Search results for: forecasting accuracy

Authors: N. R. Badurally Adam, S. R. Monebhurrun, M. Z. Dauhoo, A. Khoodaruth

Abstract:

Transport energy demand is vital for the economic growth of any country. Globalisation and better standard of living plays an important role in transport energy demand. Recently, transport energy demand in Mauritius has increased significantly, thus leading to an abuse of natural resources and thereby contributing to global warming. Forecasting the transport energy demand is therefore important for controlling and managing the demand. In this paper, we develop a model to predict the transport energy demand. The model developed is based on a system of five stochastic differential equations (SDEs) consisting of five endogenous variables: fuel price, population, gross domestic product (GDP), number of vehicles and transport energy demand and three exogenous parameters: crude birth rate, crude death rate and labour force. An interval of seven years is used to avoid any falsification of result since Mauritius is a developing country. Data available for Mauritius from year 2003 up to 2009 are used to obtain the values of design variables by applying genetic algorithm. The model is verified and validated for 2010 to 2012 by substituting the values of coefficients obtained by GA in the model and using particle swarm optimisation (PSO) to predict the values of the exogenous parameters. This model will help to control the transport energy demand in Mauritius which will in turn foster Mauritius towards a pollution-free country and decrease our dependence on fossil fuels.

Keywords: genetic algorithm, modeling, particle swarm optimization, stochastic differential equations, transport energy demand

Procedia PDF Downloads 360

Authors: Timilehin Martins Oyinloye, Won Byong Yoon

Abstract:

Simulation technology is being adopted in many industries, with research focusing on the development of new ways in which technology becomes embedded within production, services, and society in general. 3D printing (3DP) technology is fast developing in the food industry. However, the limited processability of high-performance material restricts the robustness of the process in some cases. Significantly, the printability of materials becomes the foundation for extrusion-based 3DP, with residual stress being a major challenge in the printing of complex geometry. In many situations, the trial-a-error method is being used to determine the optimum printing condition, which results in time and resource wastage. In this report, the analysis of 3 moisture levels for surimi paste was investigated for an optimum 3DP material and printing conditions by probing its rheology, flow characteristics in the nozzle, and post-deposition process using the finite element method (FEM) model. Rheological tests revealed that surimi pastes with 82% moisture are suitable for 3DP. According to the FEM model, decreasing the nozzle diameter from 1.2 mm to 0.6 mm, increased the die swell from 9.8% to 14.1%. The die swell ratio increased due to an increase in the pressure gradient (1.15107 Pa to 7.80107 Pa) at the nozzle exit. The nozzle diameter influenced the fluid properties, i.e., the shear rate, velocity, and pressure in the flow field, as well as the residual stress and the deformation of the printed sample, according to FEM simulation. The post-printing stability of the model was investigated using the additive layer manufacturing (ALM) model. The ALM simulation revealed that the residual stress and total deformation of the sample were dependent on the nozzle diameter. A small nozzle diameter (0.6 mm) resulted in a greater total deformation (0.023), particularly at the top part of the model, which eventually resulted in the sample collapsing. As the nozzle diameter increased, the accuracy of the model improved until the optimum nozzle size (1.0 mm). Validation with 3D-printed surimi products confirmed that the nozzle diameter was a key parameter affecting the geometry accuracy of 3DP of surimi paste.

Keywords: 3D printing, deformation analysis, die swell, numerical simulation, surimi paste

Procedia PDF Downloads 54

Authors: Alan C. Lin, Tzu-Kuan Lin, Tsong Der Lin

Abstract:

This paper proposes a new method to find the equations of transformation matrix for the rotation angles of the two rotational axes and the coordinates of the three linear axes of an orthogonal multi-axis milling machine. This approach provides intuitive physical meanings for rotation angles of multi-axis machines, which can be used to evaluate the accuracy of the conversion from CL data to NC data.

Keywords: CAM, multi-axis milling machining, transformation matrix, rotation angles

Procedia PDF Downloads 464

Authors: M. Abuziarov

Abstract:

The designed complex is intended for the numerical simulation of fast dynamic processes of interaction of heterogeneous environments susceptible to the significant formability. The main challenges in solving such problems are associated with the construction of the numerical meshes. Currently, there are two basic approaches to solve this problem. One is using of Lagrangian or Lagrangian Eulerian grid associated with the boundaries of media and the second is associated with the fixed Eulerian mesh, boundary cells of which cut boundaries of the environment medium and requires the calculation of these cut volumes. Both approaches require the complex grid generators and significant time for preparing the code’s data for simulation. In this codes these problems are solved using two grids, regular fixed and mobile local Euler Lagrange - Eulerian (ALE approach) accompanying the contact and free boundaries, the surfaces of shock waves and phase transitions, and other possible features of solutions, with mutual interpolation of integrated parameters. For modeling of both liquids and gases, and deformable solids the Godunov scheme of increased accuracy is used in Lagrangian - Eulerian variables, the same for the Euler equations and for the Euler- Cauchy, describing the deformation of the solid. The increased accuracy of the scheme is achieved by using 3D spatial time dependent solution of the discontinuity problem (3D space time dependent Riemann's Problem solver). The same solution is used to calculate the interaction at the liquid-solid surface (Fluid Structure Interaction problem). The codes does not require complex 3D mesh generators, only the surfaces of the calculating objects as the STL files created by means of engineering graphics are given by the user, which greatly simplifies the preparing the task and makes it convenient to use directly by the designer at the design stage. The results of the test solutions and applications related to the generation and extension of the detonation and shock waves, loading the constructions are presented.

Keywords: fluid structure interaction, Riemann's solver, Euler variables, 3D codes

Procedia PDF Downloads 424

Authors: L. Guevara, Glória L. S., Corea E. E, A. Ramírez-Zamora M., Salinas-Martinez J. A., Angeles-Hernandez J. C.

