrine of equivalence. This study is important to determine the limits of the protection provided by the patent right holder and to indicate the importance of the equivalent elements of the protection granted to the patent right holder.

Keywords: patent, infringement, intellectual property, the doctrine of equivalent

Procedia PDF Downloads 199

Authors: R. K. Pateriya, Jyoti Bharti

Abstract:

This paper presents an approach for data hiding methods which provides a secret communication between sender and receiver. The data is hidden in gray-scale images and the boundary of gray-scale image is used to store the mapping information. In this an approach data is in ASCII format and the mapping is in between ASCII value of hidden message and pixel value of cover image, since pixel value of an image as well as ASCII value is in range of 0 to 255 and this mapping information is occupying only 1 bit per character of hidden message as compared to 8 bit per character thus maintaining good quality of stego image.

Keywords: ASCII value, cover image, PSNR, pixel value, stego image, secret message

Procedia PDF Downloads 399

Authors: Vitor M. Vilas Boas, Cleison D. Silva, Gustavo S. Mafra, Alexandre Trofino Neto

Abstract:

Motor imagery (MI) based brain-computer interfaces (BCI) uses event-related (de)synchronization (ERS/ ERD), typically recorded using electroencephalography (EEG), to translate brain electrical activity into control commands. To mitigate undesirable artifacts and noise measurements on EEG signals, methods based on band-pass filters defined by a specific frequency band (i.e., 8 – 30Hz), such as the Infinity Impulse Response (IIR) filters, are typically used. Spatial techniques, such as Common Spatial Patterns (CSP), are also used to estimate the variations of the filtered signal and extract features that define the imagined motion. The CSP effectiveness depends on the subject's discriminative frequency, and approaches based on the decomposition of the band of interest into sub-bands with smaller frequency ranges (SBCSP) have been suggested to EEG signals classification. However, despite providing good results, the SBCSP approach generally increases the computational cost of the filtering step in IM-based BCI systems. This paper proposes the use of the Fast Fourier Transform (FFT) algorithm in the IM-based BCI filtering stage that implements SBCSP. The goal is to apply the FFT algorithm to reduce the computational cost of the processing step of these systems and to make them more efficient without compromising classification accuracy. The proposal is based on the representation of EEG signals in a matrix of coefficients resulting from the frequency decomposition performed by the FFT, which is then submitted to the SBCSP process. The structure of the SBCSP contemplates dividing the band of interest, initially defined between 0 and 40Hz, into a set of 33 sub-bands spanning specific frequency bands which are processed in parallel each by a CSP filter and an LDA classifier. A Bayesian meta-classifier is then used to represent the LDA outputs of each sub-band as scores and organize them into a single vector, and then used as a training vector of an SVM global classifier. Initially, the public EEG data set IIa of the BCI Competition IV is used to validate the approach. The first contribution of the proposed method is that, in addition to being more compact, because it has a 68% smaller dimension than the original signal, the resulting FFT matrix maintains the signal information relevant to class discrimination. In addition, the results showed an average reduction of 31.6% in the computational cost in relation to the application of filtering methods based on IIR filters, suggesting FFT efficiency when applied in the filtering step. Finally, the frequency decomposition approach improves the overall system classification rate significantly compared to the commonly used filtering, going from 73.7% using IIR to 84.2% using FFT. The accuracy improvement above 10% and the computational cost reduction denote the potential of FFT in EEG signal filtering applied to the context of IM-based BCI implementing SBCSP. Tests with other data sets are currently being performed to reinforce such conclusions.

Keywords: brain-computer interfaces, fast Fourier transform algorithm, motor imagery, sub-band common spatial patterns

Procedia PDF Downloads 110

Authors: Gunasekaran Raja, Kottilingam Kottursamy, Rajakumar Arul, Ramkumar Jayaraman, Krithika Sairam, Lakshmi Ravi

Abstract:

The synchronization server maintains a dynamically changing cache, which contains the data items which were requested and collected by the mobile node from the server. The order and presence of tuples in the cache changes dynamically according to the frequency of updates performed on the data, by the server and client. To synchronize, the data which has been modified by client and the server at an instant are collected, batched together by the type of modification (insert/ update/ delete), and sorted according to their update frequencies. This ensures that the DCASH (Dynamic Cache Synchronization Algorithm for Heterogeneous Reverse Y synchronizing Mobile Database Systems) gives priority to the frequently accessed data with high usage. The optimal memory management algorithm is proposed to manage data items according to their frequency, theorems were written to show the current mobile data activity is reverse Y in nature and the experiments were tested with 2g and 3g networks for various mobile devices to show the reduced response time and energy consumption.

