Search results for: data fitting

Authors: Djamila Benhaddouche, Abdelkader Benyettou

Abstract:

In this paper, we used data mining to extract biomedical knowledge. In general, complex biomedical data collected in studies of populations are treated by statistical methods, although they are robust, they are not sufficient in themselves to harness the potential wealth of data. For that you used in step two learning algorithms: the Decision Trees and Support Vector Machine (SVM). These supervised classification methods are used to make the diagnosis of thyroid disease. In this context, we propose to promote the study and use of symbolic data mining techniques.

Keywords: biomedical data, learning, classifier, algorithms decision tree, knowledge extraction

Procedia PDF Downloads 518

Authors: Usama Ahmed

Abstract:

Data mining is the process of analyze data which are used to predict helpful information. It is the field of research which solve various type of problem. In data mining, classification is an important technique to classify different kind of data. Diabetes is most common disease. This paper implements different classification technique using Waikato Environment for Knowledge Analysis (WEKA) on diabetes dataset and find which algorithm is suitable for working. The best classification algorithm based on diabetic data is Naïve Bayes. The accuracy of Naïve Bayes is 76.31% and take 0.06 seconds to build the model.

Keywords: data mining, classification, diabetes, WEKA

Procedia PDF Downloads 124

Authors: Asifa Amara, Prachi Singh, Kanishka, Debargho Pathak, Akshat Kumar, Jayakumar Eravelly

Abstract:

Today's generation is totally dependent on technology that uses data as its fuel. The present study is all about innovations and developments in data science and gives an idea about how efficiently to use the data provided. This study will help to understand the core concepts of data science. The concept of artificial intelligence was introduced by Alan Turing in which the main principle was to create an artificial system that can run independently of human-given programs and can function with the help of analyzing data to understand the requirements of the users. Data science comprises business understanding, analyzing data, ethical concerns, understanding programming languages, various fields and sources of data, skills, etc. The usage of data science has evolved over the years. In this review article, we have covered a part of data science, i.e., machine learning. Machine learning uses data science for its work. Machines learn through their experience, which helps them to do any work more efficiently. This article includes a comparative study image between human understanding and machine understanding, advantages, applications, and real-time examples of machine learning. Data science is an important game changer in the life of human beings. Since the advent of data science, we have found its benefits and how it leads to a better understanding of people, and how it cherishes individual needs. It has improved business strategies, services provided by them, forecasting, the ability to attend sustainable developments, etc. This study also focuses on a better understanding of data science which will help us to create a better world.

Keywords: data science, machine learning, data analytics, artificial intelligence

Procedia PDF Downloads 50

Authors: Londeka Ngubane

Abstract:

The use of corporal punishment is not a new phenomenon in the South African education system as it was, for a long time, recognised as a fitting form of punishment for ill-disciplined and disobedient children. The growing recognition that corporal punishment is an act of violence against children has resulted in the abolishment of this form of punishment in society and particularly in schools. However, regardless of criminalising corporal punishment, it appears to be a disciplinary measure that is persistently used by some educators. Historically and currently, the intimate connection between corporal punishment and discipline has not merely been a convention of human thinking, as this practice is given recognition in various definitions in dictionaries. ‘To discipline’ is habitually stated to mean ‘to punish’. The notion of ‘disciplining children’ also comes from entrenched common conceptions about children and their relationship with adults. Corporal punishment has, for a long time, been associated with the rearing and education of children, and this practice thus pervades schooling across nations. In many societies, punishment is a term that is closely linked with the self-perception of teachers who feel that they must be ‘in control’ and have ‘the upper hand’ in order to be respected. This impression of control is evident in the widespread conception of education which is to ‘socialize’ children in ‘desirable ways’ of ‘sitting in a formal classroom’, ‘behaving’ in school, ‘following instructions’ from the teacher, talking only when asked to, and finishing tasks on time. It was against this backdrop that a comprehensive review of relevant literature was undertaken and that individual interviews were conducted with fifty learners from four schools (two junior secondary and two senior secondary schools) in a selected township area in KwaZulu-Natal Province. The main aim of the study was to explore and thus understand learners’ views on the administration of corporal punishment regardless of the fact that it was legally abolished. It was envisaged that the interviews with the learners would elicit rich data that would enhance the researcher’s insight into their perceptions of the persistent use of corporal punishment as a disciplinary measure in their schools. The study was thus premised on the assumption, which had been strengthened by anecdotal and media evidence, that corporal punishment was still administered in some schools in South Africa and in schools in the study area in particular.

