Search results for: algorithms

Authors: Yindi Zhao, Yun Zhang, Silu Xia, Lixin Wu

Abstract:

The contours and areas of water surfaces, especially lakes, often change due to natural disasters and construction activities. It is an effective way to extract and update water contours from satellite images using image processing algorithms. However, to produce optimal water surface contours that are close to true boundaries is still a challenging task. This paper compares the performances of three different level set models, including the Chan-Vese (CV) model, the signed pressure force (SPF) model, and the region-scalable fitting (RSF) energy model for extracting lake contours. After experiment testing, it is indicated that the RSF model, in which a region-scalable fitting (RSF) energy functional is defined and incorporated into a variational level set formulation, is superior to CV and SPF, and it can get desirable contour lines when there are “holes” in the regions of waters, such as the islands in the lake. Therefore, the RSF model is applied to extracting lake contours from Landsat satellite images. Four temporal Landsat satellite images of the years of 2000, 2005, 2010, and 2014 are used in our study. All of them were acquired in May, with the same path/row (121/036) covering Xuzhou City, Jiangsu Province, China. Firstly, the near infrared (NIR) band is selected for water extraction. Image registration is conducted on NIR bands of different temporal images for information update, and linear stretching is also done in order to distinguish water from other land cover types. Then for the first temporal image acquired in 2000, lake contours are extracted via the RSF model with initialization of user-defined rectangles. Afterwards, using the lake contours extracted the previous temporal image as the initialized values, lake contours are updated for the current temporal image by means of the RSF model. Meanwhile, the changed and unchanged lakes are also detected. The results show that great changes have taken place in two lakes, i.e. Dalong Lake and Panan Lake, and RSF can actually extract and effectively update lake contours using multi-temporal satellite image.

Keywords: level set model, multi-temporal image, lake contour extraction, contour update

Procedia PDF Downloads 335

Authors: O’Day Luke

Abstract:

Purpose: To validate virtual non-contrast reconstructions generated from dual-layer spectral computed tomography (DL-CT) data as an alternative for the acquisition of a dedicated true non-contrast dataset during multiphase contrast studies. Material and methods: Thirty-three patients underwent a routine multiphase clinical CT examination, using Dual-Layer Spectral CT, from March to August 2021. True non-contrast (TNC) and virtual non-contrast (VNC) datasets, generated from both portal venous and arterial phase imaging were evaluated. For every patient in both true and virtual non-contrast datasets, a region-of-interest (ROI) was defined in aorta, liver, fluid (i.e. gallbladder, urinary bladder), kidney, muscle, fat and spongious bone, resulting in 693 ROIs. Differences in attenuation for VNC and TNV images were compared, both separately and combined. Consistency between VNC reconstructions obtained from the arterial and portal venous phase was evaluated. Results: Comparison of CT density (HU) on the VNC and TNC images showed a high correlation. The mean difference between TNC and VNC images (excluding bone results) was 5.5 ± 9.1 HU and > 90% of all comparisons showed a difference of less than 15 HU. For all tissues but spongious bone, the mean absolute difference between TNC and VNC images was below 10 HU. VNC images derived from the arterial and the portal venous phase showed a good correlation in most tissue types. The aortic attenuation was somewhat dependent however on which dataset was used for reconstruction. Bone evaluation with VNC datasets continues to be a problem, as spectral CT algorithms are currently poor in differentiating bone and iodine. Conclusion: Given the increasing availability of DL-CT and proven accuracy of virtual non-contrast processing, VNC is a promising tool for generating additional data during routine contrast-enhanced studies. This study shows the utility of virtual non-contrast scans as an alternative for true non-contrast studies during multiphase CT, with potential for dose reduction, without loss of diagnostic information.

Keywords: dual-layer spectral computed tomography, virtual non-contrast, true non-contrast, clinical comparison

Procedia PDF Downloads 112

Authors: Sokkhey Phauk, Takeo Okazaki

Abstract:

The challenging task in educational institutions is to maximize the high performance of students and minimize the failure rate of poor-performing students. An effective method to leverage this task is to know student learning patterns with highly influencing factors and get an early prediction of student learning outcomes at the timely stage for setting up policies for improvement. Educational data mining (EDM) is an emerging disciplinary field of data mining, statistics, and machine learning concerned with extracting useful knowledge and information for the sake of improvement and development in the education environment. The study is of this work is to propose techniques in EDM and integrate it into a web-based system for predicting poor-performing students. A comparative study of prediction models is conducted. Subsequently, high performing models are developed to get higher performance. The hybrid random forest (Hybrid RF) produces the most successful classification. For the context of intervention and improving the learning outcomes, a feature selection method MICHI, which is the combination of mutual information (MI) and chi-square (CHI) algorithms based on the ranked feature scores, is introduced to select a dominant feature set that improves the performance of prediction and uses the obtained dominant set as information for intervention. By using the proposed techniques of EDM, an academic performance prediction system (APPS) is subsequently developed for educational stockholders to get an early prediction of student learning outcomes for timely intervention. Experimental outcomes and evaluation surveys report the effectiveness and usefulness of the developed system. The system is used to help educational stakeholders and related individuals for intervening and improving student performance.

Keywords: academic performance prediction system, educational data mining, dominant factors, feature selection method, prediction model, student performance

Procedia PDF Downloads 85

Authors: Gianpaolo Ghiani, Emanuela Guerriero, Emanuele Manni, Alessandro Romano

Abstract:

In this article, we deal with a variant of the classical course timetabling problem that has a practical application in many areas of education. In particular, in this paper we are interested in high schools remedial courses. The purpose of such courses is to provide under-prepared students with the skills necessary to succeed in their studies. In particular, a student might be under prepared in an entire course, or only in a part of it. The limited availability of funds, as well as the limited amount of time and teachers at disposal, often requires schools to choose which courses and/or which teaching units to activate. Thus, schools need to model the training offer and the related timetabling, with the goal of ensuring the highest possible teaching quality, by meeting the above-mentioned financial, time and resources constraints. Moreover, there are some prerequisites between the teaching units that must be satisfied. We first present a Mixed-Integer Programming (MIP) model to solve this problem to optimality. However, the presence of many peculiar constraints contributes inevitably in increasing the complexity of the mathematical model. Thus, solving it through a general purpose solver may be performed for small instances only, while solving real-life-sized instances of such model requires specific techniques or heuristic approaches. For this purpose, we also propose a heuristic approach, in which we make use of a fast constructive procedure to obtain a feasible solution. To assess our exact and heuristic approaches we perform extensive computational results on both real-life instances (obtained from a high school in Lecce, Italy) and randomly generated instances. Our tests show that the MIP model is never solved to optimality, with an average optimality gap of 57%. On the other hand, the heuristic algorithm is much faster (in about the 50% of the considered instances it converges in approximately half of the time limit) and in many cases allows achieving an improvement on the objective function value obtained by the MIP model. Such an improvement ranges between 18% and 66%.

