Search results for: pre-trained models

Authors: Suren Chilingaryan

Abstract:

We study two models describing a single two-level system coupled to two boson field modes in either a parallel or orthogonal setup. Both models may be feasible for experimental realization through Raman adiabatic driving in cavity QED. We study their ground state configurations; that is, we find the quantum precursors of the corresponding semi-classical phase transitions. We found that the ground state configurations of both models present the same critical coupling as the quantum Rabi model. Around this critical coupling, the ground state goes from the so-called normal configuration with no excitation, the qubit in the ground state and the fields in the quantum vacuum state, to a ground state with excitations, the qubit in a superposition of ground and excited state, while the fields are not in the vacuum anymore, for the first model. The second model shows a more complex ground state configuration landscape where we find the normal configuration mentioned above, two single-mode configurations, where just one of the fields and the qubit are excited, and a dual-mode configuration, where both fields and the qubit are excited.

Keywords: quantum optics, quantum phase transition, cavity QED, circuit QED

Procedia PDF Downloads 368

Authors: Salah Alrabeei, Mohammad Yousuf

Abstract:

The goal of option pricing theory is to help the investors to manage their money, enhance returns and control their financial future by theoretically valuing their options. However, most of the option pricing models have no analytical solution. Furthermore, not all the numerical methods are efficient to solve these models because they have nonsmoothing payoffs or discontinuous derivatives at the exercise price. In this paper, we solve the American option under jump diffusion models by using efficient time-dependent numerical methods. several techniques are integrated to reduced the overcome the computational complexity. Fast Fourier Transform (FFT) algorithm is used as a matrix-vector multiplication solver, which reduces the complexity from O(M2) into O(M logM). Partial fraction decomposition technique is applied to rational approximation schemes to overcome the complexity of inverting polynomial of matrices. The proposed method is easy to implement on serial or parallel versions. Numerical results are presented to prove the accuracy and efficiency of the proposed method.

Keywords: integral differential equations, jump–diffusion model, American options, rational approximation

Procedia PDF Downloads 122

Authors: Hany Mohamed Abdu, Mohamed Hassan Gadallah, El-Giushi Mokhtar, Yehia Mahmoud Ismail

Abstract:

Laser beam cutting is one nontraditional machining process. This paper optimizes the parameters of Laser beam cutting machining parameters of Stainless steel (316L) by considering the effect of input parameters viz. power, oxygen pressure, frequency and cutting speed. Statistical design of experiments are carried in three different levels and process responses such as 'Average kerf taper (Ta)' and 'Surface Roughness (Ra)' are measured accordingly. A quadratic mathematical model (RSM) for each of the responses is developed as a function of the process parameters. Responses predicted by the models (as per Taguchi’s L27 OA) are employed to search for an optimal parametric combination to achieve desired yield of the process. RSM models are developed for mean responses, S/N ratio, and standard deviation of responses. Optimization models are formulated as single objective problem subject to process constraints. Models are formulated based on Analysis of Variance (ANOVA) using MATLAB environment. Optimum solutions are compared with Taguchi Methodology results.

Keywords: optimization, laser cutting, robust design, kerf width, Taguchi method, RSM and DOE

Procedia PDF Downloads 621

Authors: D. S. C. Soares, D. G. Costa, J. T. S., A. K. S. Abud, T. P. Nunes, A. M. Oliveira Júnior

Abstract:

Mathematical models of drying are used for the purpose of understanding the drying process in order to determine important parameters for design and operation of the dryer. The jackfruit is a fruit with high consumption in the Northeast and perishability. It is necessary to apply techniques to improve their conservation for longer in order to diffuse it by regions with low consumption. This study aimed to analyse several mathematical models (Page, Lewis, and Midilli) to indicate one that best fits the conditions of convective drying process using performance indicators associated with each model: accuracy (Af) and noise factors (Bf), mean square error (RMSE) and standard error of prediction (% SEP). Jackfruit drying was carried out in convective type tray dryer at a temperature of 50°C for 9 hours. It is observed that the model Midili was more accurate with Af: 1.39, Bf: 1.33, RMSE: 0.01%, and SEP: 5.34. However, the use of the Model Midilli is not appropriate for purposes of control process due to need four tuning parameters. With the performance indicators used in this paper, the Page model showed similar results with only two parameters. It is concluded that the best correlation between the experimental and estimated data is given by the Page’s model.