Keywords: mobile databases, synchronization, cache, response time

Procedia PDF Downloads 382

Authors: Supunmali Ahangama, Danny Chiang Choon Poo

Abstract:

Health analytics (HA) is used in healthcare systems for effective decision-making, management, and planning of healthcare and related activities. However, user resistance, the unique position of medical data content, and structure (including heterogeneous and unstructured data) and impromptu HA projects have held up the progress in HA applications. Notably, the accuracy of outcomes depends on the skills and the domain knowledge of the data analyst working on the healthcare data. The success of HA depends on having a sound process model, effective project management and availability of supporting tools. Thus, to overcome these challenges through an effective process model, we propose an HA process model with features from the rational unified process (RUP) model and agile methodology.

Keywords: agile methodology, health analytics, unified process model, UML

Procedia PDF Downloads 490

Authors: Maximilian Winkens, Matthias Goerke

Abstract:

The state-oriented maintenance enables the preventive intervention before the failure of a component and guarantees avoidance of expensive breakdowns. Because the timing of the maintenance is defined by the component’s state, the remaining service life can be exhausted to the limit. The basic requirement for the state-oriented maintenance is the ability to define the component’s state. New potential for this is offered by gentelligent components. They are developed at the Corporative Research Centre 653 of the German Research Foundation (DFG). Because of their sensory ability they enable the registration of stresses during the component’s use. The data is gathered and evaluated. The methodology developed determines the current state of the gentelligent component based on the gathered data. This article presents this methodology as well as current research. The main focus of the current scientific work is to improve the quality of the state determination based on the life-cycle data analysis. The methodology developed until now evaluates the data of the usage phase and based on it predicts the timing of the gentelligent component’s failure. The real failure timing though, deviate from the predicted one because the effects from the production phase aren’t considered. The goal of the current research is to develop a methodology for state determination which considers both production and usage data.

Keywords: state-oriented maintenance, life-cycle data, gentelligent component, preventive intervention

Procedia PDF Downloads 480

Authors: Rudi Kurniawan Arief

Abstract:

Manufacturing cost and setup time are the hot topics to improve in Metal Stamping industry because material and components price are always rising up while costumer requires to cut down the component price year by year. The Single Minute Exchange of Die (SMED) is one of many methods to reduce waste in stamping industry. The Japanese Quick Die Change (QDC) dies system is one of SMED systems that could reduce both of setup time and manufacturing cost. However, this system is rarely used in stamping industries. This paper will analyze how deep the QDC dies system could reduce setup time and the manufacturing cost. The research is conducted by direct observation, simulating and comparing of QDC dies system with conventional dies system. In this research, we found that the QDC dies system could save up to 35% of manufacturing cost and reduce 70% of setup times. This simulation proved that the QDC die system is effective for cost reduction but must be applied in several parallel production processes.

Keywords: press die, metal stamping, QDC system, single minute exchange die, manufacturing cost saving, SMED

Procedia PDF Downloads 156

Authors: James Ladzekpo

Abstract:

Background: The urgent need to identify new pharmacological targets for diabetes treatment and prevention has been amplified by the disease's extensive impact on individuals and healthcare systems. A deeper insight into the biological underpinnings of diabetes is crucial for the creation of therapeutic strategies aimed at these biological processes. Current predictive models based on genetic variations fall short of accurately forecasting diabetes. Objectives: Our study aims to pinpoint key epigenetic factors that predispose individuals to diabetes. These factors will inform the development of an advanced predictive model that estimates diabetes risk from genetic profiles, utilizing state-of-the-art statistical and data mining methods. Methodology: We have implemented a recursive feature elimination with cross-validation using the support vector machine (SVM) approach for refined feature selection. Building on this, we developed six machine learning models, including logistic regression, k-Nearest Neighbors (k-NN), Naive Bayes, Random Forest, Gradient Boosting, and Multilayer Perceptron Neural Network, to evaluate their performance. Findings: The Gradient Boosting Classifier excelled, achieving a median recall of 92.17% and outstanding metrics such as area under the receiver operating characteristics curve (AUC) with a median of 68%, alongside median accuracy and precision scores of 76%. Through our machine learning analysis, we identified 31 genes significantly associated with diabetes traits, highlighting their potential as biomarkers and targets for diabetes management strategies. Conclusion: Particularly noteworthy were the Gradient Boosting Classifier and Multilayer Perceptron Neural Network, which demonstrated potential in diabetes outcome prediction. We recommend future investigations to incorporate larger cohorts and a wider array of predictive variables to enhance the models' predictive capabilities.