Keywords: corporal punishment, ban, school learners, South Africa

Procedia PDF Downloads 127

Authors: Jennifer Buz, Alvin Spivey

Abstract:

The most common greenhouse gas emitted through human activities, carbon dioxide (CO2), is naturally consumed by plants during photosynthesis. This process is actively being monetized by companies wishing to offset their carbon dioxide emissions. For example, companies are now able to purchase protections for vegetated land due-to-be clear cut or purchase barren land for reforestation. Therefore, by actively preventing the destruction/decay of plant matter or by introducing more plant matter (reforestation), a company can theoretically offset some of their emissions. One of the biggest issues in the carbon credit market is validating and verifying carbon offsets. There is a need for a system that can accurately and frequently ensure that the areas sold for carbon credits have the vegetation mass (and therefore for carbon offset capability) they claim. Traditional techniques for measuring vegetation mass and determining health are costly and require many person-hours. Orbital Sidekick offers an alternative approach that accurately quantifies carbon mass and assesses vegetation health through satellite hyperspectral imagery, a technique which enables us to remotely identify material composition (including plant species) and condition (e.g., health and growth stage). How much carbon a plant is capable of storing ultimately is tied to many factors, including material density (primarily species-dependent), plant size, and health (trees that are actively decaying are not effectively storing carbon). All of these factors are capable of being observed through satellite hyperspectral imagery. This abstract focuses on speciation. To build a species classification model, we matched pixels in our remote sensing imagery to plants on the ground for which we know the species. To accomplish this, we collaborated with the researchers at the Teakettle Experimental Forest. Our remote sensing data comes from our airborne “Kato” sensor, which flew over the study area and acquired hyperspectral imagery (400-2500 nm, 472 bands) at ~0.5 m/pixel resolution. Coverage of the entire teakettle experimental forest required capturing dozens of individual hyperspectral images. In order to combine these images into a mosaic, we accounted for potential variations of atmospheric conditions throughout the data collection. To do this, we ran an open source atmospheric correction routine called ISOFIT1 (Imaging Spectrometer Optiman FITting), which converted all of our remote sensing data from radiance to reflectance. A database of reflectance spectra for each of the tree species within the study area was acquired using the Teakettle stem map and the geo-referenced hyperspectral images. We found that a wide variety of machine learning classifiers were able to identify the species within our images with high (>95%) accuracy. For the most robust quantification of carbon mass and the best assessment of the health of a vegetated area, speciation is critical. Through the use of high resolution hyperspectral data, ground-truth databases, and complex analytical techniques, we are able to determine the species present within a pixel to a high degree of accuracy. These species identifications will feed directly into our carbon mass model.

Keywords: hyperspectral, satellite, carbon, imagery, python, machine learning, speciation

Procedia PDF Downloads 96

Authors: S.K. Bhandari, E. Veneklaas, L. McCaw, R. Mazanec, K. Whitford, M. Renton

Abstract:

Eucalyptus marginata, E. diversicolor and Corymbia calophylla form widespread forests in south-west Western Australia (SWWA). These forests have economic and ecological importance, and therefore, tree growth and sustainable management are of high priority. This paper aimed to analyse and model the growth of these species at both stand and individual levels, but this presentation will focus on predicting the growth of E. Marginata at the individual tree level. More specifically, the study wanted to investigate how well individual E. marginata tree growth could be predicted by considering the diameter and height of the tree at the start of the growth period, and whether this prediction could be improved by also accounting for the competition from neighbouring trees in different ways. The study also wanted to investigate how many neighbouring trees or what neighbourhood distance needed to be considered when accounting for competition. To achieve this aim, the Pearson correlation coefficient was examined among competition indices (CIs), between CIs and dbh growth, and selected the competition index that can best predict the diameter growth of individual trees of E. marginata forest managed under different thinning regimes at Inglehope in SWWA. Furthermore, individual tree growth models were developed using simple linear regression, multiple linear regression, and linear mixed effect modelling approaches. Individual tree growth models were developed for thinned and unthinned stand separately. The developed models were validated using two approaches. In the first approach, models were validated using a subset of data that was not used in model fitting. In the second approach, the model of the one growth period was validated with the data of another growth period. Tree size (diameter and height) was a significant predictor of growth. This prediction was improved when the competition was included in the model. The fit statistic (coefficient of determination) of the model ranged from 0.31 to 0.68. The model with spatial competition indices validated as being more accurate than with non-spatial indices. The model prediction can be optimized if 10 to 15 competitors (by number) or competitors within ~10 m (by distance) from the base of the subject tree are included in the model, which can reduce the time and cost of collecting the information about the competitors. As competition from neighbours was a significant predictor with a negative effect on growth, it is recommended including neighbourhood competition when predicting growth and considering thinning treatments to minimize the effect of competition on growth. These model approaches are likely to be useful tools for the conservations and sustainable management of forests of E. marginata in SWWA. As a next step in optimizing the number and distance of competitors, further studies in larger size plots and with a larger number of plots than those used in the present study are recommended.