Keywords: drying, models, jackfruit, biotechnology

Procedia PDF Downloads 379

Authors: T. Y. Liu, C. H. Lin, Y. M. Ferng

Abstract:

Wind energy is one of the clean renewable energy. However, the low frequency (20-200HZ) noise generated from the wind turbine blades, which bothers the residents, becomes the major problem to be developed. It is useful for predicting the aerodynamic noise by flow field and pressure distribution analysis on the wind turbine blades. Therefore, the main objective of this study is to use different turbulence models to analyse the flow field and pressure distributions of the wing blades. Three-dimensional Computation Fluid Dynamics (CFD) simulation of the flow field was used to calculate the flow phenomena for the National Renewable Energy Laboratory (NREL) Phase VI horizontal axis wind turbine rotor. Two different flow cases with different wind speeds were investigated: 7m/s with 72rpm and 15m/s with 72rpm. Four kinds of RANS-based turbulence models, Standard k-ε, Realizable k-ε, SST k-ω, and v2f, were used to predict and analyse the results in the present work. The results show that the predictions on pressure distributions with SST k-ω and v2f turbulence models have good agreements with experimental data.

Keywords: horizontal axis wind turbine, turbulence model, noise, fluid dynamics

Procedia PDF Downloads 265

Authors: Abdellatif Chebboub

Abstract:

Agricultural production is sensitive to weather and thus directly affected by climate change. Plausible estimates of these climate change impacts require combined use of climate, crop, and economic models. Results from previous studies vary substantially due to differences in models, scenarios, and data. This paper is part of a collective effort to systematically integrate these three types of models. We focus on the economic component of the assessment, investigating how nine global economic models of agriculture represent endogenous responses to seven standardized climate change scenarios produced by two climate and five crop models. These responses include adjustments in yields, area, consumption, and international trade. We apply biophysical shocks derived from the Intergovernmental Panel on Climate Change’s representative concentration pathway with end-of-century radiative forcing of 8.5 W/m2. The mean biophysical yield effect with no incremental CO2 fertilization is a 17% reduction globally by 2050 relative to a scenario with unchanging climate. Endogenous economic responses reduce yield loss to 11%, increase area of major crops by 11%, and reduce consumption by 3%. Agricultural production, cropland area, trade, and prices show the greatest degree of variability in response to climate change, and consumption the lowest. The sources of these differences include model structure and specification; in particular, model assumptions about ease of land use conversion, intensification, and trade. This study identifies where models disagree on the relative responses to climate shocks and highlights research activities needed to improve the representation of agricultural adaptation responses to climate change.

Keywords: climate change, agriculture, weather change, danger of climate change

Procedia PDF Downloads 317

Authors: Ferhat Demir

Abstract:

Even if strategic management is highly critical for all types of organizations, only a few maturity models have been proposed in business literature for the area of strategic management activities. This paper updates previous studies and presents a new conceptual model for assessing the maturity of strategic management in any organization. Strategic management maturity model (S-3M) is basically composed of 6 maturity levels with 7 dimensions. The biggest contribution of S-3M is to put innovation into agenda of strategic management. The main objective of this study is to propose a model to align innovation with business strategies. This paper suggests that innovation (breakthrough new products/services and business models) is the only way of creating sustainable growth and strategy studies cannot ignore this aspect. Maturity models should embrace innovation to respond dynamic business environment and rapidly changing customer behaviours.

Keywords: strategic management, innovation, business model, maturity model

Procedia PDF Downloads 194

Authors: Leah Lee

Abstract:

This paper examines the correlation between deep learning models and emotions with noises to see whether or not noises mask emotions. The deep learning models used are plain convolutional neural networks (CNN), auto-encoder, long short-term memory (LSTM), and Visual Geometry Group-16 (VGG-16). Emotion datasets used are Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS), Crowd-sourced Emotional Multimodal Actors Dataset (CREMA-D), Toronto Emotional Speech Set (TESS), and Surrey Audio-Visual Expressed Emotion (SAVEE). To make it four times bigger, audio set files, stretch, and pitch augmentations are utilized. From the augmented datasets, five different features are extracted for inputs of the models. There are eight different emotions to be classified. Noise variations are white noise, dog barking, and cough sounds. The variation in the signal-to-noise ratio (SNR) is 0, 20, and 40. In summation, per a deep learning model, nine different sets with noise and SNR variations and just augmented audio files without any noises will be used in the experiment. To compare the results of the deep learning models, the accuracy and receiver operating characteristic (ROC) are checked.