Abstract:

Mathematical models are useful for identifying the characteristics of sheep lactation curves to develop and implement improved strategies. However, the accuracy of these models is influenced by factors such as the recording regime, mainly the intervals between test day records (TDR). The current study aimed to evaluate the effect of different TDR intervals on the goodness of fit of the Wood model (WM) applied to dairy sheep lactations. A total of 4,494 weekly TDRs from 156 lactations of dairy crossbred sheep were analyzed. Three new databases were generated from the original weekly TDR data (7D), comprising intervals of 14(14D), 21(21D), and 28(28D) days. The parameters of WM were estimated using the “minpack.lm” package in the R software. The shape of the lactation curve (typical and atypical) was defined based on the WM parameters. The goodness of fit was evaluated using the mean square of prediction error (MSPE), Root of MSPE (RMSPE), Akaike´s Information Criterion (AIC), Bayesian´s Information Criterion (BIC), and the coefficient of correlation (r) between the actual and estimated total milk yield (TMY). WM showed an adequate estimate of TMY regardless of the TDR interval (P=0.21) and shape of the lactation curve (P=0.42). However, we found higher values of r for typical curves compared to atypical curves (0.9vs.0.74), with the highest values for the 28D interval (r=0.95). In the same way, we observed an overestimated peak yield (0.92vs.6.6 l) and underestimated time of peak yield (21.5vs.1.46) in atypical curves. The best values of RMSPE were observed for the 28D interval in both lactation curve shapes. The significant lowest values of AIC (P=0.001) and BIC (P=0.001) were shown by the 7D interval for typical and atypical curves. These results represent the first approach to define the adequate interval to record the regime of dairy sheep in Latin America and showed a better fitting for the Wood model using a 7D interval. However, it is possible to obtain good estimates of TMY using a 28D interval, which reduces the sampling frequency and would save additional costs to dairy sheep producers.

Keywords: gamma incomplete, ewes, shape curves, modeling

Procedia PDF Downloads 56

Authors: Marju Ben Sayed, Shigeko Haruyama

Abstract:

Among all kinds of natural disaster, the flood is a common feature in rapidly urbanizing Dhaka city. In this research, assessment of flood risk of Dhaka metropolitan area has been investigated by using an integrated approach of GIS, remote sensing and socio-economic data. The purpose of the study is to measure the flooding risk concerning with the flood protection embankment in big flooding events (1988, 1998 and 2004) and urbanization of Dhaka metropolitan zone. In this research, we considered the Dhaka city into two parts; East Dhaka (outside the flood protection embankment) and West Dhaka (inside the flood protection embankment). Using statistical data, we explored the socio-economic status of the study area population by comparing the density of population, land price and income level. We have drawn the cross section profile of the flood protection embankment into three different points for realizing the flooding risk in the study area, especially in the big flooding year (1988, 1998 and 2004). According to the physical condition of the study area, the land use/land cover map has been classified into five classes. Comparing with each land cover unit, historical weather station data and the socio-economic data, the flooding risk has been evaluated. Moreover, we compared between DEM data and each land cover units to find out the relationship with flood. It is expected that, this study could contribute to effective flood forecasting, relief and emergency management for a future flood event in Dhaka city.

Keywords: land use, land cover change, socio-economic, Dhaka city, GIS, flood

Procedia PDF Downloads 280

Authors: Sofiqul Islam, V. Murugan, Prema Kumari, Hari

Abstract:

Gentamicin is a broad spectrum water-soluble aminoglycoside antibiotics produced by the fermentation process of microorganism known as Micromonospora purpurea. It is widely used for the treatment of infection caused by both gram positive and gram negative bacteria. Gentamicin consists of a mixture of aminoglycoside components like C1, C1a, C2a, and C2. The molecular structure of Gentamicin and its related substances showed that it has lack of presence of chromophore group in the molecule due to which the detection of such components were quite critical and challenging. In this study, a simple Reversed Phase-High Performance Liquid Chromatographic (RP-HPLC) method using ultraviolet (UV) detector was developed and validated for quantification of the related substances present in Gentamicin drug substances. The method was achieved by using Thermo Scientific Hypersil Gold analytical column (150 x 4.6 mm, 5 µm particle size) with isocratic elution composed of methanol: water: glacial acetic acid: sodium hexane sulfonate in the ratio 70:25:5:3 % v/v/v/w as a mobile phase at a flow rate of 0.5 mL/min, column temperature was maintained at 30 °C and detection wavelength of 330 nm. The four components of Gentamicin namely Gentamicin C1, C1a, C2a, and C2 were well separated along with the related substance present in Gentamicin. The Limit of Quantification (LOQ) values were found to be at 0.0075 mg/mL. The accuracy of the method was quite satisfactory in which the % recovery was resulted between 95-105% for the related substances. The correlation coefficient (≥ 0.995) shows the linearity response against concentration over the range of Limit of Quantification (LOQ). Precision studies showed the % Relative Standard Deviation (RSD) values less than 5% for its related substance. The method was validated in accordance with the International Conference of Harmonization (ICH) guideline with various parameters like system suitability, specificity, precision, linearity, accuracy, limit of quantification, and robustness. This proposed method was easy and suitable for use for the quantification of related substances in routine analysis of Gentamicin formulations.

Keywords: reversed phase-high performance liquid chromatographic (RP-HPLC), high performance liquid chromatography, gentamicin, isocratic, ultraviolet

Procedia PDF Downloads 150

Authors: Xiaobin Liang, Wei Liang, Laibin Zhang, Xiaoyan Guo

Abstract:

Pipelines pass through coal mined gobs inevitably in the mining area, the stability of which has great influence on the safety of pipelines. After extensive literature study and field research, it was found that there are a few risk assessment methods for coal mined gob pipelines, and there is a lack of data on the gob sites. Therefore, the fuzzy comprehensive evaluation method is widely used based on expert opinions. However, the subjective opinions or lack of experience of individual experts may lead to inaccurate evaluation results. Hence the accuracy of the results needs to be further improved. This paper presents a comprehensive approach to achieve this purpose by combining bow-tie model and cloud inference. The specific evaluation process is as follows: First, a bow-tie model composed of a fault tree and an event tree is established to graphically illustrate the probability and consequence indicators of pipeline failure. Second, the interval estimation method can be scored in the form of intervals to improve the accuracy of the results, and the censored mean algorithm is used to remove the maximum and minimum values of the score to improve the stability of the results. The golden section method is used to determine the weight of the indicators and reduce the subjectivity of index weights. Third, the failure probability and failure consequence scores of the pipeline are converted into three numerical features by using cloud inference. The cloud inference can better describe the ambiguity and volatility of the results which can better describe the volatility of the risk level. Finally, the cloud drop graphs of failure probability and failure consequences can be expressed, which intuitively and accurately illustrate the ambiguity and randomness of the results. A case study of a coal mine gob pipeline carrying natural gas has been investigated to validate the utility of the proposed method. The evaluation results of this case show that the probability of failure of the pipeline is very low, the consequences of failure are more serious, which is consistent with the reality.