Keywords: heuristic, MIP model, remedial course, school, timetabling

Procedia PDF Downloads 581

Authors: Andrew R. Willis, Kevin Brink, Kathleen Dipple

Abstract:

This article describes a software package called ROS-georegistration intended for use with the robot operating system (ROS) and the Gazebo 3D simulation environment. ROSgeoregistration provides tools for the simulation, test, and deployment of aerial georegistration algorithms and is available at github.com/uncc-visionlab/rosgeoregistration. A model creation package is provided which downloads multi-spectral images from the Google Earth Engine database and, if necessary, incorporates these images into a single, possibly very large, reference image. Additionally a Gazebo plugin which uses the real-time sensor pose and image formation model to generate simulated imagery using the specified reference image is provided along with related plugins for UAV relevant data. The novelty of this work is threefold: (1) this is the first system to link the massive multi-spectral imaging database of Google’s Earth Engine to the Gazebo simulator, (2) this is the first example of a system that can simulate geospatially and radiometrically accurate imagery from multiple sensor views of the same terrain region, and (3) integration with other UAS tools creates a new holistic UAS simulation environment to support UAS system and subsystem development where real-world testing would generally be prohibitive. Sensed imagery and ground truth registration information is published to client applications which can receive imagery synchronously with telemetry from other payload sensors, e.g., IMU, GPS/GNSS, barometer, and windspeed sensor data. To highlight functionality, we demonstrate ROSgeoregistration for simulating Electro-Optical (EO) and Synthetic Aperture Radar (SAR) image sensors and an example use case for developing and evaluating image-based UAS position feedback, i.e., pose for image-based Guidance Navigation and Control (GNC) applications.

Keywords: EO-to-EO, EO-to-SAR, flight simulation, georegistration, image generation, robot operating system, vision-based navigation

Procedia PDF Downloads 78

Authors: Hung-Sheng Lin, Cheng-Hsuan Li

Abstract:

Over the past few years, kernel-based algorithms have been widely used to extend some linear feature extraction methods such as principal component analysis (PCA), linear discriminate analysis (LDA), and nonparametric weighted feature extraction (NWFE) to their nonlinear versions, kernel principal component analysis (KPCA), generalized discriminate analysis (GDA), and kernel nonparametric weighted feature extraction (KNWFE), respectively. These nonlinear feature extraction methods can detect nonlinear directions with the largest nonlinear variance or the largest class separability based on the given kernel function. Moreover, they have been applied to improve the target detection or the image classification of hyperspectral images. The double nearest proportion feature extraction (DNP) can effectively reduce the overlap effect and have good performance in hyperspectral image classification. The DNP structure is an extension of the k-nearest neighbor technique. For each sample, there are two corresponding nearest proportions of samples, the self-class nearest proportion and the other-class nearest proportion. The term “nearest proportion” used here consider both the local information and other more global information. With these settings, the effect of the overlap between the sample distributions can be reduced. Usually, the maximum likelihood estimator and the related unbiased estimator are not ideal estimators in high dimensional inference problems, particularly in small data-size situation. Hence, an improved estimator by shrinkage estimation (regularization) is proposed. Based on the DNP structure, LDA is included as a special case. In this paper, the kernel method is applied to extend DNP to kernel-based DNP (KDNP). In addition to the advantages of DNP, KDNP surpasses DNP in the experimental results. According to the experiments on the real hyperspectral image data sets, the classification performance of KDNP is better than that of PCA, LDA, NWFE, and their kernel versions, KPCA, GDA, and KNWFE.

Keywords: feature extraction, kernel method, double nearest proportion feature extraction, kernel double nearest feature extraction

Procedia PDF Downloads 299

Authors: Anderson S. Fonseca, Italo F. S. Da Silva, Robert D. A. Santos, Mayara G. Da Silva, Pedro H. C. Vieira, Antonio M. S. Sobrinho, Victor H. B. Lemos, Petterson S. Diniz, Anselmo C. Paiva, Eliana M. G. Monteiro

Abstract:

In Brazil, the National Electric Energy Agency (ANEEL) establishes that electrical energy companies are responsible for measuring and billing their customers. Among these regulations, it’s defined that a company must bill your customers within 27-33 days. If a relocation or a change of period is required, the consumer must be notified in writing, in advance of a billing period. To make it easier to organize a workday’s measurements, these companies create a reading plan. These plans consist of grouping customers into reading groups, which are visited by an employee responsible for measuring consumption and billing. The creation process of a plan efficiently and optimally is a capacitated clustering problem with constraints related to homogeneity and compactness, that is, the employee’s working load and the geographical position of the consuming unit. This process is a work done manually by several experts who have experience in the geographic formation of the region, which takes a large number of days to complete the final planning, and because it’s human activity, there is no guarantee of finding the best optimization for planning. In this paper, the GBKMeans method presents a technique based on K-Means and genetic algorithms for creating a capacitated cluster that respects the constraints established in an efficient and balanced manner, that minimizes the cost of relocating consumer units and the time required for final planning creation. The results obtained by the presented method are compared with the current planning of a real city, showing an improvement of 54.71% in the standard deviation of working load and 11.97% in the compactness of the groups.