Keywords: auto-encoder, convolutional neural networks, long short-term memory, speech emotion recognition, visual geometry group-16

Procedia PDF Downloads 78

Authors: Dev Ras Pandey, Bharat Mishra, S. K. Tripathi

Abstract:

Cloud computing is ready to transform the structure of businesses and learning through supplying the real-time applications and provide an immediate help for small to medium sized businesses. The ability to run a hypervisor inside a virtual machine is important feature of virtualization and it is called nested virtualization. In today’s growing field of information technology, many of the virtualization models are available, that provide a convenient approach to implement, but decision for a single model selection is difficult. This paper explains the applications of operating system based virtualization in cloud computing with an appropriate/suitable model with their different specifications and user’s requirements. In the present paper, most popular models are selected, and the selection was based on container and hypervisor based virtualization. Selected models were compared with a wide range of user’s requirements as number of CPUs, memory size, nested virtualization supports, live migration and commercial supports, etc. and we identified a most suitable model of virtualization.

Keywords: virtualization, OS based virtualization, container based virtualization, hypervisor based virtualization

Procedia PDF Downloads 330

Authors: S. Belhour, A. K. Ferouani, C. Azizi

Abstract:

In recent years, a considerable effort (experience, numerical simulation, and theoretical prediction models) has characterised by high efficiency and low cost. Then an improved physics analytical model for simulating is proposed. The performance of GaAs MESFETs has been developed for use in device design for high frequency. This model is based on mathematical analysis, and a new approach for the standard model is proposed, this approach allowed to conceive applicable model for MESFET’s operating in the turn-one or pinch-off region and valid for the short-channel and the long channel MESFET’s in which the two dimensional potential distribution contributed by the depletion layer under the gate is obtained by conventional approximation. More ever, comparisons between the analytical models with different values of mobility are proposed, and a good agreement is obtained.

Keywords: analytical, gallium arsenide, MESFET, mobility, models

Procedia PDF Downloads 75

Authors: Mst. Tuhin Akter, Sharun Akter Khushbu, S. M. Shaqib

Abstract:

A real-time system architecture is highly effective for monitoring and detecting various damaged products or fruits that may deteriorate over time or become infected with diseases. Deep learning models have proven to be effective in building such architectures. However, building a deep learning model from scratch is a time-consuming and costly process. A more efficient solution is to utilize deep neural network (DNN) based transfer learning models in the real-time monitoring architecture. This study focuses on using a novel strawberry dataset to develop effective transfer learning models for the proposed real-time monitoring system architecture, specifically for evaluating and detecting strawberry freshness. Several state-of-the-art transfer learning models were employed, and the best performing model was found to be Xception, demonstrating higher performance across evaluation metrics such as accuracy, recall, precision, and F1-score.

Keywords: strawberry freshness evaluation, deep neural network, transfer learning, image augmentation

Procedia PDF Downloads 91

Authors: Elif Aykin Dincer, Mustafa Erbas

Abstract:

The vacuum drying behavior of beef slices (10 x 4 x 0.2 cm3) was experimentally investigated at the temperature of 60, 70, and 80°C under 25 mbar ultimate vacuum pressure and the mathematical models (Lewis, Page, Midilli, Two-term, Wangh and Singh and Modified Henderson and Pabis) were used to fit the vacuum drying of beef slices. The increase in drying air temperature resulted in a decrease in drying time. It took approximately 206, 180 and 157 min to dry beef slices from an initial moisture content to a final moisture content of 0.05 kg water/kg dry matter at 60, 70 and 80 °C of vacuum drying, respectively. It is also observed that the drying rate increased with increasing drying temperature. The coefficients (R2), the reduced chi-square (x²) and root mean square error (RMSE) values were obtained by application of six models to the experimental drying data. The best model with the highest R2 and, the lowest x² and RMSE values was selected to describe the drying characteristics of beef slices. The Page model has shown a better fit to the experimental drying data as compared to other models. In addition, the effective moisture diffusivities of beef slices in the vacuum drying at 60 - 80 °C varied in the range of 1.05 – 1.09 x 10-10 m2/s. Consequently, this results can be used to simulate vacuum drying process of beef slices and improve efficiency of the drying process.

Keywords: beef slice, drying models, effective diffusivity, vacuum

Procedia PDF Downloads 291

Authors: Alemaw Kefale, Aramde Fetene, Hayal Desta

Abstract:

This study analyzed the users' and experts' views on the green space quality and planning models in Debre Berhan (DB) and Debre Markos (DM) cities in Ethiopia. A questionnaire survey was conducted on 350 park users (148 from DB and 202 from DM) to rate the accessibility, size, shape, vegetation cover, social and cultural context, conservation and heritage, community participation, attractiveness, comfort, safety, inclusiveness, and maintenance of green spaces using a Likert scale. A key informant interview was held with 13 experts in DB and 12 in DM. Descriptive statistics and tests of independence of variables using the chi-square test were done. A statistically significant association existed between the perception of green space quality attributes and users' occupation (χ² (160, N = 350) = 224.463, p < 0.001), age (χ² (128, N = 350) = 212.812, p < 0.001), gender (χ² (32, N = 350) = 68.443, p < 0.001), and education level (χ² (192, N = 350) = 293.396, p < 0.001). 61.7 % of park users were unsatisfied with the quality of urban green spaces. The users perceived dense vegetation cover as "good," with a mean value of 3.41, while the remaining were perceived as "medium with a mean value of 2.62 – 3.32". Only quantitative space standards are practiced as a green space planning model, while other models are unfamiliar and never used in either city. Therefore, experts need to be aware of and practice urban green models during urban planning to ensure that new developments include green spaces to accommodate the community's and the environment's needs.

Keywords: urban green space, quality, users and experts, green space planning models, Ethiopia

Procedia PDF Downloads 59

Authors: Forrest Kaatz, Adhemar Bultheel

Abstract:

Au-Cu and Pt-M (M = Fe, Co, and Ni) nanocluster alloys are currently being investigated worldwide by many researchers for their interesting catalytic and nanophase properties. The low-temperature behavior of the phase diagrams is not well understood for alloys with nanometer sizes and shapes. These systems have similar bulk phase diagrams with the L12 (Au3Cu, Pt3M, AuCu3, and PtM3) structurally ordered intermetallics and the L10 structure for the AuCu and PtM intermetallics. We consider three models for low temperature ordering in the phase diagrams of Au–Cu and Pt–M nanocluster alloys. These models are valid for sizes ~ 5 nm and approach bulk values for sizes ~ 20 nm. We study the phase transition in nanoclusters with cubic, octahedral, and cuboctahedral shapes, covering the compositions of interest. These models are based on studying the melting temperatures in nanoclusters using the regular solution, mixing model for alloys. Experimentally, it is extremely challenging to determine thermodynamic data on nano–sized alloys. Reasonable agreement is found between these models and recent experimental data on nanometer clusters in the Au–Cu and Pt–M nanophase systems. From our data, experiments on nanocubes about 5 nm in size, of stoichiometric AuCu and PtM composition, could help differentiate between the models. Some available evidence indicates that ordered intermetallic nanoclusters have better catalytic properties than disordered ones. We conclude with a discussion of physical mechanisms whereby ordering could improve the catalytic properties of nanocluster alloys.

Keywords: catalytic reactions, gold nanoalloys, phase transitions, platinum nanoalloys

Procedia PDF Downloads 176

Authors: Gustavo Rodrigues, Hans Weber, Larissa Driemeier

Abstract:

The prevision of post-impact conditions and the behavior of the bodies during the impact have been object of several collision models. The formulation from Hertz’s theory is generally used dated from the 19^th century. These models consider the repulsive force as proportional to the deformation of the bodies under contact and may consider it proportional to the rate of deformation. The objective of the present work is to analyze the behavior of the bodies during impact using the Finite Element Method (FEM) with elastic and plastic material models. The main parameters to evaluate are, the contact force, the time of contact and the deformation of the bodies. An advantage of using the FEM approach is the possibility to apply a plastic deformation to the model according to the material definition: there will be used Johnson–Cook plasticity model whose parameters are obtained through empirical tests of real materials. This model allows analyzing the permanent deformation caused by impact, phenomenon observed in real world depending on the forces applied to the body. These results are compared between them and with the model-based Hertz theory.

Keywords: collision, impact models, finite element method, Hertz Theory

Procedia PDF Downloads 175

Authors: Saeed Moarrefi, Shou-Han Zhou, Liyuan Fan

Abstract:

Coupling with solid oxide fuel cells, methane dry reforming is a promising pathway for energy production while mitigating carbon emissions. However, the influence of carbon dioxide and electrochemical reactions on the internal dry reforming reaction within the fuel cells remains debatable, requiring accurate kinetic models to describe the internal reforming behaviors. We employed the Power-Law and Langmuir Hinshelwood–Hougen Watson models in an electrolyte-supported solid oxide fuel cell with a NiO-GDC-YSZ anode. The current density used in this study ranges from 0 to 1000 A/m2 at 973 K to 1173 K to estimate various kinetic parameters. The influence of the electrochemical reactions on the adsorption terms, the equilibrium of the reactions, the activation energy, the pre-exponential factor of the rate constant, and the adsorption equilibrium constant were studied. This study provides essential parameters for future simulations and highlights the need for a more detailed examination of reforming kinetic models.