Keywords: bow-tie model, natural gas pipeline, coal mine gob, cloud inference

Procedia PDF Downloads 238

Authors: Thomas Monahan, Nicolas Gorius, Thanh Nguyen

Abstract:

Exoplanet atmospheric parameters retrieval is a complex, computationally intensive, inverse modeling problem in which an exoplanet’s atmospheric composition is extracted from an observed spectrum. Traditional Bayesian sampling methods require extensive time and computation, involving algorithms that compare large numbers of known atmospheric models to the input spectral data. Runtimes are directly proportional to the number of parameters under consideration. These increased power and runtime requirements are difficult to accommodate in space missions where model size, speed, and power consumption are of particular importance. The use of traditional Bayesian sampling methods, therefore, compromise model complexity or sampling accuracy. The Atmospheric Retrievals conditional Generative Adversarial Network (ARcGAN) is a deep convolutional generative adversarial network that improves on the previous model’s speed and accuracy. We demonstrate the efficacy of artificial intelligence to quickly and reliably predict atmospheric parameters and present it as a viable alternative to slow and computationally heavy Bayesian methods. In addition to its broad applicability across instruments and planetary types, ARcGAN has been designed to function on low power application-specific integrated circuits. The application of edge computing to atmospheric retrievals allows for real or near-real-time quantification of atmospheric constituents at the instrument level. Additionally, edge computing provides both high-performance and power-efficient computing for AI applications, both of which are critical for space missions. With the edge computing chip implementation, ArcGAN serves as a strong basis for the development of a similar machine-learning algorithm to reduce the downlinked data volume from the Compact Ultraviolet to Visible Imaging Spectrometer (CUVIS) onboard the DAVINCI mission to Venus.

Keywords: deep learning, generative adversarial network, edge computing, atmospheric parameters retrieval

Procedia PDF Downloads 156

Authors: Zhenzhen Xu, Chunfu Shao, Shengyou Wang, Chunjiao Dong

Abstract:

To improve the satisfaction of residents' travel, this paper analyzes the characteristics and influencing factors of urban residents' travel behavior. First, a Multinominal Logit Model (MNL) model is built to analyze the characteristics of residents' travel behavior, reveal the influence of individual attributes, family attributes and travel characteristics on the choice of travel mode, and identify the significant factors. Then put forward suggestions for policy improvement. Finally, Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP) models are introduced to evaluate the policy effect. This paper selects Futian Street in Futian District, Shenzhen City for investigation and research. The results show that gender, age, education, income, number of cars owned, travel purpose, departure time, journey time, travel distance and times all have a significant influence on residents' choice of travel mode. Based on the above results, two policy improvement suggestions are put forward from reducing public transportation and non-motor vehicle travel time, and the policy effect is evaluated. Before the evaluation, the prediction effect of MNL, SVM and MLP models was evaluated. After parameter optimization, it was found that the prediction accuracy of the three models was 72.80%, 71.42%, and 76.42%, respectively. The MLP model with the highest prediction accuracy was selected to evaluate the effect of policy improvement. The results showed that after the implementation of the policy, the proportion of public transportation in plan 1 and plan 2 increased by 14.04% and 9.86%, respectively, while the proportion of private cars decreased by 3.47% and 2.54%, respectively. The proportion of car trips decreased obviously, while the proportion of public transport trips increased. It can be considered that the measures have a positive effect on promoting green trips and improving the satisfaction of urban residents, and can provide a reference for relevant departments to formulate transportation policies.

Keywords: neural network, travel characteristics analysis, transportation choice, travel sharing rate, traffic resource allocation

Procedia PDF Downloads 124

Authors: Zama Mahmood

Abstract:

The increasing demand of food management, monitoring of the crop growth and forecasting its yield well before harvest is very important. These days, yield assessment together with monitoring of crop development and its growth are being identified with the help of satellite and remote sensing images. Studies using remote sensing data along with field survey validation reported high correlation between vegetation indices and yield. With the development of remote sensing technique, the detection of crop and its mechanism using remote sensing data on regional or global scales have become popular topics in remote sensing applications. Punjab, specially the southern Punjab region is extremely favourable for wheat production. But measuring the exact amount of wheat production is a tedious job for the farmers and workers using traditional ground based measurements. However, remote sensing can provide the most real time information. In this study, using the Normalized Differentiate Vegetation Index (NDVI) indicator developed from Landsat satellite images, the yield of wheat has been estimated during the season of 2013-2014 for the agricultural area around Bahawalpur. The average yield of the wheat was found 35 kg/acre by analysing field survey data. The field survey data is in fair agreement with the NDVI values extracted from Landsat images. A correlation between wheat production (ton) and number of wheat pixels has also been calculated which is in proportional pattern with each other. Also a strong correlation between the NDVI and wheat area was found (R2=0.71) which represents the effectiveness of the remote sensing tools for crop monitoring and production estimation.

Keywords: landsat, NDVI, remote sensing, satellite images, yield

Procedia PDF Downloads 318

Authors: Naina Mahajan, Bikram Pal Kaur

Abstract:

The purpose of traffic accident analysis is to find the possible causes of an accident. Road accidents cannot be totally prevented but by suitable traffic engineering and management the accident rate can be reduced to a certain extent. This paper discusses the classification techniques C4.5 and ID3 using the WEKA Data mining tool. These techniques use on the NH (National highway) dataset. With the C4.5 and ID3 technique it gives best results and high accuracy with less computation time and error rate.

Keywords: C4.5, ID3, NH(National highway), WEKA data mining tool

Procedia PDF Downloads 320

Authors: Isaac Mugume, Charles Basalirwa, Daniel Waiswa, Mary Nsabagwa, Triphonia Jacob Ngailo, Joachim Reuder, Sch¨attler Ulrich, Musa Semujju

Abstract:

The Numerical weather prediction (NWP) models are considered powerful tools for guiding quantitative rainfall prediction. A couple of NWP models exist and are used at many operational weather prediction centers. This study considers two models namely the Consortium for Small–scale Modeling (COSMO) model and the Weather Research and Forecasting (WRF) model. It compares the models’ ability to predict rainfall over Uganda for the period 21st April 2013 to 10th May 2013 using the root mean square (RMSE) and the mean error (ME). In comparing the performance of the models, this study assesses their ability to predict light rainfall events and extreme rainfall events. All the experiments used the default parameterization configurations and with same horizontal resolution (7 Km). The results show that COSMO model had a tendency of largely predicting no rain which explained its under–prediction. The COSMO model (RMSE: 14.16; ME: -5.91) presented a significantly (p = 0.014) higher magnitude of error compared to the WRF model (RMSE: 11.86; ME: -1.09). However the COSMO model (RMSE: 3.85; ME: 1.39) performed significantly (p = 0.003) better than the WRF model (RMSE: 8.14; ME: 5.30) in simulating light rainfall events. All the models under–predicted extreme rainfall events with the COSMO model (RMSE: 43.63; ME: -39.58) presenting significantly higher error magnitudes than the WRF model (RMSE: 35.14; ME: -26.95). This study recommends additional diagnosis of the models’ treatment of deep convection over the tropics.