Keywords: competition, growth, model, thinning

Procedia PDF Downloads 102

Authors: Azadeh Bashiri

Abstract:

Introduction: Lack of proper diagnosis and inadequate treatment of asthma leads to physical and financial complications. This study aimed to use data mining techniques and creating a neural network intelligent system for diagnosis of asthma. Methods: The study population is the patients who had visited one of the Lung Clinics in Tehran. Data were analyzed using the SPSS statistical tool and the chi-square Pearson's coefficient was the basis of decision making for data ranking. The considered neural network is trained using back propagation learning technique. Results: According to the analysis performed by means of SPSS to select the top factors, 13 effective factors were selected, in different performances, data was mixed in various forms, so the different models were made for training the data and testing networks and in all different modes, the network was able to predict correctly 100% of all cases. Conclusion: Using data mining methods before the design structure of system, aimed to reduce the data dimension and the optimum choice of the data, will lead to a more accurate system. Therefore, considering the data mining approaches due to the nature of medical data is necessary.

Keywords: asthma, data mining, Artificial Neural Network, intelligent system

Procedia PDF Downloads 246

Authors: Christopher Muhawe, Masooda Bashir

Abstract:

With increased Internet Communication Technology(ICT), the virtual world has become the new normal. At the same time, there is an unprecedented collection of massive amounts of data by both private and public entities. Unfortunately, this increase in data collection has been in tandem with an increase in data misuse and data breach. Regrettably, the majority of data breach and data misuse claims have been unsuccessful in the United States courts for the failure of proof of direct injury to physical or economic interests. The requirement to express data privacy harms from an economic or physical stance negates the fact that not all data harms are physical or economic in nature. The challenge is compounded by the fact that data breach harms and risks do not attach immediately. This research will use a descriptive and normative approach to show that not all data harms can be expressed in economic or physical terms. Expressing privacy harms purely from an economic or physical harm perspective negates the fact that data insecurity may result into harms which run counter the functions of privacy in our lives. The promotion of liberty, selfhood, autonomy, promotion of human social relations and the furtherance of the existence of a free society. There is no economic value that can be placed on these functions of privacy. The proposed approach addresses data harms from a psychological and social perspective.

Keywords: data breach and misuse, economic harms, privacy harms, psychological harms

Procedia PDF Downloads 168

Authors: Jose Alberto Duarte Moller, Cynthia Deisy Gomez Esparza

Abstract:

The cobalt-titanium system was grown as thin films in an INTERCOVAMEX V3 sputtering system, equipped with four magnetrons assisted by DC pulsed and direct DC. A polished highly oriented (400) silicon wafer was used as substrate and the growing temperature was 500 oC. Xray Absorption Spectroscopy experiments were carried out in the SSRL in the 4-3 beam line. The Extenden X-Ray Absorption Fine Structure spectra have been numerically processed by WINXAS software from the background subtraction until the normalization and FFT adjustment. Analyzing the absorption spectra of cobalt in the CoTi2 phase we can appreciate that they agree in energy with the reference spectra that corresponds to the CoO, which indicates that the valence where upon working is Co2+. The RDF experimental results were then compared with those RDF´s generated theoretically by using FEFF software, from a model compound of CoTi2 phase obtained by XRD. The fitting procedure is a highly iterative process. Fits are also checked in R-space using both the real and imaginary parts of Fourier transform. Finally, the presence of overlapping coordination shells and the correctness of the assumption about the nature of the coordinating atom were checked.