Keywords: dry reforming kinetics, Langmuir Hinshelwood–Hougen Watson, power-law, SOFC

Procedia PDF Downloads 28

Authors: Mohan Li

Abstract:

Nowadays, intelligent research technology is more and more important than traditional research methods in urban research work, and this proportion will greatly increase in the next few decades. Frequently such analyzing work cannot be carried without some software engineering knowledge. And here, domain models of urban research will be necessary when applying software engineering knowledge to urban work. In many urban plan practice projects, making rational models, feeding reliable data, and providing enough computation all make indispensable assistance in producing good urban planning. During the whole work process, domain models can optimize workflow design. At present, human beings have entered the era of big data. The amount of digital data generated by cities every day will increase at an exponential rate, and new data forms are constantly emerging. How to select a suitable data set from the massive amount of data, manage and process it has become an ability that more and more planners and urban researchers need to possess. This paper summarizes and makes predictions of the emergence of technologies and technological iterations that may affect urban research in the future, discover urban problems, and implement targeted sustainable urban strategies. They are summarized into seven major domain models. They are urban and rural regional domain model, urban ecological domain model, urban industry domain model, development dynamic domain model, urban social and cultural domain model, urban traffic domain model, and urban space domain model. These seven domain models can be used to guide the construction of systematic urban research topics and help researchers organize a series of intelligent analytical tools, such as Python, R, GIS, etc. These seven models make full use of quantitative spatial analysis, machine learning, and other technologies to achieve higher efficiency and accuracy in urban research, assisting people in making reasonable decisions.

Keywords: big data, domain model, urban planning, urban quantitative analysis, machine learning, workflow design

Procedia PDF Downloads 177

Authors: Mojtaba Farzaneh, Majid Baghaei Nejad

Abstract:

This article proposes a new Flexible Capacitive Tactile Sensors based on paper sheets. This method combines the parameters of sensor's material and dielectric, and forms a new model of flexible capacitive sensors. The present article tries to present a practical explanation of this method's application and advantages. With the use of this new method, it is possible to make a more flexibility and accurate sensor in comparison with the current models. To assess the performance of this model, the common capacitive sensor is simulated and the proposed model of this article and one of the existing models are assessed. The results of this article indicate that the proposed model of this article can enhance the speed and accuracy of tactile sensor and has less error in comparison with the current models. Based on the results of this study, it can be claimed that in comparison with the current models, the proposed model of this article is capable of representing more flexibility and more accurate output parameters for touching the sensor, especially in abnormal situations and uneven surfaces, and increases accuracy and practicality.

Keywords: capacitive sensor, paper sheets, flexible, tactile, uneven

Procedia PDF Downloads 353

Authors: Bi-Huei Tsai

Abstract:

Previous studies on financial distress prediction choose the conventional failing and non-failing dichotomy; however, the distressed extent differs substantially among different financial distress events. To solve the problem, “non-distressed”, “slightly-distressed” and “reorganization and bankruptcy” are used in our article to approximate the continuum of corporate financial health. This paper explains different financial distress events using the two-stage method. First, this investigation adopts firm-specific financial ratios, corporate governance and market factors to measure the probability of various financial distress events based on multinomial logit models. Specifically, the bootstrapping simulation is performed to examine the difference of estimated misclassifying cost (EMC). Second, this work further applies macroeconomic factors to establish the credit cycle index and determines the distressed cut-off indicator of the two-stage models using such index. Two different models, one-stage and two-stage prediction models, are developed to forecast financial distress, and the results acquired from different models are compared with each other, and with the collected data. The findings show that the two-stage model incorporating financial ratios, corporate governance and market factors has the lowest misclassification error rate. The two-stage model is more accurate than the one-stage model as its distressed cut-off indicators are adjusted according to the macroeconomic-based credit cycle index.

Keywords: Multinomial logit model, corporate governance, company failure, reorganization, bankruptcy

Procedia PDF Downloads 378

Authors: Kudzanayi Chiteka, Wellington Makondo

Abstract:

The whole world is on the drive to go green owing to the negative effects of burning fossil fuels. Therefore, there is immediate need to identify and utilise alternative renewable energy sources. Among these energy sources solar energy is one of the most dominant in Zimbabwe. Solar power plants used to generate electricity are entirely dependent on solar radiation. For planning purposes, solar radiation values should be known in advance to make necessary arrangements to minimise the negative effects of the absence of solar radiation due to cloud cover and other naturally occurring phenomena. This research focused on the prediction of Global Horizontal Irradiation values for the sixth day given values for the past five days. Artificial intelligence techniques were used in this research. Three models were developed based on Support Vector Machines, Radial Basis Function, and Feed Forward Back-Propagation Artificial neural network. Results revealed that Support Vector Machines gives the best results compared to the other two with a mean absolute percentage error (MAPE) of 2%, Mean Absolute Error (MAE) of 0.05kWh/m²/day root mean square (RMS) error of 0.15kWh/m²/day and a coefficient of determination of 0.990. The other predictive models had prediction accuracies of MAPEs of 4.5% and 6% respectively for Radial Basis Function and Feed Forward Back-propagation Artificial neural network. These two models also had coefficients of determination of 0.975 and 0.970 respectively. It was found that prediction of GHI values for the future days is possible using artificial intelligence-based predictive models.

