Search results for: Mindlin’s 2nd gradient model

Authors: Hsin Lee, Hsuan Lee

Abstract:

Legal artificial intelligence (LegalAI) has been increasing applications within legal systems, propelled by advancements in natural language processing (NLP). Compared with general documents, legal case documents are typically long text sequences with intrinsic logical structures. Most existing language models have difficulty understanding the long-distance dependencies between different structures. Another unique challenge is that while the Judiciary of Taiwan has released legal judgments from various levels of courts over the years, there remains a significant obstacle in the lack of labeled datasets. This deficiency makes it difficult to train models with strong generalization capabilities, as well as accurately evaluate model performance. To date, models in Taiwan have yet to be specifically trained on judgment data. Given these challenges, this research proposes a Longformer-based pre-trained language model explicitly devised for retrieving similar judgments in Taiwanese legal documents. This model is trained on a self-constructed dataset, which this research has independently labeled to measure judgment similarities, thereby addressing a void left by the lack of an existing labeled dataset for Taiwanese judgments. This research adopts strategies such as early stopping and gradient clipping to prevent overfitting and manage gradient explosion, respectively, thereby enhancing the model's performance. The model in this research is evaluated using both the dataset and the Average Entropy of Offense-charged Clustering (AEOC) metric, which utilizes the notion of similar case scenarios within the same type of legal cases. Our experimental results illustrate our model's significant advancements in handling similarity comparisons within extensive legal judgments. By enabling more efficient retrieval and analysis of legal case documents, our model holds the potential to facilitate legal research, aid legal decision-making, and contribute to the further development of LegalAI in Taiwan.

Keywords: legal artificial intelligence, computation and language, language model, Taiwanese legal cases

Procedia PDF Downloads 65

Authors: Nuhu Dalhat Mu’azu, Abdullahi Usman, A. Bukhari, Muhammad Hussain Essa, Salihu Lukman

Abstract:

Electrokinetic remediation process was employed for the removal of four (4) heavy metals (Cr, Cu, Hg and Pb) from contaminated clay and bentonite soils under pulsed current supply mode. The effects of voltage gradient, pulse duty cycle and bentonite/clay ratio on the simultaneous removal efficiencies of the heavy metals were investigated. A total of thirteen experiments were designed and conducted according to factorial design with each experiment allowed to continuously ran for 3 weeks. Results obtained showed that increase in bentonite ratio decreased the removal efficiency of the heavy metals with no significant effect on the energy consumption. Conversely, increase in both voltage gradient and pulse duty cycle increased the heavy metals removal efficiencies with increased in energy consumption. Additionally, increase in voltage gradient increased the electrical conductivity and the soil pH due to due to continuous refill and replacement of process fluids as they decomposed under the induced voltage gradient. Under different operating conditions, the maximum removal efficiencies obtained for Cr, Cu, Hg, and Pb were 21.87, 83.2, 62.4, 78.06 and 16.65% respectively.

Keywords: clay, bentonite, soil remediation, mixed contaminants, heavy metals, and electrokinetic-adsorption

Procedia PDF Downloads 422

Authors: João Paulo Pascon

Abstract:

A constitutive model accounting for large strains, thermoviscoplasticity, and ductile damage evolution is proposed in the present work. To this end, a fully Lagrangian framework is employed, considering plane stress conditions and multiplicative split of the deformation gradient. The full model includes Gurson’s void growth, nucleation and coalescence, plastic work heating, strain and strain-rate hardening, thermal softening, and heat conductivity. The contribution of the work is the combination of all the above-mentioned features within the finite-strain setting. The model is implemented in a computer code using triangular finite elements and nonlinear analysis. Two mechanical examples involving ductile damage and finite strain levels are analyzed: an inhomogeneous tension specimen and the necking problem. Results demonstrate the capabilities of the developed formulation regarding ductile fracture and large deformations.

Keywords: ductile damage model, finite element method, large strains, thermoviscoplasticity

Procedia PDF Downloads 80

Authors: Satish Kumar Dewangan, Santosh Kumar Senapati

Abstract:

The present work focuses on the investigation of the effect of non-Newtonian behavior on the oil-water stratified flow in a horizontal channel using ANSYS Fluent. Coupled level set and volume of fluid (CLSVOF) has been used to capture the evolving interface assuming unsteady, coaxial flow with constant fluid properties. The diametric variation of oil volume fraction, mixture velocity, total pressure and pressure gradient has been studied. Non-Newtonian behavior of oil has been represented by the power law model in order to investigate the effect of flow behavior index. Stratified flow pattern tends to assume dispersed flow pattern with the change in the behavior of oil to non-Newtonian. The pressure gradient is found to be very much sensitive to the flow behavior index. The findings could be useful in designing the transportation pipe line in petroleum industries.