Keywords: capacitated clustering, k-means, genetic algorithm, districting problems

Procedia PDF Downloads 168

Authors: Eric Lee G. Escobedo-Wu, Nidhi Rohatgi, Fouzel Dhebar

Abstract:

It is challenging for patients to navigate through healthcare systems after-hours. This leads to delays in care, patient/provider dissatisfaction, inappropriate resource utilization, readmissions, and higher costs. It is important to provide patients and providers with effective clinical decision-making tools to allow seamless connectivity and coordinated care. In August 2015, patient-centric Stanford Health Care established Clinical Advice Services (CAS) to provide clinical decision support after-hours. CAS is founded on key Lean principles: Value stream mapping, empathy mapping, waste walk, takt time calculations, standard work, plan-do-check-act cycles, and active daily management. At CAS, Clinical Assistants take the initial call and manage all non-clinical calls (e.g., appointments, directions, general information). If the patient has a clinical symptom, the CAS nurses take the call and utilize standardized clinical algorithms to triage the patient to home, clinic, urgent care, emergency department, or 911. Nurses may also contact the on-call physician based on the clinical algorithm for further direction and consultation. Since August 2015, CAS has managed 228,990 calls from 26 clinical specialties. Reporting is built into the electronic health record for analysis and data collection. 65.3% of the after-hours calls are clinically related. Average clinical algorithm adherence rate has been 92%. An average of 9% of calls was escalated by CAS nurses to the physician on call. An average of 5% of patients was triaged to the Emergency Department by CAS. Key learnings indicate that a seamless connectivity vision, cascading, multidisciplinary ownership of the problem, and synergistic enterprise improvements have contributed to this success while striving for continuous improvement.

Keywords: after hours phone calls, clinical advice services, nurse triage, Stanford Health Care

Procedia PDF Downloads 145

Authors: Shih-Wen Hsiao, Yi-Cheng Tsao

Abstract:

In the field of reverse engineering and creative industries, applying 3D scanning process to obtain geometric forms of the objects is a mature and common technique. For instance, organic objects such as faces and nonorganic objects such as products could be scanned to acquire the geometric information for further application. However, although the data resolution of 3D scanning device is increasing and there are more and more abundant complementary applications, the penetration rate of 3D scanning for the public is still limited by the relative high price of the devices. On the other hand, Kinect, released by Microsoft, is known for its powerful functions, considerably low price, and complete technology and database support. Therefore, related studies can be done with the applying of Kinect under acceptable cost and data precision. Due to the fact that Kinect utilizes optical mechanism to extracting depth information, limitations are found due to the reason of the straight path of the light. Thus, various angles are required sequentially to obtain the complete 3D information of the object when applying a single Kinect for 3D scanning. The integration process which combines the 3D data from different angles by certain algorithms is also required. This sequential scanning process costs much time and the complex integration process often encounter some technical problems. Therefore, this paper aimed to apply multiple Kinects simultaneously on the field of developing a rapid 3D mannequin scan platform and proposed suggestions on the number and angles of Kinects. In the content, a method of establishing the coordination based on the relation between mannequin and the specifications of Kinect is proposed, and a suggestion of angles and number of Kinects is also described. An experiment of applying multiple Kinect on the scanning of 3D mannequin is constructed by Microsoft API, and the results show that the time required for scanning and technical threshold can be reduced in the industries of fashion and garment design.

Keywords: 3D scan, depth sensor, fashion and garment design, mannequin, multiple Kinect sensor

Procedia PDF Downloads 343

Authors: Nelson Kimeli Kemboi Yego, Juma Kasozi, Joseph Nkruzinza, Francis Kipkogei

Abstract:

The study investigated the low uptake of health insurance in Kenya despite efforts to achieve universal health coverage through various health insurance schemes. Unsupervised machine learning techniques were employed to identify patterns in health insurance uptake based on sociodemographic factors among Kenyan households. The aim was to identify key demographic groups that are underinsured and to provide insights for the development of effective policies and outreach programs. Using the 2021 FinAccess Survey, the study clustered Kenyan households based on their health insurance uptake and sociodemographic features to reveal patterns in health insurance uptake across the country. The effectiveness of k-prototypes clustering, hierarchical clustering, and agglomerative hierarchical clustering in clustering based on sociodemographic factors was compared. The k-prototypes approach was found to be the most effective at uncovering distinct and well-separated clusters in the Kenyan sociodemographic data related to health insurance uptake based on silhouette, Calinski-Harabasz, Davies-Bouldin, and Rand indices. Hence, it was utilized in uncovering the patterns in uptake. The results of the analysis indicate that inclusivity in health insurance is greatly related to affordability. The findings suggest that targeted policy interventions and outreach programs are necessary to increase health insurance uptake in Kenya, with the ultimate goal of achieving universal health coverage. The study provides important insights for policymakers and stakeholders in the health insurance sector to address the low uptake of health insurance and to ensure that healthcare services are accessible and affordable to all Kenyans, regardless of their socio-demographic status. The study highlights the potential of unsupervised machine learning techniques to provide insights into complex health policy issues and improve decision-making in the health sector.

Keywords: health insurance, unsupervised learning, clustering algorithms, machine learning

Procedia PDF Downloads 84

Authors: Misganaw Chere Siferih

Abstract:

This study assessed the impact of crossbreed dairy cows on household income and food security. The study area is found in Dejen Woreda, East Gojam Zone, and Amhara region of Ethiopia. Random sampling technique was used to obtain a sample of 80 crossbreed dairy cow owners and 176 indigenous dairy cow owners. The study employed food consumption score analytical framework to measure food security status of the household. No Statistical significant mean difference is found between crossbreed owners and indigenous owners. Logistic regression was employed to investigate crossbreed dairy cow adoption determinants , the result indicates that gender, education, labor number, land size cultivated, dairy cooperatives membership, net income and food security status of the household are statistically significant independent variables, which explained the binary dependent variable, crossbreed dairy cow adoption. Propensity score matching (PSM) was employed to analyze the impact of crossbreed dairy cow owners on farmers’ income and food security. The average net income of crossbreed dairy cow owners was found to be significantly higher than indigenous dairy cow owners. Estimates of average treatment effect of the treated (ATT) indicated that crossbreed dairy cow is able to impact households’ net income by 42%, 38.5%, 30.8% and 44.5% higher in kernel, radius, nearest neighborhood and stratification matching algorithms respectively as compared to indigenous dairy cow owners. However, estimates of average treatment of the treated (ATT) suggest that being an owner of crossbreed dairy cow is not able to affect food security significantly. Thus, crossbreed dairy cow enables farmers to increase income but not their food security in the study area. Finally, the study recommended establishing dairy cooperatives and advice farmers to become a member of them, attention to promoting the impact of crossbreed dairy cows and promotion of nutrition focus projects.