Keywords: solar energy, global horizontal irradiation, artificial intelligence, predictive models

Procedia PDF Downloads 274

Authors: Alexandros Pournaras, Anastasia Papadopoulou, Serafim Kontos, Anastasios Karakostas

Abstract:

The air quality of modern cities is an important concern, as poor air quality contributes to human health and environmental issues. Reliable air quality forecasting has, thus, gained scientific and governmental attention as an essential tool that enables authorities to take proactive measures for public safety. In this study, the potential of Machine Learning (ML) models to forecast PM2.5 at local scale is investigated in the city of Thessaloniki, the second largest city in Greece, which has been struggling with the persistent issue of air pollution. ML models, with proven ability to address timeseries forecasting, are employed to predict the PM2.5 concentrations and the respective Air Quality Index 5-days ahead by learning from daily historical air quality and meteorological data from 2014 to 2016 and gathered from two stations with different land use characteristics in the urban fabric of Thessaloniki. The performance of the ML models on PM2.5 concentrations is evaluated with common statistical methods, such as R squared (r²) and Root Mean Squared Error (RMSE), utilizing a portion of the stations’ measurements as test set. A multi-categorical evaluation is utilized for the assessment of their performance on respective AQIs. Several conclusions were made from the experiments conducted. Experimenting on MLs’ configuration revealed a moderate effect of various parameters and training schemas on the model’s predictions. Their performance of all these models were found to produce satisfactory results on PM2.5 concentrations. In addition, their application on untrained stations showed that these models can perform well, indicating a generalized behavior. Moreover, their performance on AQI was even better, showing that the MLs can be used as predictors for AQI, which is the direct information provided to the general public.

Keywords: Air Quality, AQ Forecasting, AQI, Machine Learning, PM2.5

Procedia PDF Downloads 78

Authors: M. Ouassaf, S. Belaid

Abstract:

A series of quinoline derivatives with antimalarial activity were subjected to two-dimensional quantitative structure-activity relationship (2D-QSAR) studies. Three models were implemented using multiple regression linear MLR, a regression partial least squares (PLS), nonlinear regression (MNLR), to see which descriptors are closely related to the activity biologic. We relied on a principal component analysis (PCA). Based on our results, a comparison of the quality of, MLR, PLS, and MNLR models shows that the MNLR (R = 0.914 and R² = 0.835, RCV= 0.853) models have substantially better predictive capability because the MNLR approach gives better results than MLR (R = 0.835 and R² = 0,752, RCV=0.601)), PLS (R = 0.742 and R² = 0.552, RCV=0.550) The model of MNLR gave statistically significant results and showed good stability to data variation in leave-one-out cross-validation. The obtained results suggested that our proposed model MNLR may be useful to predict the biological activity of derivatives of quinoline.

Keywords: antimalarial, quinoline, QSAR, PCA, MLR , MNLR, MLR

Procedia PDF Downloads 157

Authors: Xiongxiong You, Zhanwen Niu

Abstract:

Choosing an appropriate surrogate model plays an important role in surrogates-assisted evolutionary algorithms (SAEAs) since there are many types and different kernel functions in the surrogate model. In this paper, an adaptive selection of the best suitable surrogate model method is proposed to solve different kinds of expensive optimization problems. Firstly, according to the prediction residual error sum of square (PRESS) and different model selection strategies, the excellent individual surrogate models are integrated into multiple ensemble models in each generation. Then, based on the minimum root of mean square error (RMSE), the best suitable surrogate model is selected dynamically. Secondly, two methods with dynamic number of models and selection strategies are designed, which are used to show the influence of the number of individual models and selection strategy. Finally, some compared studies are made to deal with several commonly used benchmark problems, as well as a rotor system optimization problem. The results demonstrate the accuracy and robustness of the proposed method.