Keywords: oil-water stratified flow, horizontal channel, CLSVOF, non–Newtonian behaviour.

Procedia PDF Downloads 484

Authors: L. Melzerová, T. Janda, M. Šejnoha, J. Šejnoha

Abstract:

Two finite element (FEM) models are presented in this paper to address the random nature of the response of glued timber structures made of wood segments with variable elastic moduli evaluated from 3600 indentation measurements. This total database served to create the same number of ensembles as was the number of segments in the tested beam. Statistics of these ensembles were then assigned to given segments of beams and the Latin Hypercube Sampling (LHS) method was called to perform 100 simulations resulting into the ensemble of 100 deflections subjected to statistical evaluation. Here, a detailed geometrical arrangement of individual segments in the laminated beam was considered in the construction of two-dimensional FEM model subjected to in four-point bending to comply with the laboratory tests. Since laboratory measurements of local elastic moduli may in general suffer from a significant experimental error, it appears advantageous to exploit the full scale measurements of timber beams, i.e. deflections, to improve their prior distributions with the help of the Bayesian statistical method. This, however, requires an efficient computational model when simulating the laboratory tests numerically. To this end, a simplified model based on Mindlin’s beam theory was established. The improved posterior distributions show that the most significant change of the Young’s modulus distribution takes place in laminae in the most strained zones, i.e. in the top and bottom layers within the beam center region. Posterior distributions of moduli of elasticity were subsequently utilized in the 2D FEM model and compared with the original simulations.

Keywords: Bayesian inference, FEM, four point bending test, laminated timber, parameter estimation, prior and posterior distribution, Young’s modulus

Procedia PDF Downloads 279

Authors: Sol Girouard, Zona Kostic

Abstract:

A home is often the largest and most expensive purchase a person makes. Whether the decision leads to a successful outcome will be determined by a combination of critical factors. In this paper, we propose a method that efficiently handles all the factors in residential real estate and performs predictions given a feature space with high dimensionality while controlling for overfitting. The proposed method was built on gradient descent and boosting algorithms and uses a mixed optimizing technique to improve the prediction power. Usually, a single model cannot handle all the cases thus our approach builds multiple models based on different subsets of the predictors. The algorithm was tested on 3 million homes across the U.S., and the experimental results demonstrate the efficiency of this approach by outperforming techniques currently used in forecasting prices. With everyday changes on the real estate market, our proposed algorithm capitalizes from new events allowing more efficient predictions.

Keywords: real estate prediction, gradient descent, boosting, ensemble methods, active learning, training

Procedia PDF Downloads 271

Authors: Dua Hişam, Serhat İkizoğlu

Abstract:

Identifying the problem behind balance disorder is one of the most interesting topics in the medical literature. This study has considerably enhanced the development of artificial intelligence (AI) algorithms applying multiple machine learning (ML) models to sensory data on gait collected from humans to classify between normal people and those suffering from Vestibular System (VS) problems. Although AI is widely utilized as a diagnostic tool in medicine, AI models have not been used to perform feature extraction and identify VS disorders through training on raw data. In this study, three machine learning (ML) models, the Random Forest Classifier (RF), Extreme Gradient Boosting (XGB), and K-Nearest Neighbor (KNN), have been trained to detect VS disorder, and the performance comparison of the algorithms has been made using accuracy, recall, precision, and f1-score. With an accuracy of 95.28 %, Random Forest Classifier (RF) was the most accurate model.

Keywords: vestibular disorder, machine learning, random forest classifier, k-nearest neighbor, extreme gradient boosting

Procedia PDF Downloads 63

Authors: Md. Fazlul Karim, Ahmad Izani Md. Ismail, Mohammed Ashaque Meah

Abstract:

This paper focuses on the development of a 2-D Boundary Fitted and Nested Grid (BFNG) model to compute the tsunami propagation of Indonesian tsunami 2004 along the coastal region of Penang in Peninsular Malaysia. In the presence of a curvilinear coastline, boundary fitted grids are suitable to represent the model boundaries accurately. On the other hand, when large gradient of velocity within a confined area is expected, the use of a nested grid system is appropriate to improve the numerical accuracy with the least grid numbers. This paper constructs a shallow water nested and orthogonal boundary fitted grid model and presents computational results of the tsunami impact on the Penang coast due to the Indonesian tsunami of 2004. The results of the numerical simulations are compared with available data.