Keywords: XAS, EXAFS, FEFF, CoTi

Procedia PDF Downloads 266

Authors: Chandra Prayaga, Aaron Wade, Lakshmi Prayaga, Gopi Shankar Mallu

Abstract:

This paper presents the use of machine learning algorithms to predict the success of students in an introductory physics course. Data having 140 rows pertaining to the performance of two batches of students was used. The lack of sufficient data to train robust machine learning models was compensated for by generating synthetic data similar to the real data. CTGAN and CTGAN with Gaussian Copula (Gaussian) were used to generate synthetic data, with the real data as input. To check the similarity between the real data and each synthetic dataset, pair plots were made. The synthetic data was used to train machine learning models using the PyCaret package. For the CTGAN data, the Ada Boost Classifier (ADA) was found to be the ML model with the best fit, whereas the CTGAN with Gaussian Copula yielded Logistic Regression (LR) as the best model. Both models were then tested for accuracy with the real data. ROC-AUC analysis was performed for all the ten classes of the target variable (Grades A, A-, B+, B, B-, C+, C, C-, D, F). The ADA model with CTGAN data showed a mean AUC score of 0.4377, but the LR model with the Gaussian data showed a mean AUC score of 0.6149. ROC-AUC plots were obtained for each Grade value separately. The LR model with Gaussian data showed consistently better AUC scores compared to the ADA model with CTGAN data, except in two cases of the Grade value, C- and A-.

Keywords: machine learning, student success, physics course, grades, synthetic data, CTGAN, gaussian copula CTGAN

Procedia PDF Downloads 20

Authors: Lucille Alonso, Florent Renard

Abstract:

The characterizations of urban heat island (UHI) and their interactions with climate change and urban climates are the main research and public health issue, due to the increasing urbanization of the population. These solutions require a better knowledge of the UHI and micro-climate in urban areas, by combining measurements and modelling. This study is part of this topic by evaluating microclimatic conditions in dense urban areas in the Lyon Metropolitan Area (France) using a combination of data traditionally used such as topography, but also from LiDAR (Light Detection And Ranging) data, Landsat 8 satellite observation and Sentinel and ground measurements by bike. These bicycle-dependent weather data collections are used to build the database of the variable to be modelled, the air temperature, over Lyon’s hyper-center. This study aims to model the air temperature, measured during 6 mobile campaigns in Lyon in clear weather, using multiple linear regressions based on 33 explanatory variables. They are of various categories such as meteorological parameters from remote sensing, topographic variables, vegetation indices, the presence of water, humidity, bare soil, buildings, radiation, urban morphology or proximity and density to various land uses (water surfaces, vegetation, bare soil, etc.). The acquisition sources are multiple and come from the Landsat 8 and Sentinel satellites, LiDAR points, and cartographic products downloaded from an open data platform in Greater Lyon. Regarding the presence of low, medium, and high vegetation, the presence of buildings and ground, several buffers close to these factors were tested (5, 10, 20, 25, 50, 100, 200 and 500m). The buffers with the best linear correlations with air temperature for ground are 5m around the measurement points, for low and medium vegetation, and for building 50m and for high vegetation is 100m. The explanatory model of the dependent variable is obtained by multiple linear regression of the remaining explanatory variables (Pearson correlation matrix with a |r| < 0.7 and VIF with < 5) by integrating a stepwise sorting algorithm. Moreover, holdout cross-validation is performed, due to its ability to detect over-fitting of multiple regression, although multiple regression provides internal validation and randomization (80% training, 20% testing). Multiple linear regression explained, on average, 72% of the variance for the study days, with an average RMSE of only 0.20°C. The impact on the model of surface temperature in the estimation of air temperature is the most important variable. Other variables are recurrent such as distance to subway stations, distance to water areas, NDVI, digital elevation model, sky view factor, average vegetation density, or building density. Changing urban morphology influences the city's thermal patterns. The thermal atmosphere in dense urban areas can only be analysed on a microscale to be able to consider the local impact of trees, streets, and buildings. There is currently no network of fixed weather stations sufficiently deployed in central Lyon and most major urban areas. Therefore, it is necessary to use mobile measurements, followed by modelling to characterize the city's multiple thermal environments.

Keywords: air temperature, LIDAR, multiple linear regression, surface temperature, urban heat island

Procedia PDF Downloads 108

Authors: Luke Ukpebor, C. E. Abhulimen

Abstract:

Stiff initial value problems in ordinary differential equations are problems for which a typical solution is rapidly decaying exponentially, and their numerical investigations are very tedious. Conventional numerical integration solvers cannot cope effectively with stiff problems as they lack adequate stability characteristics. In this article, we developed a new family of four-step second derivative exponentially fitted method of order six for the numerical integration of stiff initial value problem of general first order differential equations. In deriving our method, we employed the idea of breaking down the general multi-derivative multistep method into predator and corrector schemes which possess free parameters that allow for automatic fitting into exponential functions. The stability analysis of the method was discussed and the method was implemented with numerical examples. The result shows that the method is A-stable and competes favorably with existing methods in terms of efficiency and accuracy.