Keywords: adaptive selection, expensive optimization, rotor system, surrogates assisted evolutionary algorithms

Procedia PDF Downloads 141

Authors: Sajjad Negahban, Ruihe Wang, Baojiang Sun

Abstract:

Gas-lift dual gradient drilling is a solution for deepwater drilling challenges. As well, Continuous development of drilling technology leads to increase employment of mineral oil based drilling fluids and synthetic-based drilling fluids, which have adequate characteristics such as: high rate of penetration, lubricity, shale inhibition and low toxicity. The paper discusses utilization of nitrified mineral oil base drilling for deepwater drilling and for more accurate prediction of pressure in DGD at marine riser, solubility and bubble pressure were considered in steady state hydraulic model. The Standing bubble pressure and solubility correlations, and two models which were acquired from experimental determination were applied in hydraulic model. The effect of the black oil correlations, and new solubility and bubble pressure models was evaluated on the PVT parameters such as oil formation volume factor, density, viscosity, volumetric flow rate. Eventually, the consequent simulated pressure profile due to these models was presented.

Keywords: solubility, bubble pressure, gas-lift dual gradient drilling, steady state hydraulic model

Procedia PDF Downloads 277

Authors: Shunzuo Wu, Xudong Luo, Yuanxiu Liao

Abstract:

Recently, the rapid development of deep learning makes artificial intelligence (AI) penetrate into many fields, replacing manual work there. In particular, AI systems also become a research focus in the field of automatic office. To meet real needs in automatic officiating, in this paper we develop an automatic form filling system. Specifically, it uses two classical neural network models and several word embedding models to classify various relevant information elicited from the Internet. When training the neural network models, we use less noisy and balanced data for training. We conduct a series of experiments to test my systems and the results show that our system can achieve better classification results.

Keywords: artificial intelligence and office, NLP, deep learning, text classification

Procedia PDF Downloads 201

Authors: Elison Eduardo Jardim Bierhals, Claudineia Brazil, Deivid Pires, Rafael Haag, Elton Gimenez Rossini

Abstract:

The objective of this work is to evaluate the results of solar radiation projections between 2006 and 2013 for the state of Rio Grande do Sul, Brazil. The projections are provided by the General Circulation Models (MCGs) belonging to the Coupled Model Intercomparison Phase 5 (CMIP5). In all, the results of the simulation of six models are evaluated, compared to monthly data, measured by a network of thirteen meteorological stations of the National Meteorological Institute (INMET). The performance of the models is evaluated by the Nash coefficient and the Bias. The results are presented in the form of tables, graphs and spatialization maps. The ACCESS1-0 RCP 4.5 model presented the best results for the solar radiation simulations, for the most optimistic scenario, in much of the state. The efficiency coefficients (CEF) were between 0.95 and 0.98. In the most pessimistic scenario, HADGen2-AO RCP 8.5 had the best accuracy among the analyzed models, presenting coefficients of efficiency between 0.94 and 0.98. From this validation, solar radiation projection maps were elaborated, indicating a seasonal increase of this climatic variable in some regions of the Brazilian territory, mainly in the spring.

Keywords: climate change, projections, solar radiation, validation

Procedia PDF Downloads 208

Authors: Sumit Sen, Sohan Khedekar, Umang Shinde, Shivam Bhargava

Abstract:

This study has been undertaken to investigate whether the deep learning models are able to predict the future stock prices by training the model with the historical stock price data. Since this work required time series analysis, various models are present today to perform time series analysis such as Recurrent Neural Network LSTM, ARIMA and Facebook Prophet. Applying these models the movement of stock price of stocks are predicted and also tried to provide the future prediction of the stock price of a stock. Final product will be a stock price prediction web application that is developed for providing the user the ease of analysis of the stocks and will also provide the predicted stock price for the next seven days.

Keywords: Autoregressive Integrated Moving Average, Deep Learning, Long Short Term Memory, Time-series

Procedia PDF Downloads 143

Authors: Toktam Khatibi

Abstract:

Self-supervised learning (SSL) includes machine learning models that are trained on one aspect and/or one part of the input to learn other aspects and/or part of it. SSL models are divided into two different categories, including pre-text task-based models and contrastive learning ones. Pre-text tasks are some auxiliary tasks learning pseudo-labels, and the trained models are further fine-tuned for downstream tasks. However, one important disadvantage of SSL using pre-text task solving is defining an appropriate pre-text task for each image dataset with a variety of image modalities. Therefore, it is required to design an appropriate pretext task automatically for each dataset and each downstream task. To the best of our knowledge, the automatic designing of pretext tasks for image analysis has not been considered yet. In this paper, we present a framework based on In-context learning that describes each task based on its input and output data using a pre-trained image transformer. Our proposed method combines the input image and its learned description for optimizing the pre-text task design and its hyper-parameters using Meta-learning models. The representations learned from the pre-text tasks are fine-tuned for solving the downstream tasks. We demonstrate that our proposed framework outperforms the compared ones on unseen tasks and image modalities in addition to its superior performance for previously known tasks and datasets.