Keywords: boundary fitted nested model, tsunami, Penang Island, 2004 Indonesian Tsunami

Procedia PDF Downloads 318

Authors: M. D. D. Chathurangi, M. G. K. Nayanathara, K. M. H. M. M. Gunapala, G. M. R. G. Dayananda, Kavinga Yapa Abeywardena, Deemantha Siriwardana

Abstract:

Due to the growing popularity of social media platforms at present, there are various concerns, mostly cyberbullying, spam, bot accounts, and the spread of incorrect information. To develop a risk score calculation system as a thorough method for deciphering and exposing unethical social media profiles, this research explores the most suitable algorithms to our best knowledge in detecting the mentioned concerns. Various multiple models, such as Naïve Bayes, CNN, KNN, Stochastic Gradient Descent, Gradient Boosting Classifier, etc., were examined, and the best results were taken into the development of the risk score system. For cyberbullying, the Logistic Regression algorithm achieved an accuracy of 84.9%, while the spam-detecting MLP model gained 98.02% accuracy. The bot accounts identifying the Random Forest algorithm obtained 91.06% accuracy, and 84% accuracy was acquired for fake news detection using SVM.

Keywords: cyberbullying, spam behavior, bot accounts, fake news, machine learning

Procedia PDF Downloads 29

Authors: Benjamin K. Ghansah, Richard K. Appoh, Iliya Nababa, Eric K. Forkuo

Abstract:

The Nasia River is an important source of water for domestic and agricultural purposes to the inhabitants of its catchment. Major farming activities takes place within the floodplain of the river and its network of tributaries. The actual inundation extent of the river system is; however, unknown. Reasons for this lack of information include financial constraints and inadequate human resources as flood modelling is becoming increasingly complex by the day. Knowledge of the inundation extent will help in the assessment of risk posed by the annual flooding of the river, and help in the planning of flood recession agricultural activities. This study used a simple terrain based algorithm, Height Above Nearest Drainage (HAND), to delineate the floodplain of the Nasia River and its tributaries. The HAND model is a drainage normalized digital elevation model, which has its height reference based on the local drainage systems rather than the average mean sea level (AMSL). The underlying principle guiding the development of the HAND model is that hillslope flow paths behave differently when the reference gradient is to the local drainage network as compared to the seaward gradient. The new terrain model of the catchment was created using the NASA’s SRTM Digital Elevation Model (DEM) 30m as the only data input. Contours (HAND Contour) were then generated from the normalized DEM. Based on field flood inundation survey, historical information of flooding of the area as well as satellite images, a HAND Contour of 2m was found to best correlates with the flood inundation extent of the river and its tributaries. A percentage accuracy of 75% was obtained when the surface area created by the 2m contour was compared with surface area of the floodplain computed from a satellite image captured during the peak flooding season in September 2016. It was estimated that the flooding of the Nasia River and its tributaries created a floodplain area of 1011 km².

Keywords: digital elevation model, floodplain, HAND contour, inundation extent, Nasia River

Procedia PDF Downloads 447

Authors: Chin-Chen Chang, I-Ta Lee, Tsung-Ta Ke, Wen-Kai Tai

Abstract:

Common methods for resizing image size include scaling and cropping. However, these two approaches have some quality problems for reduced images. In this paper, we propose an image resizing algorithm by separating the main objects and the background. First, we extract two feature maps, namely, an enhanced visual saliency map and an improved gradient map from an input image. After that, we integrate these two feature maps to an importance map. Finally, we generate the target image using the importance map. The proposed approach can obtain desired results for a wide range of images.

Keywords: energy map, visual saliency, gradient map, seam carving

Procedia PDF Downloads 473

Authors: Jiwon Lee, Chanho Jung, Si-Hwan Jang, Kyung-Ill Kim, Sanghyun Joo, Wook-Ho Son

Abstract:

Since the advances in digital imaging technologies have led to development of high quality digital devices, there are a lot of illegal copies of copyrighted video content on the internet. Thus, we propose a high-quality (HQ) video watermarking scheme that can prevent these illegal copies from spreading out. The proposed scheme is applied spatial and temporal gradient methods to improve the fidelity and detection performance. Also, the scheme duplicates the watermark signal temporally to alleviate the signal reduction caused by geometric and signal-processing distortions. Experimental results show that the proposed scheme achieves better performance than previously proposed schemes and it has high fidelity. The proposed scheme can be used in broadcast monitoring or traitor tracking applications which need fast detection process to prevent illegally recorded video content from spreading out.