Keywords: comparative performance, the COSMO model, the WRF model, light rainfall events, extreme rainfall events

Procedia PDF Downloads 246

Authors: Marielcio Lacerda, Angelo Paulino, Elcio Shiguemori, Alvaro Damiao, Lamartine Guimaraes, Camila Anjos

Abstract:

Accurate information about the terrain is extremely important in disaster management activities or conflict. This paper proposes the use of the Unmanned Aircraft Systems (UAS) at the identification of Airdrop Zones (AZs) and Helicopter Landing Zones (HLZs). In this paper we consider the AZs the zones where troops or supplies are dropped by parachute, and HLZs areas where victims can be rescued. The use of digital image processing enables the automatic generation of an orthorectified mosaic and an actual Digital Surface Model (DSM). This methodology allows obtaining this fundamental information to the terrain’s comprehension post-disaster in a short amount of time and with good accuracy. In order to get the identification and classification of AZs and HLZs images from DJI drone, model Phantom 4 have been used. The images were obtained with the knowledge and authorization of the responsible sectors and were duly registered in the control agencies. The flight was performed on May 24, 2017, and approximately 1,300 images were obtained during approximately 1 hour of flight. Afterward, new attributes were generated by Feature Extraction (FE) from the original images. The use of multispectral images and complementary attributes generated independently from them increases the accuracy of classification. The attributes of this work include the Declivity Map and Principal Component Analysis (PCA). For the classification four distinct classes were considered: HLZ 1 – small size (18m x 18m); HLZ 2 – medium size (23m x 23m); HLZ 3 – large size (28m x 28m); AZ (100m x 100m). The Decision Tree method Random Forest (RF) was used in this work. RF is a classification method that uses a large collection of de-correlated decision trees. Different random sets of samples are used as sampled objects. The results of classification from each tree and for each object is called a class vote. The resulting classification is decided by a majority of class votes. In this case, we used 200 trees for the execution of RF in the software WEKA 3.8. The classification result was visualized on QGIS Desktop 2.12.3. Through the methodology used, it was possible to classify in the study area: 6 areas as HLZ 1, 6 areas as HLZ 2, 4 areas as HLZ 3; and 2 areas as AZ. It should be noted that an area classified as AZ covers the classifications of the other classes, and may be used as AZ, HLZ of large size (HLZ3), medium size (HLZ2) and small size helicopters (HLZ1). Likewise, an area classified as HLZ for large rotary wing aircraft (HLZ3) covers the smaller area classifications, and so on. It was concluded that images obtained through small UAV are of great use in calamity situations since they can provide data with high accuracy, with low cost, low risk and ease and agility in obtaining aerial photographs. This allows the generation, in a short time, of information about the features of the terrain in order to serve as an important decision support tool.

Keywords: disaster management, unmanned aircraft systems, helicopter landing zones, airdrop zones, random forest

Procedia PDF Downloads 162

Authors: Panatda Intanin, Sangutid Thongsawad, Chirapha Tannanonta, Todsaporn Fuangrod

Abstract:

Stereotactic radiosurgery (SRS) is a highly precision delivery technique that requires comprehensive quality assurance (QA) tests prior to treatment delivery. An isocenter of delivery beam plays a critical role that affect the treatment accuracy. The uncertainty of isocenter is traditionally accessed using circular cone equipment, Winston-Lutz (WL) phantom and film. This technique is considered time consuming and highly dependent on the observer. In this work, the development of multileaf collimator (MLC)-based isocenter verification tool using electronic portal imaging device (EPID) was proposed and evaluated. A mechanical isocenter alignment with ball bearing diameter 5 mm and circular cone diameter 10 mm fixed to gantry head defines the radiation field was set as the conventional WL test method. The conventional setup was to compare to the proposed setup; using MLC (10 x 10 mm) to define the radiation filed instead of cone. This represents more realistic delivery field than using circular cone equipment. The acquisition from electronic portal imaging device (EPID) and radiographic film were performed in both experiments. The gantry angles were set as following: 0°, 90°, 180° and 270°. A software tool was in-house developed using MATLAB/SIMULINK programming to determine the centroid of radiation field and shadow of WL phantom automatically. This presents higher accuracy than manual measurement. The deviation between centroid of both cone-based and MLC-based WL tests were quantified. To compare between film and EPID image, the deviation for all gantry angle was 0.26±0.19mm and 0.43±0.30 for cone-based and MLC-based WL tests. For the absolute deviation calculation on EPID images between cone and MLC-based WL test was 0.59±0.28 mm and the absolute deviation on film images was 0.14±0.13 mm. Therefore, the MLC-based isocenter verification using EPID present high sensitivity tool for SRS QA.

Keywords: isocenter verification, quality assurance, EPID, SRS

Procedia PDF Downloads 136

Authors: Richard Ren

Abstract:

Air pollution is a serious danger to international well-being and economies - it will kill an estimated 7 million people every year, costing world economies $2.6 trillion by 2060 due to sick days, healthcare costs, and reduced productivity. In the United States alone, 60,000 premature deaths are caused by poor air quality. For this reason, there is a crucial need to develop effective methods to forecast air quality, which can mitigate air pollution’s detrimental public health effects and associated costs by helping people plan ahead and avoid exposure. The goal of this study is to propose an artificial intelligence framework for predicting future air quality based on timing variables (i.e. season, weekday/weekend), future weather forecasts, as well as past pollutant and air quality measurements. The proposed framework utilizes multiple machine learning algorithms (logistic regression, random forest, neural network) with different specifications and averages the results of the three top-performing models to eliminate inaccuracies, weaknesses, and biases from any one individual model. Over time, the proposed framework uses new data to self-adjust model parameters and increase prediction accuracy. To demonstrate its applicability, a prototype of this framework was created to forecast air quality in Los Angeles, California using datasets from the RP4 weather data repository and EPA pollutant measurement data. The results showed good agreement between the framework’s predictions and real-life observations, with an overall 92% model accuracy. The combined model is able to predict more accurately than any of the individual models, and it is able to reliably forecast season-based variations in air quality levels. Top air quality predictor variables were identified through the measurement of mean decrease in accuracy. This study proposed and demonstrated the efficacy of a comprehensive air quality prediction framework leveraging multiple machine learning algorithms to overcome individual algorithm shortcomings. Future enhancements should focus on expanding and testing a greater variety of modeling techniques within the proposed framework, testing the framework in different locations, and developing a platform to automatically publish future predictions in the form of a web or mobile application. Accurate predictions from this artificial intelligence framework can in turn be used to save and improve lives by allowing individuals to protect their health and allowing governments to implement effective pollution control measures.Air pollution is a serious danger to international wellbeing and economies - it will kill an estimated 7 million people every year, costing world economies $2.6 trillion by 2060 due to sick days, healthcare costs, and reduced productivity. In the United States alone, 60,000 premature deaths are caused by poor air quality. For this reason, there is a crucial need to develop effective methods to forecast air quality, which can mitigate air pollution’s detrimental public health effects and associated costs by helping people plan ahead and avoid exposure. The goal of this study is to propose an artificial intelligence framework for predicting future air quality based on timing variables (i.e. season, weekday/weekend), future weather forecasts, as well as past pollutant and air quality measurements. The proposed framework utilizes multiple machine learning algorithms (logistic regression, random forest, neural network) with different specifications and averages the results of the three top-performing models to eliminate inaccuracies, weaknesses, and biases from any one individual model. Over time, the proposed framework uses new data to self-adjust model parameters and increase prediction accuracy. To demonstrate its applicability, a prototype of this framework was created to forecast air quality in Los Angeles, California using datasets from the RP4 weather data repository and EPA pollutant measurement data. The results showed good agreement between the framework’s predictions and real-life observations, with an overall 92% model accuracy. The combined model is able to predict more accurately than any of the individual models, and it is able to reliably forecast season-based variations in air quality levels. Top air quality predictor variables were identified through the measurement of mean decrease in accuracy. This study proposed and demonstrated the efficacy of a comprehensive air quality prediction framework leveraging multiple machine learning algorithms to overcome individual algorithm shortcomings. Future enhancements should focus on expanding and testing a greater variety of modeling techniques within the proposed framework, testing the framework in different locations, and developing a platform to automatically publish future predictions in the form of a web or mobile application. Accurate predictions from this artificial intelligence framework can in turn be used to save and improve lives by allowing individuals to protect their health and allowing governments to implement effective pollution control measures.Air pollution is a serious danger to international wellbeing and economies - it will kill an estimated 7 million people every year, costing world economies $2.6 trillion by 2060 due to sick days, healthcare costs, and reduced productivity. In the United States alone, 60,000 premature deaths are caused by poor air quality. For this reason, there is a crucial need to develop effective methods to forecast air quality, which can mitigate air pollution’s detrimental public health effects and associated costs by helping people plan ahead and avoid exposure. The goal of this study is to propose an artificial intelligence framework for predicting future air quality based on timing variables (i.e. season, weekday/weekend), future weather forecasts, as well as past pollutant and air quality measurements. The proposed framework utilizes multiple machine learning algorithms (logistic regression, random forest, neural network) with different specifications and averages the results of the three top-performing models to eliminate inaccuracies, weaknesses, and biases from any one individual model. Over time, the proposed framework uses new data to self-adjust model parameters and increase prediction accuracy. To demonstrate its applicability, a prototype of this framework was created to forecast air quality in Los Angeles, California using datasets from the RP4 weather data repository and EPA pollutant measurement data. The results showed good agreement between the framework’s predictions and real-life observations, with an overall 92% model accuracy. The combined model is able to predict more accurately than any of the individual models, and it is able to reliably forecast season-based variations in air quality levels. Top air quality predictor variables were identified through the measurement of mean decrease in accuracy. This study proposed and demonstrated the efficacy of a comprehensive air quality prediction framework leveraging multiple machine learning algorithms to overcome individual algorithm shortcomings. Future enhancements should focus on expanding and testing a greater variety of modeling techniques within the proposed framework, testing the framework in different locations, and developing a platform to automatically publish future predictions in the form of a web or mobile application. Accurate predictions from this artificial intelligence framework can in turn be used to save and improve lives by allowing individuals to protect their health and allowing governments to implement effective pollution control measures.

Keywords: air quality prediction, air pollution, artificial intelligence, machine learning algorithms

Procedia PDF Downloads 107

Authors: Jinjun Jiang, Lianzhong Chen, Rui Xu

Abstract:

The accurate of model attitude angel plays an important role on the aerodynamic test results in the wind tunnel test. The original method applies the spherical coordinate system transformation to obtain attitude angel calculation.The model attitude angel is obtained by coordinate transformation and spherical surface mapping applying the nominal attitude angel (the balance attitude angel in the wind tunnel coordinate system) indicated by the mechanism. First, the coordinate transformation of this method is not only complex but also difficult to establish the transformed relationship between the space coordinate systems especially after many steps of coordinate transformation, moreover it cannot realize the iterative calculation of the interference relationship between attitude angels; Second, during the calculate process to solve the problem the arc is approximately used to replace the straight line, the angel for the tangent value, and the inverse trigonometric function is applied. Therefore, in the calculation of attitude angel, the process is complex and inaccurate, which can be solved approximately when calculating small attack angel. However, with the advancing development of modern aerodynamic unsteady research, the aircraft tends to develop high or super large attack angel and unsteadyresearch field.According to engineering practice and vector theory, the concept of vector angel coordinate systemis proposed for the first time, and the vector angel coordinate system of attitude angel is established.With the iterative correction calculation and avoiding the problem of approximate and inverse trigonometric function solution, the model attitude calculation process is carried out in detail, which validates that the calculation accuracy and accuracy of model attitude angels are improved.Based on engineering and theoretical methods, a vector angel coordinate systemis established for the first time, which gives the transformation and angel definition relations between different flight attitude coordinate systems, that can accurately calculate the attitude angel of the corresponding coordinate systemand determine its direction, especially in the channel coupling calculation, the calculation of the attitude angel between the coordinate systems is only related to the angel, and has nothing to do with the change order s of the coordinate system, whichsimplifies the calculation process.

Keywords: attitude angel, angel vector coordinate system, iterative calculation, spherical coordinate system, wind tunnel test

Procedia PDF Downloads 117

Authors: Li Xu, Huanbao Jiang, Zhenfei Huang, Lailai Zhang

Abstract:

In recent years, as global warming, the sea-ice extent of North Arctic undergoes an evident decrease and Arctic channel has attracted the attention of shipping industry. Ice crystals existing in the seawater of Arctic channel which enter the seawater system of the ship with the seawater were found blocking the seawater pipe. The appearance of cooler paralysis, auxiliary machine error and even ship power system paralysis may be happened if seriously. In order to reduce the effect of high temperature in auxiliary equipment, seawater system will use external ice-water to participate in the cooling cycle and achieve the state of its flow. The distribution of ice crystals in seawater pipe can be achieved. As the ice slurry system is solid liquid two-phase system, the flow process of ice-water mixture is very complex and diverse. In this paper, the flow process in seawater pipe of ice slurry is simulated with fluid dynamics simulation software based on k-ε turbulence model. As the ice packing fraction is a key factor effecting the distribution of ice crystals, the influence of ice packing fraction on the flowing process of ice slurry is analyzed. In this work, the simulation results show that as the ice packing fraction is relatively large, the distribution of ice crystals is uneven in the flowing process of the seawater which has such disadvantage as increase the possibility of blocking, that will provide scientific forecasting methods for the forming of ice block in seawater piping system. It has important significance for the reliability of the operating of polar ships in the future.