Keywords: crossbreed dairy cow, net income, food security, propensity score matching

Procedia PDF Downloads 17

Authors: Cristy Stagnar, Nina Hubig, Diana Ivankovic

Abstract:

The epigenome is a compelling candidate for mediating long-term responses to environmental effects modifying disease risk. The main goal of this research is to develop a machine learning-based DNA methylation score, which will be valuable in delineating the unique contribution of paternal epigenetic modifications to the germline impacting childhood health outcomes. It will also be a useful tool in validating self-reports of nonsmoking and in adjusting epigenome-wide DNA methylation association studies for this early-life exposure. Using secondary data from two population-based methylation profiling studies, our DNA methylation score is based on CpG DNA methylation measurements from cord blood gathered from children whose fathers smoked pre- and peri-conceptually. Each child’s mother and father fell into one of three class labels in the accompanying questionnaires -never smoker, former smoker, or current smoker. By applying different machine learning algorithms to the accessible resource for integrated epigenomic studies (ARIES) sub-study of the Avon longitudinal study of parents and children (ALSPAC) data set, which we used for training and testing of our model, the best-performing algorithm for classifying the father smoker and mother never smoker was selected based on Cohen’s κ. Error in the model was identified and optimized. The final DNA methylation score was further tested and validated in an independent data set. This resulted in a linear combination of methylation values of selected probes via a logistic link function that accurately classified each group and contributed the most towards classification. The result is a unique, robust DNA methylation score which combines information on DNA methylation and early life exposure of offspring to paternal smoking during pregnancy and which may be used to examine the paternal contribution to offspring health outcomes.

Keywords: epigenome, health outcomes, paternal preconception environmental exposures, supervised machine learning

Procedia PDF Downloads 163

Authors: M. Varin, A. M. Dubois, R. Gadbois-Langevin, B. Chalghaf

Abstract:

Good knowledge of the physical environment is essential for an integrated forest planning. This information enables better forecasting of operating costs, determination of cutting volumes, and preservation of ecologically sensitive areas. The use of satellite images in stereoscopic pairs gives the capacity to generate high precision 3D models, which are scale-adapted for harvesting operations. These models could represent an alternative to 3D LiDAR data, thanks to their advantageous cost of acquisition. The objective of the study was to assess the quality of stereo-derived canopy height models (CHM) in comparison to a traditional LiDAR CHM and ground tree-height samples. Two study sites harboring two different forest stand types (broadleaf and conifer) were analyzed using stereo pairs and tri-stereo images from the WorldView-3 satellite to calculate CHM. Acquisition of multispectral images from an Unmanned Aerial Vehicle (UAV) was also realized on a smaller part of the broadleaf study site. Different algorithms using two softwares (PCI Geomatica and Correlator3D) with various spatial resolutions and band selections were tested to select the 3D modeling technique, which offered the best performance when compared with LiDAR. In the conifer study site, the CHM produced with Corelator3D using only the 50-cm resolution panchromatic band was the one with the smallest Root-mean-square deviation (RMSE: 1.31 m). In the broadleaf study site, the tri-stereo model provided slightly better performance, with an RMSE of 1.2 m. The tri-stereo model was also compared to the UAV, which resulted in an RMSE of 1.3 m. At individual tree level, when ground samples were compared to satellite, lidar, and UAV CHM, RMSE were 2.8, 2.0, and 2.0 m, respectively. Advanced analysis was done for all of these cases, and it has been noted that RMSE is reduced when the canopy cover is higher when shadow and slopes are lower and when clouds are distant from the analyzed site.

Keywords: very high spatial resolution, satellite imagery, WorlView-3, canopy height models, CHM, LiDAR, unmanned aerial vehicle, UAV

Procedia PDF Downloads 92

Authors: Gayathri Nagarajan, L. D. Dhinesh Babu

Abstract:

Health care is one of the prominent industries that generate voluminous data thereby finding the need of machine learning techniques with big data solutions for efficient processing and prediction. Missing data, incomplete data, real time streaming data, sensitive data, privacy, heterogeneity are few of the common challenges to be addressed for efficient processing and mining of health care data. In comparison with other applications, accuracy and fast processing are of higher importance for health care applications as they are related to the human life directly. Though there are many machine learning techniques and big data solutions used for efficient processing and prediction in health care data, different techniques and different frameworks are proved to be effective for different applications largely depending on the characteristics of the datasets. In this paper, we present a framework that uses ensemble machine learning technique gradient boosted trees for data classification in health care big data. The framework is built on Spark platform which is fast in comparison with other traditional frameworks. Unlike other works that focus on a single technique, our work presents a comparison of six different machine learning techniques along with gradient boosted trees on datasets of different characteristics. Five benchmark health care datasets are considered for experimentation, and the results of different machine learning techniques are discussed in comparison with gradient boosted trees. The metric chosen for comparison is misclassification error rate and the run time of the algorithms. The goal of this paper is to i) Compare the performance of gradient boosted trees with other machine learning techniques in Spark platform specifically for health care big data and ii) Discuss the results from the experiments conducted on datasets of different characteristics thereby drawing inference and conclusion. The experimental results show that the accuracy is largely dependent on the characteristics of the datasets for other machine learning techniques whereas gradient boosting trees yields reasonably stable results in terms of accuracy without largely depending on the dataset characteristics.

Keywords: big data analytics, ensemble machine learning, gradient boosted trees, Spark platform

Procedia PDF Downloads 216

Authors: Anand Malik

Abstract:

The Himalayas with high tectonic activities poses a great threat to human life and property. Climate change is another reason which triggering extreme events multiple fold effect on high mountain glacial environment, rock falls, landslides, debris flows, flash flood and snow avalanches. One such extreme event of cloud burst along with breach of moraine dammed Chorabri Lake occurred from June 14 to June 17, 2013, triggered flooding of Saraswati and Mandakini rivers in the Kedarnath Valley of Rudraprayag district of Uttrakhand state of India. As a result, huge volume of water with its high velocity created a catastrophe of the century, which resulted into loss of large number of human/animals, pilgrimage, tourism, agriculture and property. Thus a comprehensive assessment of debris flow hazards requires GIS-based modeling using numerical methods. The aim of present study is to focus on analysis and mapping of debris flow movements using geospatial data with flow-r (developed by team at IGAR, University of Lausanne). The model is based on combined probabilistic and energetic algorithms for the assessment of spreading of flow with maximum run out distances. Aster Digital Elevation Model (DEM) with 30m x 30m cell size (resolution) is used as main geospatial data for preparing the run out assessment, while Landsat data is used to analyze land use land cover change in the study area. The results of the study area show that model can be applied with great accuracy as the model is very useful in determining debris flow areas. The results are compared with existing available landslides/debris flow maps. ArcGIS software is used in preparing run out susceptibility maps which can be used in debris flow mitigation and future land use planning.