Keywords: editing prevention technique, gradient method, luminance change, video watermarking

Procedia PDF Downloads 451

Authors: Yuanjin Liu

Abstract:

Retirement withdrawal strategies are very important to minimize the probability of ruin in retirement. The ruin probability is modeled as a function of initial withdrawal age, gender, asset allocation, inflation rate, and initial withdrawal rate. The ruin probability is obtained based on the 2019 period life table for the Social Security, IRS Required Minimum Distribution (RMD) Worksheets, US historical bond and equity returns, and inflation rates using simulation. Several popular machine learning algorithms of the generalized additive model, random forest, support vector machine, extreme gradient boosting, and artificial neural network are built. The model validation and selection are based on the test errors using hyperparameter tuning and train-test split. The optimal model is recommended for retirees to monitor the ruin probability. The optimal withdrawal strategy can be obtained based on the optimal predictive model.

Keywords: ruin probability, retirement withdrawal strategies, predictive models, optimal model

Procedia PDF Downloads 71

Authors: Sergey Kopysov, Nikita Nedozhogin, Leonid Tonkov

Abstract:

The article presents a parallel iterative solver for large sparse linear systems which can be used on a heterogeneous platform. Traditionally, the problem of solving linear systems does not scale well on multi-CPU/multi-GPUs clusters. For example, most of the attempts to implement the classical conjugate gradient method were at best counted in the same amount of time as the problem was enlarged. The paper proposes the pipelined variant of the conjugate gradient method (PCG), a formulation that is potentially better suited for hybrid CPU/GPU computing since it requires only one synchronization point per one iteration instead of two for standard CG. The standard and pipelined CG methods need the vector entries generated by the current GPU and other GPUs for matrix-vector products. So the communication between GPUs becomes a major performance bottleneck on multi GPU cluster. The article presents an approach to minimize the communications between parallel parts of algorithms. Additionally, computation and communication can be overlapped to reduce the impact of data exchange. Using the pipelined version of the CG method with one synchronization point, the possibility of asynchronous calculations and communications, load balancing between the CPU and GPU for solving the large linear systems allows for scalability. The algorithm is implemented with the combined use of technologies: MPI, OpenMP, and CUDA. We show that almost optimum speed up on 8-CPU/2GPU may be reached (relatively to a one GPU execution). The parallelized solver achieves a speedup of up to 5.49 times on 16 NVIDIA Tesla GPUs, as compared to one GPU.

Keywords: conjugate gradient, GPU, parallel programming, pipelined algorithm

Procedia PDF Downloads 156

Authors: Minto Rattan, Tania Bose, Neeraj Chamoli

Abstract:

The present paper investigates the effect of linear thermal gradient on the steady-state creep behavior of rotating isotropic disc using threshold stress based Sherby’s creep law. The composite discs made of aluminum matrix reinforced with silicon carbide particulate has been taken for analysis. The stress and strain rate distributions have been calculated for discs rotating at linear thermal gradation using von Mises’ yield criterion. The material parameters have been estimated by regression fit of the available experimental data. The results are displayed and compared graphically in designer friendly format for the above said temperature profile with the disc operating under uniform temperature profile. It is observed that radial and tangential stresses show minor variation and the strain rates vary significantly in the presence of thermal gradation as compared to disc having uniform temperature.

Keywords: creep, isotropic, steady-state, thermal gradient

Procedia PDF Downloads 263

Authors: Ahmed I. Raheem

Abstract:

In this thesis, stratified and stratified wavy flow regimes have been investigated numerically for the oil (1.57 mPa s viscosity and 780 kg/m3 density) and water twophase flow in small and large horizontal steel pipes with a diameter between 0.0254 to 0.508 m by ANSYS Fluent software. Volume of fluid (VOF) with two phases flows using two equations family models (Realizable k-

Keywords: CFD, two-phase flow, pressure gradient, volume of fluid, large diameter, horizontal pipe, oil-water stratified and stratified wavy flow

Procedia PDF Downloads 426

Authors: Jisun Lee, Jay Hyoun Kwon

Abstract:

As an alternative way to compensate the INS (inertial navigation system) error in non-GNSS (Global Navigation Satellite System) environment, geophysical database referenced navigation is being studied. In this study, both gravity gradient and terrain data were combined to complement the weakness of sole geophysical data as well as to improve the stability of the positioning. The main process to compensate the INS error using geophysical database was constructed on the basis of the EKF (Extended Kalman Filter). In detail, two type of combination method, centralized and decentralized filter, were applied to check the pros and cons of its algorithm and to find more robust results. The performance of each navigation algorithm was evaluated based on the simulation by supposing that the aircraft flies with precise geophysical DB and sensors above nine different trajectories. Especially, the results were compared to the ones from sole geophysical database referenced navigation to check the improvement due to a combination of the heterogeneous geophysical database. It was found that the overall navigation performance was improved, but not all trajectories generated better navigation result by the combination of gravity gradient with terrain data. Also, it was found that the centralized filter generally showed more stable results. It is because that the way to allocate the weight for the decentralized filter could not be optimized due to the local inconsistency of geophysical data. In the future, switching of geophysical data or combining different navigation algorithm are necessary to obtain more robust navigation results.

Keywords: Extended Kalman Filter, geophysical database referenced navigation, gravity gradient, terrain

Procedia PDF Downloads 342

Authors: Dheyaa W. Abbood, Rafa H. AL-Suhaili, May S. Saleh

Abstract:

A model using the artificial neural networks and genetic algorithm technique is developed for obtaining optimum dimensions of the foundation length and protections of small hydraulic structures. The procedure involves optimizing an objective function comprising a weighted summation of the state variables. The decision variables considered in the optimization are the upstream and downstream cutoffs length sand their angles of inclination, the foundation length, and the length of the downstream soil protection. These were obtained for a given maximum difference in head, depth of impervious layer and degree of anisotropy.The optimization carried out subjected to constraints that ensure a safe structure against the uplift pressure force and sufficient protection length at the downstream side of the structure to overcome an excessive exit gradient. The Geo-studios oft ware, was used to analyze 1200 different cases. For each case the length of protection and volume of structure required to satisfy the safety factors mentioned previously were estimated. An ANN model was developed and verified using these cases input-output sets as its data base. A MatLAB code was written to perform a genetic algorithm optimization modeling coupled with this ANN model using a formulated optimization model. A sensitivity analysis was done for selecting the cross-over probability, the mutation probability and level ,the number of population, the position of the crossover and the weights distribution for all the terms of the objective function. Results indicate that the most factor that affects the optimum solution is the number of population required. The minimum value that gives stable global optimum solution of this parameters is (30000) while other variables have little effect on the optimum solution.

Keywords: inclined cutoff, optimization, genetic algorithm, artificial neural networks, geo-studio, uplift pressure, exit gradient, factor of safety

Procedia PDF Downloads 318

Authors: Adrian Szczyrba, Aneta Halka-Grysinska, Tomasz Baj, Tadeusz H. Dzido

Abstract:

Since prehistory, plants were constituted as an essential source of biologically active substances in folk medicine. One of the examples of medicinal plants is Ruta graveolens L. For a long time, Ruta g. herb has been famous for its spasmolytic, diuretic, or anti-inflammatory therapeutic effects. The wide spectrum of secondary metabolites produced by Ruta g. includes flavonoids (eg. rutin, quercetin), coumarins (eg. bergapten, umbelliferone) phenolic acids (eg. rosmarinic acid, chlorogenic acid), and limonoids. Unfortunately, the presence of produced substances is highly dependent on environmental factors like temperature, humidity, or soil acidity; therefore standardization is necessary. There were many attempts of characterization of various phytochemical groups (eg. coumarins) of Ruta graveolens using the normal – phase thin-layer chromatography (TLC). However, due to the so-called general elution problem, usually, some components remained unseparated near the start or finish line. Therefore Ruta graveolens is a very good model plant. Methanol and petroleum ether extract from its aerial parts were used to demonstrate the capabilities of the new device for gradient thin-layer chromatogram development. The development of gradient thin-layer chromatograms in the reversed-phase system in conventional horizontal chambers can be disrupted by problems associated with an excessive flux of the mobile phase to the surface of the adsorbent layer. This phenomenon is most likely caused by significant differences between the surface tension of the subsequent fractions of the mobile phase. An excessive flux of the mobile phase onto the surface of the adsorbent layer distorts the flow of the mobile phase. The described effect produces unreliable, and unrepeatable results, causing blurring and deformation of the substance zones. In the prototype device, the mobile phase solution is delivered onto the surface of the adsorbent layer with controlled velocity (by moving pipette driven by 3D machine). The delivery of the solvent to the adsorbent layer is equal to or lower than that of conventional development. Therefore chromatograms can be developed with optimal linear mobile phase velocity. Furthermore, under such conditions, there is no excess of eluent solution on the surface of the adsorbent layer so the higher performance of the chromatographic system can be obtained. Directly feeding the adsorbent layer with eluent also enables to perform convenient continuous gradient elution practically without the so-called gradient delay. In the study, unique fingerprints of methanol and petroleum ether extracts of Ruta graveolens aerial parts were obtained with stepwise gradient reversed-phase thin-layer chromatography. Obtained fingerprints under different chromatographic conditions will be compared. The advantages and disadvantages of the proposed approach to chromatogram development with controlled solvent velocity will be discussed.