Keywords: in-context learning (ICL), meta learning, self-supervised learning (SSL), vision-language domain, transformers

Procedia PDF Downloads 81

Authors: Mian Huang, Chi Ma, Junyu Lin, William Lu

Abstract:

Introduction: Bone health attracts more attention recently and an intelligent question and answer (QA) chatbot for osteoporosis is helpful for science popularization. With Generative Pretrained Transformer (GPT) technology developing, we build an osteoporosis corpus dataset and then fine-tune LLaMA, a famous open-source GPT foundation large language model(LLM), on our self-constructed osteoporosis corpus. Evaluated by clinical orthopedic experts, our fine-tuned model outperforms vanilla LLaMA on osteoporosis QA task in Chinese. Three-dimensional quantitative computed tomography (QCT) measured bone mineral density (BMD) is considered as more accurate than DXA for BMD measurement in recent years. We develop an automatic Phantom-less QCT(PL-QCT) that is more efficient for BMD measurement since no need of an external phantom for calibration. Combined with LLM on osteoporosis, our PL-QCT provides efficient and accurate BMD measurement for our chatbot users. Material and Methods: We build an osteoporosis corpus containing about 30,000 Chinese literatures whose titles are related to osteoporosis. The whole process is done automatically, including crawling literatures in .pdf format, localizing text/figure/table region by layout segmentation algorithm and recognizing text by OCR algorithm. We train our model by continuous pre-training with Low-rank Adaptation (LoRA, rank=10) technology to adapt LLaMA-7B model to osteoporosis domain, whose basic principle is to mask the next word in the text and make the model predict that word. The loss function is defined as cross-entropy between the predicted and ground-truth word. Experiment is implemented on single NVIDIA A800 GPU for 15 days. Our automatic PL-QCT BMD measurement adopt AI-associated region-of-interest (ROI) generation algorithm for localizing vertebrae-parallel cylinder in cancellous bone. Due to no phantom for BMD calibration, we calculate ROI BMD by CT-BMD of personal muscle and fat. Results & Discussion: Clinical orthopaedic experts are invited to design 5 osteoporosis questions in Chinese, evaluating performance of vanilla LLaMA and our fine-tuned model. Our model outperforms LLaMA on over 80% of these questions, understanding ‘Expert Consensus on Osteoporosis’, ‘QCT for osteoporosis diagnosis’ and ‘Effect of age on osteoporosis’. Detailed results are shown in appendix. Future work may be done by training a larger LLM on the whole orthopaedics with more high-quality domain data, or a multi-modal GPT combining and understanding X-ray and medical text for orthopaedic computer-aided-diagnosis. However, GPT model gives unexpected outputs sometimes, such as repetitive text or seemingly normal but wrong answer (called ‘hallucination’). Even though GPT give correct answers, it cannot be considered as valid clinical diagnoses instead of clinical doctors. The PL-QCT BMD system provided by Bone’s QCT(Bone’s Technology(Shenzhen) Limited) achieves 0.1448mg/cm2(spine) and 0.0002 mg/cm2(hip) mean absolute error(MAE) and linear correlation coefficient R2=0.9970(spine) and R2=0.9991(hip)(compared to QCT-Pro(Mindways)) on 155 patients in three-center clinical trial in Guangzhou, China. Conclusion: This study builds a Chinese osteoporosis corpus and develops a fine-tuned and domain-adapted LLM as well as a PL-QCT BMD measurement system. Our fine-tuned GPT model shows better capability than LLaMA model on most testing questions on osteoporosis. Combined with our PL-QCT BMD system, we are looking forward to providing science popularization and early morning screening for potential osteoporotic patients.

Keywords: GPT, phantom-less QCT, large language model, osteoporosis

Procedia PDF Downloads 71

Authors: Paul Ojeaga

Abstract:

Can the level of innovation use drive un-validated business models across regions? To what extent does industrial sector attractiveness drive firm’s success across regions at the time of start-up? This study examines the role of innovation on start-up success in six regions of the world (namely Sub Saharan Africa, the Middle East and North Africa, Latin America, South East Asia Pacific, the European Union and the United States representing North America) using macroeconomic variables. While there have been studies using firm level data, results from such studies are not suitable for national policy decisions. The need to drive a regional innovation policy also begs for an answer, therefore providing room for this study. Results using dynamic panel estimation show that innovation counts in the early infancy stage of new business life cycle. The results are robust even after controlling for time fixed effects and the study present variance-covariance estimation robust standard errors.

Keywords: industrial economics, un-validated business models, scalable models, entrepreneurship

Procedia PDF Downloads 283