Keywords: debris flow, geospatial data, GIS based modeling, flow-R

Procedia PDF Downloads 245

Authors: Thomas Weber, Nina Strobel, Thomas Kohne, Eberhard Abele

Abstract:

In context of the energy transition in Germany, the importance of so-called virtual power plants in the energy supply continues to increase. The progressive dismantling of the large power plants and the ongoing construction of many new decentralized plants result in great potential for optimization through synergies between the individual plants. These potentials can be exploited by mathematical optimization algorithms to calculate the optimal application planning of decentralized power and heat generators and storage systems. This also includes linear or linear mixed integer optimization. In this paper, procedures for reducing the number of decision variables to be calculated are explained and validated. On the one hand, this includes combining n similar installation types into one aggregated unit. This aggregated unit is described by the same constraints and target function terms as a single plant. This reduces the number of decision variables per time step and the complexity of the problem to be solved by a factor of n. The exact operating mode of the individual plants can then be calculated in a second optimization in such a way that the output of the individual plants corresponds to the calculated output of the aggregated unit. Another way to reduce the number of decision variables in an optimization problem is to reduce the number of time steps to be calculated. This is useful if a high temporal resolution is not necessary for all time steps. For example, the volatility or the forecast quality of environmental parameters may justify a high or low temporal resolution of the optimization. Both approaches are examined for the resulting calculation time as well as for optimality. Several optimization models for virtual power plants (combined heat and power plants, heat storage, power storage, gas turbine) with different numbers of plants are used as a reference for the investigation of both processes with regard to calculation duration and optimality.

Keywords: CHP, Energy 4.0, energy storage, MILP, optimization, virtual power plant

Procedia PDF Downloads 146

Authors: Mst Shapna Akter, Hossain Shahriar

Abstract:

One of the most important challenges in the field of software code audit is the presence of vulnerabilities in software source code. Every year, more and more software flaws are found, either internally in proprietary code or revealed publicly. These flaws are highly likely exploited and lead to system compromise, data leakage, or denial of service. C and C++ open-source code are now available in order to create a largescale, machine-learning system for function-level vulnerability identification. We assembled a sizable dataset of millions of opensource functions that point to potential exploits. We developed an efficient and scalable vulnerability detection method based on deep neural network models that learn features extracted from the source codes. The source code is first converted into a minimal intermediate representation to remove the pointless components and shorten the dependency. Moreover, we keep the semantic and syntactic information using state-of-the-art word embedding algorithms such as glove and fastText. The embedded vectors are subsequently fed into deep learning networks such as LSTM, BilSTM, LSTM-Autoencoder, word2vec, BERT, and GPT-2 to classify the possible vulnerabilities. Furthermore, we proposed a neural network model which can overcome issues associated with traditional neural networks. Evaluation metrics such as f1 score, precision, recall, accuracy, and total execution time have been used to measure the performance. We made a comparative analysis between results derived from features containing a minimal text representation and semantic and syntactic information. We found that all of the deep learning models provide comparatively higher accuracy when we use semantic and syntactic information as the features but require higher execution time as the word embedding the algorithm puts on a bit of complexity to the overall system.

Keywords: cyber security, vulnerability detection, neural networks, feature extraction

Procedia PDF Downloads 45

Authors: Nwachukwu Ifeanyi

Abstract:

Wind energy is a vital component of the global renewable energy portfolio, with wind turbines serving as the primary means of harnessing this abundant resource. However, the efficiency and stability of wind turbines remain critical challenges in maximizing energy output and ensuring long-term operational viability. This study proposes a comprehensive approach utilizing computational aerodynamics and aeromechanics to optimize wind turbine performance across multiple objectives. The proposed research aims to integrate advanced computational fluid dynamics (CFD) simulations with structural analysis techniques to enhance the aerodynamic efficiency and mechanical stability of wind turbine blades. By leveraging multi-objective optimization algorithms, the study seeks to simultaneously optimize aerodynamic performance metrics such as lift-to-drag ratio and power coefficient while ensuring structural integrity and minimizing fatigue loads on the turbine components. Furthermore, the investigation will explore the influence of various design parameters, including blade geometry, airfoil profiles, and turbine operating conditions, on the overall performance and stability of wind turbines. Through detailed parametric studies and sensitivity analyses, valuable insights into the complex interplay between aerodynamics and structural dynamics will be gained, facilitating the development of next-generation wind turbine designs. Ultimately, this research endeavours to contribute to the advancement of sustainable energy technologies by providing innovative solutions to enhance the efficiency, reliability, and economic viability of wind power generation systems. The findings have the potential to inform the design and optimization of wind turbines, leading to increased energy output, reduced maintenance costs, and greater environmental benefits in the transition towards a cleaner and more sustainable energy future.

Keywords: computation, robotics, mathematics, simulation

Procedia PDF Downloads 12

Authors: M. Uysal, M. Yilmaz, I. Tiryakioğlu

Abstract:

Digital Terrain Model (DTM) is a digital numerical representation of the Earth's surface. DTMs have been applied to a diverse field of tasks, such as urban planning, military, glacier mapping, disaster management. In the expression of the Earth' surface as a mathematical model, an infinite number of point measurements are needed. Because of the impossibility of this case, the points at regular intervals are measured to characterize the Earth's surface and DTM of the Earth is generated. Hitherto, the classical measurement techniques and photogrammetry method have widespread use in the construction of DTM. At present, RADAR, LiDAR, and stereo satellite images are also used for the construction of DTM. In recent years, especially because of its superiorities, Airborne Light Detection and Ranging (LiDAR) has an increased use in DTM applications. A 3D point cloud is created with LiDAR technology by obtaining numerous point data. However recently, by the development in image mapping methods, the use of unmanned aerial vehicles (UAV) for photogrammetric data acquisition has increased DTM generation from image-based point cloud. The accuracy of the DTM depends on various factors such as data collection method, the distribution of elevation points, the point density, properties of the surface and interpolation methods. In this study, the random data reduction method is compared for DTMs generated from image based point cloud data. The original image based point cloud data set (100%) is reduced to a series of subsets by using random algorithm, representing the 75, 50, 25 and 5% of the original image based point cloud data set. Over the ANS campus of Afyon Kocatepe University as the test area, DTM constructed from the original image based point cloud data set is compared with DTMs interpolated from reduced data sets by Kriging interpolation method. The results show that the random data reduction method can be used to reduce the image based point cloud datasets to 50% density level while still maintaining the quality of DTM.