Keywords: fingerprints, gradient thin-layer chromatography, reversed-phase TLC, Ruta graveolens

Procedia PDF Downloads 285

Authors: Mansour Al Qubeissi, Sergei S. Sazhin, Cyril Crua, Morgan R. Heikal

Abstract:

This paper presents the application of the Discrete Component Model for heating and evaporation to multi-component biodiesel fuel droplets in direct injection internal combustion engines. This model takes into account the effects of temperature gradient, recirculation and species diffusion inside droplets. A distinctive feature of the model used in the analysis is that it is based on the analytical solutions to the temperature and species diffusion equations inside the droplets. Nineteen types of biodiesel fuels are considered. It is shown that a simplistic model, based on the approximation of biodiesel fuel by a single component or ignoring the diffusion of components of biodiesel fuel, leads to noticeable errors in predicted droplet evaporation time and time evolution of droplet surface temperature and radius.

Keywords: heat/mass transfer, biodiesel, multi-component fuel, droplet

Procedia PDF Downloads 560

Authors: S. Khan, A. Khan, Abdul R. Soomrani, Raja F. Zafar, A. Waqas, G. Akbar

Abstract:

This paper proposes a low-complexity image interpolation method. Image interpolation is used to convert a low dimension video/image to high dimension video/image. The objective of a good interpolation method is to upscale an image in such a way that it provides better edge preservation at the cost of very low complexity so that real-time processing of video frames can be made possible. However, low complexity methods tend to provide real-time interpolation at the cost of blurring, jagging and other artifacts due to errors in slope calculation. Non-linear methods, on the other hand, provide better edge preservation, but at the cost of high complexity and hence they can be considered very far from having real-time interpolation. The proposed method is a linear method that uses gradient orientation for slope calculation, unlike conventional linear methods that uses the contrast of nearby pixels. Prewitt edge detection is applied to separate uniform regions and edges. Simple line averaging is applied to unknown uniform regions, whereas unknown edge pixels are interpolated after calculation of slopes using gradient orientations of neighboring known edge pixels. As a post-processing step, bilateral filter is applied to interpolated edge regions in order to enhance the interpolated edges.

Keywords: edge detection, gradient orientation, image upscaling, linear interpolation, slope tracing

Procedia PDF Downloads 254

Authors: Simone Leonardi, Luca Ardito

Abstract:

Micro-influencers are effective elements in the marketing strategies of companies and institutions because of their capability to create an hyper-engaged audience around a specific topic of interest. In recent years, many scientific approaches and commercial tools have handled the task of detecting this type of social media users. These strategies adopt solutions ranging from rule based machine learning models to deep neural networks and graph analysis on text, images, and account information. This work compares the existing solutions and proposes an ensemble method to generalize them with different input data and social media platforms. The deployed solution combines deep learning models on unstructured data with statistical machine learning models on structured data. We retrieve both social media accounts information and multimedia posts on Twitter and Instagram. These data are mapped into feature vectors for an eXtreme Gradient Boosting (XGBoost) classifier. Sixty different topics have been analyzed to build a rule based gold standard dataset and to compare the performances of our approach against baseline classifiers. We prove the effectiveness of our work by comparing the accuracy, precision, recall, and f1 score of our model with different configurations and architectures. We obtained an accuracy of 0.91 with our best performing model.