Keywords: ice slurry, seawater pipe, ice packing fraction, numerical simulation

Procedia PDF Downloads 353

Authors: Tigabu Dagne Akal

Abstract:

While advances in computer and communications technology have made the network ubiquitous, they have also rendered networked systems vulnerable to malicious attacks devised from a distance. These attacks or intrusions start with attackers infiltrating a network through a vulnerable host and then launching further attacks on the local network or Intranet. Nowadays, system administrators and network professionals can attempt to prevent such attacks by developing intrusion detection tools and systems using data mining technology. In this study, the experiments were conducted following the Knowledge Discovery in Database Process Model. The Knowledge Discovery in Database Process Model starts from selection of the datasets. The dataset used in this study has been taken from Massachusetts Institute of Technology Lincoln Laboratory. After taking the data, it has been pre-processed. The major pre-processing activities include fill in missed values, remove outliers; resolve inconsistencies, integration of data that contains both labelled and unlabelled datasets, dimensionality reduction, size reduction and data transformation activity like discretization tasks were done for this study. A total of 21,533 intrusion records are used for training the models. For validating the performance of the selected model a separate 3,397 records are used as a testing set. For building a predictive model for intrusion detection J48 decision tree and the Naïve Bayes algorithms have been tested as a classification approach for both with and without feature selection approaches. The model that was created using 10-fold cross validation using the J48 decision tree algorithm with the default parameter values showed the best classification accuracy. The model has a prediction accuracy of 96.11% on the training datasets and 93.2% on the test dataset to classify the new instances as normal, DOS, U2R, R2L and probe classes. The findings of this study have shown that the data mining methods generates interesting rules that are crucial for intrusion detection and prevention in the networking industry. Future research directions are forwarded to come up an applicable system in the area of the study.

Keywords: intrusion detection, data mining, computer science, data mining

Procedia PDF Downloads 285

Authors: Mohamed El Horri, Amine Mouden, Reda Messaoudi, Mohamed Chekkal, Driss Benlaldj, Malika Baghdadi, Lahcene Benmahdi, Fatima Seghier

Abstract:

Background/aim: Recently, the Von Willebrand factor antigen (vWF-Ag)has been identified as a new marker of portal hypertension (PH) and its complications. Few studies talked about its role in the prediction of esophageal varices. VWF-Ag is considered a non-invasive approach, In order to avoid the endoscopic burden, cost, drawbacks, unpleasant and repeated examinations to the patients. In our study, we aimed to evaluate the ability of this marker in the prediction of another complication of portal hypertension, which is portal hypertensive gastropathy (PHG), the one that is diagnosed also by endoscopic tools. Patients and methods: It is about a prospective study, which include 124 cirrhotic patients with no history of bleeding who underwent screening endoscopy for PH-related complications like esophageal varices (EVs) and PHG. Routine biological tests were performed as well as the VWF-Ag testing by both ELFA and Immunoturbidimetric techniques. The diagnostic performance of our marker was assessed using sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and receiver operating characteristic curves. Results: 124 patients were enrolled in this study, with a mean age of 58 years [CI: 55 – 60 years] and a sex ratio of 1.17. Viral etiologies were found in 50% of patients. Screening endoscopy revealed the presence of PHG in 20.2% of cases, while for EVsthey were found in 83.1% of cases. VWF-Ag levels, were significantly increased in patients with PHG compared to those who have not: 441% [CI: 375 – 506], versus 279% [CI: 253 – 304], respectively (p <0.0001). Using the area under the receiver operating characteristic curve (AUC), vWF-Ag was a good predictor for the presence of PHG. With a value higher than 320% and an AUC of 0.824, VWF-Ag had an 84% sensitivity, 74% specificity, 44.7% positive predictive value, 94.8% negative predictive value, and 75.8% diagnostic accuracy. Conclusion: VWF-Ag is a good non-invasive low coast marker for excluding the presence of PHG in patients with liver cirrhosis. Using this marker as part of a selective screening strategy might reduce the need for endoscopic screening and the coast of the management of these kinds of patients.

Keywords: von willebrand factor, portal hypertensive gastropathy, prediction, liver cirrhosis

Procedia PDF Downloads 187

Authors: Rao Fu, Jie Chen, Shutian Zeng, Yiping Zhuang, Agus Sudjianto

Abstract:

Generative Adversarial Net (GAN) has proved to be a powerful machine learning tool in image data analysis and generation. In this paper, we propose to use Conditional Generative Adversarial Net (CGAN) to learn and simulate time series data. The conditions include both categorical and continuous variables with different auxiliary information. Our simulation studies show that CGAN has the capability to learn different types of normal and heavy-tailed distributions, as well as dependent structures of different time series. It also has the capability to generate conditional predictive distributions consistent with training data distributions. We also provide an in-depth discussion on the rationale behind GAN and the neural networks as hierarchical splines to establish a clear connection with existing statistical methods of distribution generation. In practice, CGAN has a wide range of applications in market risk and counterparty risk analysis: it can be applied to learn historical data and generate scenarios for the calculation of Value-at-Risk (VaR) and Expected Shortfall (ES), and it can also predict the movement of the market risk factors. We present a real data analysis including a backtesting to demonstrate that CGAN can outperform Historical Simulation (HS), a popular method in market risk analysis to calculate VaR. CGAN can also be applied in economic time series modeling and forecasting. In this regard, we have included an example of hypothetical shock analysis for economic models and the generation of potential CCAR scenarios by CGAN at the end of the paper.

Keywords: conditional generative adversarial net, market and credit risk management, neural network, time series

Procedia PDF Downloads 125

Authors: Hazem A. Abdelfattah, Salia H. Khafaji

Abstract:

Apparel industry needs to direct scientific researches to develop it , and because of the importance of a men’s suit jacket industry, the study of the basics of men’s jacket pattern making requires a high degree of accuracy and efficiency which contain a lot of technical and skill aspects to give the jacket a drape, comfort and good fitting , prompting researchers to think about the use of men’s mannequin with sizes (M-L-XL) to devise a method to draft a paper pattern for the men's suit jacket to use it in the industry easily and quickly and achieve the required good fitting.