Keywords: DTM, Unmanned Aerial Vehicle (UAV), uniform, random, kriging

Procedia PDF Downloads 124

Authors: Zahid Ullah, Atlas Khan

Abstract:

The rapid advancements in data collection, storage, and processing capabilities have led to an explosion of data in various domains. In this era of big data, mathematical sciences play a crucial role in uncovering valuable insights and driving informed decision-making through data analytics. The purpose of this abstract is to present the latest advances in mathematical sciences and their application in harnessing the power of data analytics. This abstract highlights the interdisciplinary nature of data analytics, showcasing how mathematics intersects with statistics, computer science, and other related fields to develop cutting-edge methodologies. It explores key mathematical techniques such as optimization, mathematical modeling, network analysis, and computational algorithms that underpin effective data analysis and interpretation. The abstract emphasizes the role of mathematical sciences in addressing real-world challenges across different sectors, including finance, healthcare, engineering, social sciences, and beyond. It showcases how mathematical models and statistical methods extract meaningful insights from complex datasets, facilitating evidence-based decision-making and driving innovation. Furthermore, the abstract emphasizes the importance of collaboration and knowledge exchange among researchers, practitioners, and industry professionals. It recognizes the value of interdisciplinary collaborations and the need to bridge the gap between academia and industry to ensure the practical application of mathematical advancements in data analytics. The abstract highlights the significance of ongoing research in mathematical sciences and its impact on data analytics. It emphasizes the need for continued exploration and innovation in mathematical methodologies to tackle emerging challenges in the era of big data and digital transformation. In summary, this abstract sheds light on the advances in mathematical sciences and their pivotal role in unveiling the power of data analytics. It calls for interdisciplinary collaboration, knowledge exchange, and ongoing research to further unlock the potential of mathematical methodologies in addressing complex problems and driving data-driven decision-making in various domains.

Keywords: mathematical sciences, data analytics, advances, unveiling

Procedia PDF Downloads 58

Authors: Monireh Ahmadi Bani, Adel Khorramrouz, Lalenoor Morvarid, Bagheri Mahtab

Abstract:

Background:- Foot disorders are common in musculoskeletal problems. Plantar pressure distribution measurement is one the most important part of foot disorders diagnosis for quantitative analysis. However, the association of plantar pressure and foot disorders is not clear. With the growth of dataset and machine learning methods, the relationship between foot disorders and plantar pressures can be detected. Significance of the study:- The purpose of this study was to predict the probability of common foot disorders based on peak plantar pressure distribution and center of pressure during walking. Methodologies:- 2323 participants were assessed in a foot therapy clinic between 2015 and 2021. Foot disorders were diagnosed by an experienced physician and then they were asked to walk on a force plate scanner. After the data preprocessing, due to the difference in walking time and foot size, we normalized the samples based on time and foot size. Some of force plate variables were selected as input to a deep neural network (DNN), and the probability of any each foot disorder was measured. In next step, we used support vector machine (SVM) and run dataset for each foot disorder (classification of yes or no). We compared DNN and SVM for foot disorders prediction based on plantar pressure distributions and center of pressure. Findings:- The results demonstrated that the accuracy of deep learning architecture is sufficient for most clinical and research applications in the study population. In addition, the SVM approach has more accuracy for predictions, enabling applications for foot disorders diagnosis. The detection accuracy was 71% by the deep learning algorithm and 78% by the SVM algorithm. Moreover, when we worked with peak plantar pressure distribution, it was more accurate than center of pressure dataset. Conclusion:- Both algorithms- deep learning and SVM will help therapist and patients to improve the data pool and enhance foot disorders prediction with less expense and error after removing some restrictions properly.

Keywords: deep neural network, foot disorder, plantar pressure, support vector machine

Procedia PDF Downloads 316

Authors: Kevin Wang, Jian Ge, Yinan Zhao, Kevin Willis

Abstract:

Kepler has discovered over 4000 exoplanets and candidates. However, current transit planet detection techniques based on the wavelet analysis and the Box Least Squares (BLS) algorithm have limited sensitivity in detecting minor planets with a low signal-to-noise ratio (SNR) and long periods with only 3-4 repeated signals over the mission lifetime of 4 years. This paper presents a novel precise-period transit signal detection methodology based on a new Graphics Processing Unit (GPU) Fast Folding algorithm in conjunction with a Convolutional Neural Network (CNN) to detect low SNR and/or long-period transit planet signals. A comparison with BLS is conducted on both simulated light curves and real data, demonstrating that the new method has higher speed, sensitivity, and reliability. For instance, the new system can detect transits with SNR as low as three while the performance of BLS drops off quickly around SNR of 7. Meanwhile, the GPU Fast Folding method folds light curves 25 times faster than BLS, a significant gain that allows exoplanet detection to occur at unprecedented period precision. This new method has been tested with all known transit signals with 100% confirmation. In addition, this new method has been successfully applied to the Kepler of Interest (KOI) data and identified a few new Earth-sized Ultra-short period (USP) exoplanet candidates and habitable planet candidates. The results highlight the promise for GPU Fast Folding as a replacement to the traditional BLS algorithm for finding small and/or long-period habitable and Earth-sized planet candidates in-transit data taken with Kepler and other space transit missions such as TESS(Transiting Exoplanet Survey Satellite) and PLATO(PLAnetary Transits and Oscillations of stars).

Keywords: algorithms, astronomy data analysis, deep learning, exoplanet detection methods, small planets, habitable planets, transit photometry

Procedia PDF Downloads 179

Authors: Oscar E. Cariceo, Claudia V. Casal

Abstract:

Machine learning offers a set of techniques to promote social work interventions and can lead to support decisions of practitioners in order to predict new behaviors based on data produced by the organizations, services agencies, users, clients or individuals. Machine learning techniques include a set of generalizable algorithms that are data-driven, which means that rules and solutions are derived by examining data, based on the patterns that are present within any data set. In other words, the goal of machine learning is teaching computers through 'examples', by training data to test specifics hypothesis and predict what would be a certain outcome, based on a current scenario and improve that experience. Machine learning can be classified into two general categories depending on the nature of the problem that this technique needs to tackle. First, supervised learning involves a dataset that is already known in terms of their output. Supervising learning problems are categorized, into regression problems, which involve a prediction from quantitative variables, using a continuous function; and classification problems, which seek predict results from discrete qualitative variables. For social work research, machine learning generates predictions as a key element to improving social interventions on complex social issues by providing better inference from data and establishing more precise estimated effects, for example in services that seek to improve their outcomes. This paper exposes the results of a classification algorithm to predict cyberbullying among adolescents. Data were retrieved from the National Polyvictimization Survey conducted by the government of Chile in 2017. A logistic regression model was created to predict if an adolescent would experience cyberbullying based on the interaction and behavior of gender, age, grade, type of school, and self-esteem sentiments. The model can predict with an accuracy of 59.8% if an adolescent will suffer cyberbullying. These results can help to promote programs to avoid cyberbullying at schools and improve evidence based practice.