Keywords: deep learning, gradient boosting, image processing, micro-influencers, NLP, social media

Procedia PDF Downloads 174

Authors: Zhang Shuqi, Liu Dan

Abstract:

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

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

Procedia PDF Downloads 96

Authors: Byung Ho Jung, Dong Hoon Lim

Abstract:

Logistic regression is a statistical method for analyzing a dataset in which there are one or more independent variables that determine an outcome. Logistic regression is used extensively in numerous disciplines, including the medical and social science fields. In this paper, we address the problem of estimating parameters in the logistic regression based on MapReduce framework with RHadoop that integrates R and Hadoop environment applicable to large scale data. There exist three learning algorithms for logistic regression, namely Gradient descent method, Cost minimization method and Newton-Rhapson's method. The Newton-Rhapson's method does not require a learning rate, while gradient descent and cost minimization methods need to manually pick a learning rate. The experimental results demonstrated that our learning algorithms using RHadoop can scale well and efficiently process large data sets on commodity hardware. We also compared the performance of our Newton-Rhapson's method with gradient descent and cost minimization methods. The results showed that our newton's method appeared to be the most robust to all data tested.

Keywords: big data, logistic regression, MapReduce, RHadoop

Procedia PDF Downloads 271

Authors: Florian Kiebert, Hagen Schmidt

Abstract:

Standing surface acoustic waves (sSAWs) have already been used to manipulate particles in a microfluidic channel made of polydimethylsiloxan (PDMS). Usually two identical facing interdigital transducers (IDTs) are exploited to form an sSAW. Further, it has been reported that an sSAW can be generated by a single IDT using a superstrate resonating cavity or a PDMS post. Nevertheless, both setups utilising a traveling surface acoustic wave (tSAW) to create an sSAW for particle manipulation are costly. We present a simplified setup with a tSAW and a PDMS channel to form an sSAW. The incident tSAW is reflected at the rear PDMS channel wall and superimposed with the reflected tSAW. This superpositioned waves generates an sSAW but only at regions where the distance to the rear channel wall is smaller as the attenuation length of the tSAW minus the channel width. Therefore in a channel of 500µm width a tSAW with a wavelength λ = 120 µm causes a sSAW over the whole channel, whereas a tSAW with λ = 60 µm only forms an sSAW next to the rear wall of the channel, taken into account the attenuation length of a tSAW in water. Hence, it is possible to concentrate and trap particles in a defined region of the channel by adjusting the relation between the channel width and tSAW wavelength. Moreover, it is possible to generate a particle gradient over the channel width by picking the right ratio between channel wall and wavelength. The particles are moved towards the rear wall by the acoustic streaming force (ASF) and the acoustic radiation force (ARF) caused by the tSAW generated bulk acoustic wave (BAW). At regions in the channel were the sSAW is dominating the ARF focuses the particles in the pressure nodes formed by the sSAW caused BAW. On the one side the ARF generated by the sSAW traps the particle at the center of the tSAW beam, i. e. of the IDT aperture. On the other side, the ASF leads to two vortices, one on the left and on the right side of the focus region, deflecting the particles out of it. Through variation of the applied power it is possible to vary the number of particles trapped in the focus points, because near to the rear wall the amplitude of the reflected tSAW is higher and, therefore, the ARF of the sSAW is stronger. So in the vicinity of the rear wall the concentration of particles is higher but decreases with increasing distance to the wall, forming a gradient of particles. The particle gradient depends on the applied power as well as on the flow rate. Thus by variation of these two parameters it is possible to change the particle gradient. Furthermore, we show that the particle gradient can be modified by changing the relation between the channel width and tSAW wavelength. Concluding a single IDT generates an sSAW in a PDMS microchannel enables particle gradient generation in a well-defined microfluidic flow system utilising the ARF and ASF of a tSAW and an sSAW.

Keywords: ARF, ASF, particle manipulation, sSAW, tSAW

Procedia PDF Downloads 329

Authors: Da LI, Peng Xu

Abstract:

Recent extreme weather events, such as the 2021 European floods and North American heatwaves, have exposed the vulnerability of energy systems to both extreme demand scenarios and potential physical damage. Current integrated energy system designs often overlook performance under these challenging conditions. This research, focusing on a regional integrated energy system in Ningbo, China, proposes a distinct design method to optimize system reliability during extreme events. A multi-scenario model was developed, encompassing various extreme load conditions and potential system damages caused by severe weather. Based on this model, a comprehensive reliability improvement scheme was designed, incorporating a gradient approach to address different levels of disaster severity through the integration of advanced technologies like distributed energy storage. The scheme's effectiveness was validated through Monte Carlo simulations. Results demonstrate significant enhancements in energy supply reliability and peak load reduction capability under extreme scenarios. The findings provide several insights for improving energy system adaptability in the face of climate-induced challenges, offering valuable references for building reliable energy infrastructure capable of withstanding both extreme demands and physical threats across a spectrum of disaster intensities.