Keywords: diabetes, machine learning, prediction, biomarkers

Procedia PDF Downloads 39

Authors: S. Delgado, C. Cerrada, R. S. Gómez

Abstract:

This research introduces an approach to voxelizing the surfaces of triangular meshes with efficiency and accuracy. Our method leverages parallel equidistant scan-lines and introduces a Gap Detection technique to address the limitations of existing approaches. We present a comprehensive study showcasing the method's effectiveness, scalability, and versatility in different scenarios. Voxelization is a fundamental process in computer graphics and simulations, playing a pivotal role in applications ranging from scientific visualization to virtual reality. Our algorithm focuses on enhancing the voxelization process, especially for complex models and high resolutions. One of the major challenges in voxelization in the Graphics Processing Unit (GPU) is the high cost of discovering the same voxels multiple times. These repeated voxels incur in costly memory operations with no useful information. Our scan-line-based method ensures that each voxel is detected exactly once when processing the triangle, enhancing performance without compromising the quality of the voxelization. The heart of our approach lies in the use of parallel, equidistant scan-lines to traverse the interiors of triangles. This minimizes redundant memory operations and avoids revisiting the same voxels, resulting in a significant performance boost. Moreover, our method's computational efficiency is complemented by its simplicity and portability. Written as a single compute shader in Graphics Library Shader Language (GLSL), it is highly adaptable to various rendering pipelines and hardware configurations. To validate our method, we conducted extensive experiments on a diverse set of models from the Stanford repository. Our results demonstrate not only the algorithm's efficiency, but also its ability to produce 26 tunnel free accurate voxelizations. The Gap Detection technique successfully identifies and addresses gaps, ensuring consistent and visually pleasing voxelized surfaces. Furthermore, we introduce the Slope Consistency Value metric, quantifying the alignment of each triangle with its primary axis. This metric provides insights into the impact of triangle orientation on scan-line based voxelization methods. It also aids in understanding how the Gap Detection technique effectively improves results by targeting specific areas where simple scan-line-based methods might fail. Our research contributes to the field of voxelization by offering a robust and efficient approach that overcomes the limitations of existing methods. The Gap Detection technique fills a critical gap in the voxelization process. By addressing these gaps, our algorithm enhances the visual quality and accuracy of voxelized models, making it valuable for a wide range of applications. In conclusion, "Closing the Gap: Efficient Voxelization with Equidistant Scan-lines and Gap Detection" presents an effective solution to the challenges of voxelization. Our research combines computational efficiency, accuracy, and innovative techniques to elevate the quality of voxelized surfaces. With its adaptable nature and valuable innovations, this technique could have a positive influence on computer graphics and visualization.

Keywords: voxelization, GPU acceleration, computer graphics, compute shaders

Procedia PDF Downloads 52

Authors: Ali Ulvi Uzer

Abstract:

Data reduction is an important topic in the field of pattern recognition applications. The basic concept is the reduction of multitudinous amounts of data down to the meaningful parts. The Principal Component Analysis (PCA) method is frequently used for data reduction. The Support Vector Machine (SVM) method is a discriminative classifier formally defined by a separating hyperplane. In other words, given labeled training data, the algorithm outputs an optimal hyperplane which categorizes new examples. This study offers a hybrid approach that uses the PCA for data reduction and Support Vector Machines (SVM) for classification. In order to detect the accuracy of the suggested system, two boreholes taken from the soil sample was used. The classification accuracies for this dataset were obtained through using ten-fold cross-validation method. As the results suggest, this system, which is performed through size reduction, is a feasible system for faster recognition of dataset so our study result appears to be very promising.

Keywords: feature selection, sequential forward selection, support vector machines, soil sample

Procedia PDF Downloads 437

Authors: Soufiane Zerraf, Malika Tridane, Said Belaaouad

Abstract:

CuHPO₃.2H₂O was synthesized by the direct method. CuHPO₃.2H₂O crystallizes in the orthorhombic system, space group P2₁2₁2₁, a = 6.7036 (2) Å, b = 7.3671 (4) Å, c = 8.9749 (4) Å, Z = 4, V = 443.24 (4) ų. The crystal structure was refined to R₁= 0.0154, R₂= 0.0380 for 19018 reflections satisfying criterion I ≥ 2σ (I). The structural resolution shows the existence of chains of ions HPO₃- linked together by hydrogen bonds. The crystalline structure is formed by chains consisting of Cu[O₃(H₂O)₃] deformed octahedral, which are connected to the vertices. The chains extend parallel to b and are mutually linked by PO₃ groups. The structure is closely related to that of CuSeO₃.2H₂O and CuTeO₃.2H₂O. The experimental studies of the infrared and Raman spectra were used to confirm the presence of the phosphate ion and were compared in the (0-4000) cm-1 region with the theoretical results calculated by the density functional theory (DFT) method to provide reliable assignments of all observed bands in the experimental spectra.