Keywords: cyberbullying, evidence based practice, machine learning, social work research

Procedia PDF Downloads 141

Authors: Dovile Petkeviciute-Barysiene

Abstract:

Legal technologies not only create new ways for accessing and providing legal services but also transform the role of legal practitioners. Both lawyers and users of legal services expect automated solutions to outperform people with objectivity and impartiality. Although fairness of the automated decisions is crucial, research on assessing various characteristics of automated processes related to the perceived fairness has only begun. One of the major obstacles to this research is the lack of comprehensive understanding of what legal actions are automated and could be meaningfully automated, and to what extent. Neither public nor legal practitioners oftentimes cannot envision technological input due to the lack of general without illustrative examples. The aim of this study is to map decision making stages and automation levels which are and/or could be achieved in legal actions related to pre-trial and trial processes. Major legal decisions and judgments are identified during the consultations with legal practitioners. The dual-process model of information processing is used to describe cognitive processes taking place while making legal decisions and judgments during pre-trial and trial action. Some of the existing legal technologies are incorporated into the analysis as well. Several published automation level taxonomies are considered because none of them fit well into the legal context, as they were all created for avionics, teleoperation, unmanned aerial vehicles, etc. From the information processing perspective, analysis of the legal decisions and judgments expose situations that are most sensitive to cognitive bias, among others, also help to identify areas that would benefit from the automation the most. Automation level analysis, in turn, provides a systematic approach to interaction and cooperation between humans and algorithms. Moreover, an integrated map of legal decisions and judgments, information processing characteristics, and automation levels all together provide some groundwork for the research of legal technology perceived fairness and acceptance. Acknowledgment: This project has received funding from European Social Fund (project No 09.3.3-LMT-K-712-19-0116) under grant agreement with the Research Council of Lithuania (LMTLT).

Keywords: automation levels, information processing, legal judgment and decision making, legal technology

Procedia PDF Downloads 104

Authors: Faheem Shahzad Baloch, Muhammad Azhar Nadeem, Muhammad Amjad Nawaz, Ephrem Habyarimana, Gonul Comertpay, Tolga Karakoy, Rustu Hatipoglu, Mehmet Zahit Yeken, Vahdettin Ciftci

Abstract:

Turkey is considered a bridge between Europe, Asia, and Africa and possibly played an important role in the distribution of many crops including common bean. Hundreds of common bean landraces can be found in Turkey, particularly in farmers’ fields, and they consistently contribute to the overall production. To investigate the existing genetic diversity and hybridization events between the Andean and Mesoamerican gene pools in the Turkish common bean, 188 common bean accessions (182 landraces and 6 modern cultivars as controls) were collected from 19 different Turkish geographic regions. These accessions were characterized using phenotypic data (growth habit and seed weight), geographic provenance, 12557 high-quality whole-genome DArTseq markers, and 3767 novel DArTseq loci were also identified. The clustering algorithms resolved the Turkish common bean landrace germplasm into the two recognized gene pools, the Mesoamerican and Andean gene pools. Hybridization events were observed in both gene pools (14.36% of the accessions) but mostly in the Mesoamerican (7.97% of the accessions), and was low relative to previous European studies. The lower level of hybridization witnessed the existence of Turkish common bean germplasm in its original form as compared to Europe. Mesoamerican gene pool reflected a higher level of diversity, while the Andean gene pool was predominant (56.91% of the accessions), but genetically less diverse and phenotypically more pure, reflecting farmers greater preference for the Andean gene pool. We also found some genetically distinct landraces and overall, a meaningful level of genetic variability which can be used by the scientific community in breeding efforts to develop superior common bean strains.

Keywords: bean germplasm, DArTseq markers, genotyping by sequencing, Turkey, whole genome diversity

Procedia PDF Downloads 208

Authors: Haoxiang Desmond Lim, Jie Wu, Tze How Daniel New, Shengxian Shi

Abstract:

This paper outlines the development of a novel experimental technique in quantifying supersonic jet flows, in an attempt to avoid seeding particle problems frequently associated with particle-image velocimetry (PIV) techniques at high Mach numbers. Based on optical flow algorithms, the idea behind the technique involves using high speed cameras to capture Schlieren images of the supersonic jet shear layers, before they are subjected to an adapted optical flow algorithm based on the Horn-Schnuck method to determine the associated flow fields. The proposed method is capable of offering full-field unsteady flow information with potentially higher accuracy and resolution than existing point-measurements or PIV techniques. Preliminary study via numerical simulations of a circular de Laval jet nozzle successfully reveals flow and shock structures typically associated with supersonic jet flows, which serve as useful data for subsequent validation of the optical flow based experimental results. For experimental technique, a Z-type Schlieren setup is proposed with supersonic jet operated in cold mode, stagnation pressure of 8.2 bar and exit velocity of Mach 1.5. High-speed single-frame or double-frame cameras are used to capture successive Schlieren images. As implementation of optical flow technique to supersonic flows remains rare, the current focus revolves around methodology validation through synthetic images. The results of validation test offers valuable insight into how the optical flow algorithm can be further improved to improve robustness and accuracy. Details of the methodology employed and challenges faced will be further elaborated in the final conference paper should the abstract be accepted. Despite these challenges however, this novel supersonic flow measurement technique may potentially offer a simpler way to identify and quantify the fine spatial structures within the shock shear layer.