Keywords: extreme weather events, integrated energy systems, reliability improvement, climate change adaptation

Procedia PDF Downloads 5

Authors: Elmo Thiago Lins Cöuras Ford, Valentina Alessandra Carvalho do Vale

Abstract:

Given the unquestionable need of environmental preservation, the natural fibers have been seen as a salutary alternative for production of composites in substitution to the synthetic fibers, vitreous and metallic. In this work, the behavior of a composite was analyzed done with fiber of the peel of the coconut as reinforcement and latex as head office, when submitted the source of heat. The temperature profiles were verified in the internal surfaces and it expresses of the composite as well as the temperature gradient in the same. It was also analyzed the behavior of this composite when submitted to a cold source. As consequence, in function of the answers of the system, conclusions were reached.

Keywords: natural fiber, composite, temperature, latex, gradient

Procedia PDF Downloads 806

Authors: Shuo Huang, Huomiao Guo, Wenrui Huang

Abstract:

In navigation channels located in coasts and estuaries as the waterways connecting coastal water to ports or harbors, density-induced flow often exist due to the density-gradient or gravity gradient as the results of mixing between fresh water from coastal rivers and saline water in the coasts. The density-induced flow often carries sediment transport into navigation channels and causes sediment depositions in the channels. As a result, expensive dredging may need to maintain the water depth required for navigation. In our study, we conduct a series of experiments to investigate the characteristics of density-induced flow in the estuarine navigation channels under different density gradients. Empirical equations between density flow and salinity gradient were derived. Effects of coastal structures for regulating navigation channel on density-induced flow have also been investigated. Results will be very helpful for improving the understanding of the characteristics of density-induced flow in estuarine navigation channels. The results will also provide technical support for cost-effective waterway regulation and management to maintain coastal and estuarine navigation channels.

Keywords: density flow, estuarine, navigation channel, structure

Procedia PDF Downloads 249

Authors: Oluwatosin Olofintoye, Josiah Adeyemo, Gbemileke Shomade

Abstract:

The relationship between rainfall and runoff is an important issue in surface water hydrology therefore the understanding and development of accurate rainfall-runoff models and their applications in water resources planning, management and operation are of paramount importance in hydrological studies. This paper reviews some of the previous works on the rainfall-runoff process modeling. The hydrologic applications of conceptual models and artificial neural networks (ANNs) for the precipitation-runoff process modeling were studied. Gradient training methods such as error back-propagation (BP) and evolutionary algorithms (EAs) are discussed in relation to the training of artificial neural networks and it is shown that application of EAs to artificial neural networks training could be an alternative to other training methods. Therefore, further research interest to exploit the abundant expert knowledge in the area of artificial intelligence for the solution of hydrologic and water resources planning and management problems is needed.

Keywords: artificial intelligence, artificial neural networks, evolutionary algorithms, gradient training method, rainfall-runoff model

Procedia PDF Downloads 443

Authors: L. S. Godlevsky, N. V. Kresyun, V. P. Martsenyuk, K. S. Shakun, T. V. Tatarchuk, K. O. Prybolovets, L. F. Kalinichenko, M. Karpinski, T. Gancarczyk

Abstract:

Structures of eye bottom were extracted using multiscale texture gradient method and color characteristics of macular zone and vessels were verified in CIELAB scale. The difference of average values of L*, a* and b* coordinates of CIE (International Commision of Illumination) scale in patients with diabetes and healthy volunteers was compared. The average value of L* in diabetic patients exceeded such one in the group of practically healthy persons by 2.71 times (P < 0.05), while the value of a* index was reduced by 3.8 times when compared with control one (P < 0.05). b* index exceeded such one in the control group by 12.4 times (P < 0.05). The integrated index on color difference (ΔE) exceeded control value by 2.87 times (P < 0.05). More pronounced differences with ΔE were followed by a shorter period of MA appearance with a correlation level at -0.56 (P < 0.05). The specificity of diagnostics raised by 2.17 times (P < 0.05) and negative prognostic index exceeded such one determined with the expert method by 2.26 times (P < 0.05).

Keywords: diabetic retinopathy, multiscale texture gradient, color spectrum analysis, medical diagnostics

Procedia PDF Downloads 112