Keywords: draping, pattern, men, jacket

Procedia PDF Downloads 333

Authors: Manavjit Singh Dhindsa, Ursula Das, Kshirasagar Naik, Marzia Zaman, Richard Purcell, Srinivas Sampalli, Abdul Mutakabbir, Chung-Horng Lung, Thambirajah Ravichandran

Abstract:

These days, wildland fires, also known as forest fires, are more prevalent than ever. Wildfires have major repercussions that affect ecosystems, communities, and the environment in several ways. Wildfires lead to habitat destruction and biodiversity loss, affecting ecosystems and causing soil erosion. They also contribute to poor air quality by releasing smoke and pollutants that pose health risks, especially for individuals with respiratory conditions. Wildfires can damage infrastructure, disrupt communities, and cause economic losses. The economic impact of firefighting efforts, combined with their direct effects on forestry and agriculture, causes significant financial difficulties for the areas impacted. This research explores different forest fire spread models and presents a comprehensive review of various techniques and methodologies used in the field. A forest fire spread model is a computational or mathematical representation that is used to simulate and predict the behavior of a forest fire. By applying scientific concepts and data from empirical studies, these models attempt to capture the intricate dynamics of how a fire spreads, taking into consideration a variety of factors like weather patterns, topography, fuel types, and environmental conditions. These models assist authorities in understanding and forecasting the potential trajectory and intensity of a wildfire. Emphasizing the need for a comprehensive understanding of wildfire dynamics, this research explores the approaches, assumptions, and findings derived from various models. By using a comparison approach, a critical analysis is provided by identifying patterns, strengths, and weaknesses among these models. The purpose of the survey is to further wildfire research and management techniques. Decision-makers, researchers, and practitioners can benefit from the useful insights that are provided by synthesizing established information. Fire spread models provide insights into potential fire behavior, facilitating authorities to make informed decisions about evacuation activities, allocating resources for fire-fighting efforts, and planning for preventive actions. Wildfire spread models are also useful in post-wildfire mitigation strategies as they help in assessing the fire's severity, determining high-risk regions for post-fire dangers, and forecasting soil erosion trends. The analysis highlights the importance of customized modeling approaches for various circumstances and promotes our understanding of the way forest fires spread. Some of the known models in this field are Rothermel’s wildland fuel model, FARSITE, WRF-SFIRE, FIRETEC, FlamMap, FSPro, cellular automata model, and others. The key characteristics that these models consider include weather (includes factors such as wind speed and direction), topography (includes factors like landscape elevation), and fuel availability (includes factors like types of vegetation) among other factors. The models discussed are physics-based, data-driven, or hybrid models, also utilizing ML techniques like attention-based neural networks to enhance the performance of the model. In order to lessen the destructive effects of forest fires, this initiative aims to promote the development of more precise prediction tools and effective management techniques. The survey expands its scope to address the practical needs of numerous stakeholders. Access to enhanced early warning systems enables decision-makers to take prompt action. Emergency responders benefit from improved resource allocation strategies, strengthening the efficacy of firefighting efforts.

Keywords: artificial intelligence, deep learning, forest fire management, fire risk assessment, fire simulation, machine learning, remote sensing, wildfire modeling

Procedia PDF Downloads 68

Authors: Amira Masood, Zainah Alshouri, Noor Bantan, Samira Kutbi

Abstract:

Artificial intelligence is continuously becoming more advanced and more widespread and is present in many of our day-to-day lives as a means of assistance in numerous different fields. A growing number of people, companies, and corporations are utilizing Canva and its AI tools as a method of quick and easy media production. Hence, in order to test the integrity of the rapid growth of AI, this paper will explore the usefulness of Canva's advanced design features as well as their accuracy by determining user satisfaction through a survey-based research approach and by investigating whether or not AI is successful enough that it eliminates the need for human alterations.

Keywords: artificial intelligence, canva, features, users, satisfaction

Procedia PDF Downloads 89

Authors: Cyrillus Prabandana

Abstract:

Supply chain optimisation requires multiple factors as consideration or constraints. These factors are including but not limited to demand forecasting, raw material fulfilment, production capacity, inventory level, facilities locations, transportation means, and manpower availability. By knowing all manageable factors involved and assuming the uncertainty with pre-defined percentage factors, an integrated supply chain model could be developed to manage various business scenarios. This paper analyse the utilisation of geographical point of view to develop an integrated supply chain network model to optimise the distribution of finished product appropriately according to forecasted demand and available supply. The supply chain optimisation model shows that small change in one supply chain constraint is possible to largely impact other constraints, and the new information from the model should be able to support the decision making process. The model was focused on three areas, i.e. raw material fulfilment, production capacity and finished products transportation. To validate the model suitability, it was implemented in a project aimed to optimise the concrete supply chain in a mining location. The high level of operations complexity and involvement of multiple stakeholders in the concrete supply chain is believed to be sufficient to give the illustration of the larger scope. The implementation of this geographical supply chain network modeling resulted an optimised concrete supply chain from raw material fulfilment until finished products distribution to each customer, which indicated by lower percentage of missed concrete order fulfilment to customer.

Keywords: decision making, geographical supply chain modeling, supply chain optimisation, supply chain

Procedia PDF Downloads 339

Authors: Evangelos Tsakalos, Maria Kazantzaki, Eleni Filippaki, Yannis Bassiakos

Abstract:

The coastal areas of the Mediterranean have been extensively affected by the transgressive event that followed the Last Glacial Maximum, with many studies conducted regarding the stratigraphic configuration of coastal sediments around the Mediterranean. The coastal zone of the Messologi area, western Greece, consists of low relief beaches containing low cliffs and eroded dunes, a fact which, in combination with the rising sea level and tectonic subsidence of the area, has led to substantial coastal. Coastal vulnerability assessment is a useful means of identifying areas of coastline that are vulnerable to impacts of climate change and coastal processes, highlighting potential problem areas. Commonly, coastal vulnerability assessment takes the form of an ‘index’ that quantifies the relative vulnerability along a coastline. Here we make use of the coastal vulnerability index (CVI) methodology by Thieler and Hammar-Klose, by considering geological features, coastal slope, relative sea-level change, shoreline erosion/accretion rates, and mean significant wave height as well as mean tide range to assess the present-day vulnerability of the coastal zone of Messologi area. In light of this, an impact assessment is performed under three different sea level rise scenarios, and adaptation measures to control climate change events are proposed. This study contributes toward coastal zone management practices in low-lying areas that have little data information, assisting decision-makers in adopting best adaptations options to overcome sea level rise impact on vulnerable areas similar to the coastal zone of Messologi.