Keywords: Schlieren, optical flow, supersonic jets, shock shear layer

Procedia PDF Downloads 288

Authors: Yunus Onur Yildiz, Mesut Kirca

Abstract:

In this study, mechanical properties of a nanoporous metal coated with a different metallic material are studied through a new atomistic modelling technique and molecular dynamics (MD) simulations. This new atomistic modelling technique is based on the Voronoi tessellation method for the purpose of geometric representation of the ligaments. With the proposed technique, atomistic models of nanoporous metals which have randomly oriented ligaments with non-uniform mass distribution along the ligament axis can be generated by enabling researchers to control both ligament length and diameter. Furthermore, by the utilization of this technique, atomistic models of coated nanoporous materials can be numerically obtained for further mechanical or thermal characterization. In general, this study consists of two stages. At the first stage, we use algorithms developed for generating atomic coordinates of the coated nanoporous material. In this regard, coordinates of randomly distributed points are determined in a controlled way to be employed in the establishment of the Voronoi tessellation, which results in randomly oriented and intersected line segments. Then, line segment representation of the Voronoi tessellation is transformed to atomic structure by a special process. This special process includes generation of non-uniform volumetric core region in which atoms can be generated based on a specific crystal structure. As an extension, this technique can be used for coating of nanoporous structures by creating another volumetric region encapsulating the core region in which atoms for the coating material are generated. The ultimate goal of the study at this stage is to generate atomic coordinates that can be employed in the MD simulations of randomly organized coated nanoporous structures. At the second stage of the study, mechanical behavior of the coated nanoporous models is investigated by examining deformation mechanisms through MD simulations. In this way, the effect of coating on the mechanical behavior of the selected material couple is investigated.

Keywords: atomistic modelling, molecular dynamic, nanoporous metals, voronoi tessellation

Procedia PDF Downloads 259

Authors: Kiana Zeighami, Morteza Ozlati Moghadam

Abstract:

Designing a high accuracy and high precision motion controller is one of the important issues in today’s industry. There are effective solutions available in the industry but the real-time performance, smoothness and accuracy of the movement can be further improved. This paper discusses a complete solution to carry out the movement of three stepper motors in three dimensions. The objective is to provide a method to design a fully integrated System-on-Chip (SOC)-based motion controller to reduce the cost and complexity of production by incorporating Field Programmable Gate Array (FPGA) into the design. In the proposed method the FPGA receives its commands from a host computer via wireless internet communication and calculates the motion trajectory for three axes. A profile generator module is designed to realize the interpolation algorithm by translating the position data to the real-time pulses. This paper discusses an approach to implement the linear interpolation algorithm, since it is one of the fundamentals of robots’ movements and it is highly applicable in motion control industries. Along with full profile trajectory, the triangular drive is implemented to eliminate the existence of error at small distances. To integrate the parallelism and real-time performance of FPGA with the power of Central Processing Unit (CPU) in executing complex and sequential algorithms, the NIOS II soft-core processor was added into the design. This paper presents different operating modes such as absolute, relative positioning, reset and velocity modes to fulfill the user requirements. The proposed approach was evaluated by designing a custom-made FPGA board along with a mechanical structure. As a result, a precise and smooth movement of stepper motors was observed which proved the effectiveness of this approach.

Keywords: 3-axis linear interpolation, FPGA, motion controller, micro-stepping

Procedia PDF Downloads 185

Authors: Reza Safdari, Goli Arji, Robab Abdolkhani Maryam zahmatkeshan

Abstract:

Background and purpose: Cardiovascular diseases are among the most common diseases in all societies. The most important step in minimizing myocardial infarction and its complications is to minimize its risk factors. The amount of medical data is increasingly growing. Medical data mining has a great potential for transforming these data into information. Using data mining techniques to generate predictive models for identifying those at risk for reducing the effects of the disease is very helpful. The present study aimed to collect data related to risk factors of heart infarction from patients’ medical record and developed predicting models using data mining algorithm. Methods: The present work was an analytical study conducted on a database containing 350 records. Data were related to patients admitted to Shahid Rajaei specialized cardiovascular hospital, Iran, in 2011. Data were collected using a four-sectioned data collection form. Data analysis was performed using SPSS and Clementine version 12. Seven predictive algorithms and one algorithm-based model for predicting association rules were applied to the data. Accuracy, precision, sensitivity, specificity, as well as positive and negative predictive values were determined and the final model was obtained. Results: five parameters, including hypertension, DLP, tobacco smoking, diabetes, and A+ blood group, were the most critical risk factors of myocardial infarction. Among the models, the neural network model was found to have the highest sensitivity, indicating its ability to successfully diagnose the disease. Conclusion: Risk prediction models have great potentials in facilitating the management of a patient with a specific disease. Therefore, health interventions or change in their life style can be conducted based on these models for improving the health conditions of the individuals at risk.

Keywords: decision trees, neural network, myocardial infarction, Data Mining

Procedia PDF Downloads 403

Authors: N. N. Mosola, S. J. Molete, L. S. Masoebe, M. Letsae

Abstract:

Hand gesture recognition is a technique used to locate, detect, and recognize a hand gesture. Detection and recognition are concepts of Artificial Intelligence (AI). AI concepts are applicable in Human Computer Interaction (HCI), Expert systems (ES), etc. Hand gesture recognition can be used in sign language interpretation. Sign language is a visual communication tool. This tool is used mostly by deaf societies and those with speech disorder. Communication barriers exist when societies with speech disorder interact with others. This research aims to build a hand recognition system for Lesotho’s Sesotho and English language interpretation. The system will help to bridge the communication problems encountered by the mentioned societies. The system has various processing modules. The modules consist of a hand detection engine, image processing engine, feature extraction, and sign recognition. Detection is a process of identifying an object. The proposed system uses Canny pruning Haar and Haarcascade detection algorithms. Canny pruning implements the Canny edge detection. This is an optimal image processing algorithm. It is used to detect edges of an object. The system employs a skin detection algorithm. The skin detection performs background subtraction, computes the convex hull, and the centroid to assist in the detection process. Recognition is a process of gesture classification. Template matching classifies each hand gesture in real-time. The system was tested using various experiments. The results obtained show that time, distance, and light are factors that affect the rate of detection and ultimately recognition. Detection rate is directly proportional to the distance of the hand from the camera. Different lighting conditions were considered. The more the light intensity, the faster the detection rate. Based on the results obtained from this research, the applied methodologies are efficient and provide a plausible solution towards a light-weight, inexpensive system which can be used for sign language interpretation.

Keywords: canny pruning, hand recognition, machine learning, skin tracking

Procedia PDF Downloads 156