Keywords: crystal structure, X-ray diffraction, vibration study, thermal behavior, density functional theory

Procedia PDF Downloads 96

Authors: Samiah Alammari, Nassim Ammour

Abstract:

When providing a massive number of tasks successively to a deep learning process, a good performance of the model requires preserving the previous tasks data to retrain the model for each upcoming classification. Otherwise, the model performs poorly due to the catastrophic forgetting phenomenon. To overcome this shortcoming, we developed a successful continual learning deep model for remote sensing hyperspectral image regions classification. The proposed neural network architecture encapsulates two trainable subnetworks. The first module adapts its weights by minimizing the discrimination error between the land-cover classes during the new task learning, and the second module tries to learn how to replicate the data of the previous tasks by discovering the latent data structure of the new task dataset. We conduct experiments on HSI dataset Indian Pines. The results confirm the capability of the proposed method.

Keywords: continual learning, data reconstruction, remote sensing, hyperspectral image segmentation

Procedia PDF Downloads 232

Authors: G. M. Komaki, S. Mobin, E. Teymourian, S. Sheikh

Abstract:

This paper addresses minimizing the makespan of the distributed permutation flow shop scheduling problem. In this problem, there are several parallel identical factories or flowshops each with series of similar machines. Each job should be allocated to one of the factories and all of the operations of the jobs should be performed in the allocated factory. This problem has recently gained attention and due to NP-Hard nature of the problem, metaheuristic algorithms have been proposed to tackle it. Majority of the proposed algorithms require large computational time which is the main drawback. In this study, a general variable neighborhood search algorithm (GVNS) is proposed where several time-saving schemes have been incorporated into it. Also, the GVNS uses the sophisticated method to change the shaking procedure or perturbation depending on the progress of the incumbent solution to prevent stagnation of the search. The performance of the proposed algorithm is compared to the state-of-the-art algorithms based on standard benchmark instances.

Keywords: distributed permutation flow shop, scheduling, makespan, general variable neighborhood search algorithm

Procedia PDF Downloads 342

Authors: Veniamin Boiarkin, Bruno Bogaz Zarpelão, Muttukrishnan Rajarajan

Abstract:

The manufacturing sector is a vital component of most economies, which leads to a large number of cyberattacks on organisations, whereas disruption in operation may lead to significant economic consequences. Adversaries aim to disrupt the production processes of manufacturing companies, gain financial advantages, and steal intellectual property by getting unauthorised access to sensitive data. Access to sensitive data helps organisations to enhance the production and management processes. However, the majority of the existing data-sharing mechanisms are either susceptible to different cyber attacks or heavy in terms of computation overhead. In this paper, a privacy-preserving data-sharing scheme for smart utilities is proposed. First, a customer’s privacy adjustment mechanism is proposed to make sure that end-users have control over their privacy, which is required by the latest government regulations, such as the General Data Protection Regulation. Secondly, a local differential privacy-based mechanism is proposed to ensure the privacy of the end-users by hiding real data based on the end-user preferences. The proposed scheme may be applied to different industrial control systems, whereas in this study, it is validated for energy utility use cases consisting of smart, intelligent devices. The results show that the proposed scheme may guarantee the required level of privacy with an expected relative error in utility.

Keywords: data-sharing, local differential privacy, manufacturing, privacy-preserving mechanism, smart utility

Procedia PDF Downloads 59

Authors: Eszter Siposne Nandori

Abstract:

The paper describes how the subjective interpretation of poverty changed during the COVID-19 pandemic. The results of data collection at the end of 2020 are compared to the results of a similar survey from 2019. The methods of systematic data collection are used to collect data about the beliefs of the population about poverty. The analysis is carried out in Borsod-Abaúj-Zemplén County, one of the most backward areas in Hungary. The paper concludes that poverty is mainly linked to material values, and it did not change from 2019 to 2020. Some slight changes, however, highlight the effect of the pandemic: poverty is increasingly seen as a generational problem in 2020, and another important change is that isolation became more closely related to poverty.

Keywords: Hungary, interpretation of poverty, pandemic, systematic data collection, subjective poverty

Procedia PDF Downloads 107

Authors: Nicodemus M. J. Mbwambo, Yu-Shan Sun, Murali Sitaraman, Joan Krone

Abstract:

This paper presents a generic data abstraction that captures a navigable tree position. The mathematical modeling of the abstraction encapsulates the current tree position, which can be used to navigate and modify the tree. The encapsulation of the tree position in the data abstraction specification avoids the use of explicit references and aliasing, thereby simplifying verification of (imperative) client code that uses the data abstraction. To ease the tasks of such specification and verification, a general tree theory, rich with mathematical notations and results, has been developed. The paper contains an example to illustrate automated verification ramifications. With sufficient tree theory development, automated proving seems plausible even in the absence of a special-purpose tree solver.

Keywords: automation, data abstraction, maps, specification, tree, verification

Procedia PDF Downloads 148