Keywords: coastal vulnerability index, coastal erosion, sea level rise, GIS

Procedia PDF Downloads 164

Authors: Changju Lee, Seungmo Ku, Sondo Kim, Woojin Chang

Abstract:

How to detecting financial bubble? Addressing this simple question has been the focus of a vast amount of empirical research spanning almost half a century. However, financial bubble is hard to observe and varying over the time; there needs to be more research on this area. In this paper, we used abnormal difference between common stocks price and those preferred stocks price to explain financial bubble. First, we proposed the ‘W-index’ which indicates spread between common stocks and those preferred stocks in stock market. Second, to prove that this ‘W-index’ is valid for measuring financial bubble, we showed that there is an inverse relationship between this ‘W-index’ and S&P500 rate of return. Specifically, our hypothesis is that when ‘W-index’ is comparably higher than other periods, financial bubbles are added up in stock market and vice versa; according to our hypothesis, if investors made long term investments when ‘W-index’ is high, they would have negative rate of return; however, if investors made long term investments when ‘W-index’ is low, they would have positive rate of return. By comparing correlation values and adjusted R-squared values of between W-index and S&P500 return, VIX index and S&P500 return, and TED index and S&P500 return, we showed only W-index has significant relationship between S&P500 rate of return. In addition, we figured out how long investors should hold their investment position regard the effect of financial bubble. Using this W-index, investors could measure financial bubble in the market and invest with low risk.

Keywords: financial bubble detection, future return, forecasting, pairs trading, preferred stocks

Procedia PDF Downloads 352

Authors: Wanchalerm Patanacharoenwong, Panaya Sudta, Prachya Bumrungkun

Abstract:

The paper focuses on the development of a system that combines Internet of Things (IoT) technologies and deep learning algorithms for anomaly detection in residential Electric Vehicle (EV) charging in smart cities. With the increasing number of EVs, ensuring efficient and reliable charging systems has become crucial. The aim of this research is to develop an integrated IoT and deep learning system for detecting anomalies in residential EV charging and enhancing EV load profiling and event detection in smart cities. This approach utilizes IoT devices equipped with infrared cameras to collect thermal images and household EV charging profiles from the database of Thailand utility, subsequently transmitting this data to a cloud database for comprehensive analysis. The methodology includes the use of advanced deep learning techniques such as Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM) algorithms. IoT devices equipped with infrared cameras are used to collect thermal images and EV charging profiles. The data is transmitted to a cloud database for comprehensive analysis. The researchers also utilize feature-based Gaussian mixture models for EV load profiling and event detection. Moreover, the research findings demonstrate the effectiveness of the developed system in detecting anomalies and critical profiles in EV charging behavior. The system provides timely alarms to users regarding potential issues and categorizes the severity of detected problems based on a health index for each charging device. The system also outperforms existing models in event detection accuracy. This research contributes to the field by showcasing the potential of integrating IoT and deep learning techniques in managing residential EV charging in smart cities. The system ensures operational safety and efficiency while also promoting sustainable energy management. The data is collected using IoT devices equipped with infrared cameras and is stored in a cloud database for analysis. The collected data is then analyzed using RNN, LSTM, and feature-based Gaussian mixture models. The approach includes both EV load profiling and event detection, utilizing a feature-based Gaussian mixture model. This comprehensive method aids in identifying unique power consumption patterns among EV owners and outperforms existing models in event detection accuracy. In summary, the research concludes that integrating IoT and deep learning techniques can effectively detect anomalies in residential EV charging and enhance EV load profiling and event detection accuracy. The developed system ensures operational safety and efficiency, contributing to sustainable energy management in smart cities.

Keywords: cloud computing framework, recurrent neural networks, long short-term memory, Iot, EV charging, smart grids

Procedia PDF Downloads 48

Authors: Justas Dilys

Abstract:

ThispaperaddressestheimplementationandoptimizationofanUnscentedKalmanFilter(UKF) for the Permanent Magnet Synchronous Motor (PMSM) sensorless control using an ARM Cortex- M3 microcontroller. A various optimization levels based on arithmetic calculation reduction was implemented in ARM Cortex-M3 microcontroller. The execution time of UKF estimator was up to 90µs without loss of accuracy. Moreover, simulation studies on the Unscented Kalman filters are carried out using Matlab to explore the usability of the UKF in a sensorless PMSMdrive.

Keywords: unscented kalman filter, ARM, PMSM, implementation

Procedia PDF Downloads 147

Authors: Mpho Mokoatle, Darlington Mapiye, James Mashiyane, Stephanie Muller, Gciniwe Dlamini

Abstract:

Great advances in high-throughput sequencing technologies have resulted in availability of huge amounts of sequencing data in public and private repositories, enabling a holistic understanding of complex biological phenomena. Sequence data are used for a wide range of applications such as gene annotations, expression studies, personalized treatment and precision medicine. However, this rapid growth in sequence data poses a great challenge which calls for novel data processing and analytic methods, as well as huge computing resources. In this work, a machine and statistical learning approach for DNA sequence classification based on $k$-mer representation of sequence data is proposed. The approach is tested using whole genome sequences of Mycobacterium tuberculosis (MTB) isolates to (i) reduce the size of genomic sequence data, (ii) identify an optimum size of k-mers and utilize it to build classification models, (iii) predict the phenotype from whole genome sequence data of a given bacterial isolate, and (iv) demonstrate computing challenges associated with the analysis of whole genome sequence data in producing interpretable and explainable insights. The classification models were trained on 104 whole genome sequences of MTB isoloates. Cluster analysis showed that k-mers maybe used to discriminate phenotypes and the discrimination becomes more concise as the size of k-mers increase. The best performing classification model had a k-mer size of 10 (longest k-mer) an accuracy, recall, precision, specificity, and Matthews Correlation coeffient of 72.0%, 80.5%, 80.5%, 63.6%, and 0.4 respectively. This study provides a comprehensive approach for resampling whole genome sequencing data, objectively selecting a k-mer size, and performing classification for phenotype prediction. The analysis also highlights the importance of increasing the k-mer size to produce more biological explainable results, which brings to the fore the interplay that exists amongst accuracy, computing resources and explainability of classification results. However, the analysis provides a new way to elucidate genetic information from genomic data, and identify phenotype relationships which are important especially in explaining complex biological mechanisms.

Keywords: AWD-LSTM, bootstrapping, k-mers, next generation sequencing

Procedia PDF Downloads 152