Keywords: A-stable, exponentially fitted, four step, predator-corrector, second derivative, stiff initial value problems

Procedia PDF Downloads 228

Authors: Chuping Lo

Abstract:

This paper presents a simple model demonstrating that ceteris paribus countries with lower barriers to accessing global data tend to earn higher incomes than other countries. Therefore, large countries that inherently have greater data resources tend to have higher incomes than smaller countries, such that the former may be more hesitant than the latter to liberalize cross-border data flows to maintain this advantage. Furthermore, countries with higher artificial intelligence (AI) intensity in production technologies tend to benefit more from economies of scale in data aggregation, leading to higher income and more trade as they are better able to utilize global data.

Keywords: digital intensity, digital divide, international trade, scale of economics

Procedia PDF Downloads 40

Authors: G. R. Navaneesh, E. Nagarajan, C. H. Rajam Raju

Abstract:

Nowadays the multimedia data are used to store some secure information. All previous methods allocate a space in image for data embedding purpose after encryption. In this paper, we propose a novel method by reserving space in image with a boundary surrounded before encryption with a traditional RDH algorithm, which makes it easy for the data hider to reversibly embed data in the encrypted images. The proposed method can achieve real time performance, that is, data extraction and image recovery are free of any error. A secure transmission process is also discussed in this paper, which improves the efficiency by ten times compared to other processes as discussed.

Keywords: secure communication, reserving room before encryption, least significant bits, image encryption, reversible data hiding

Procedia PDF Downloads 379

Authors: Fuad M. Alkoot

Abstract:

DNA data have been used in forensics for decades. However, current research looks at using the DNA as a biometric identity verification modality. The goal is to improve the speed of identification. We aim at using gene data that was initially used for autism detection to find if and how accurate is this data for identification applications. Mainly our goal is to find if our data preprocessing technique yields data useful as a biometric identification tool. We experiment with using the nearest neighbor classifier to identify subjects. Results show that optimal classification rate is achieved when the test set is corrupted by normally distributed noise with zero mean and standard deviation of 1. The classification rate is close to optimal at higher noise standard deviation reaching 3. This shows that the data can be used for identity verification with high accuracy using a simple classifier such as the k-nearest neighbor (k-NN). 

Keywords: biometrics, genetic data, identity verification, k nearest neighbor

Procedia PDF Downloads 228

Authors: Dino Carpentras, Alejandro Dinkelberg, Michael Quayle

Abstract:

Within agent-based modeling, opinion dynamics is the field that focuses on modeling people's opinions. In this prolific field, most of the literature is dedicated to the exploration of the two 'degrees of freedom' and how they impact the model’s properties (e.g., the average final opinion, the number of final clusters, etc.). These degrees of freedom are (1) the interaction rule, which determines how agents update their own opinion, and (2) the network topology, which defines the possible interaction among agents. In this work, we show that the third degree of freedom exists. This can be used to change a model's output up to 100% of its initial value or to transform two models (both from the literature) into each other. Since opinion dynamics models are representations of the real world, it is fundamental to understand how people’s opinions can be measured. Even for abstract models (i.e., not intended for the fitting of real-world data), it is important to understand if the way of numerically representing opinions is unique; and, if this is not the case, how the model dynamics would change by using different representations. The process of measuring opinions is non-trivial as it requires transforming real-world opinion (e.g., supporting most of the liberal ideals) to a number. Such a process is usually not discussed in opinion dynamics literature, but it has been intensively studied in a subfield of psychology called psychometrics. In psychometrics, opinion scales can be converted into each other, similarly to how meters can be converted to feet. Indeed, psychometrics routinely uses both linear and non-linear transformations of opinion scales. Here, we analyze how this transformation affects opinion dynamics models. We analyze this effect by using mathematical modeling and then validating our analysis with agent-based simulations. Firstly, we study the case of perfect scales. In this way, we show that scale transformations affect the model’s dynamics up to a qualitative level. This means that if two researchers use the same opinion dynamics model and even the same dataset, they could make totally different predictions just because they followed different renormalization processes. A similar situation appears if two different scales are used to measure opinions even on the same population. This effect may be as strong as providing an uncertainty of 100% on the simulation’s output (i.e., all results are possible). Still, by using perfect scales, we show that scales transformations can be used to perfectly transform one model to another. We test this using two models from the standard literature. Finally, we test the effect of scale transformation in the case of finite precision using a 7-points Likert scale. In this way, we show how a relatively small-scale transformation introduces both changes at the qualitative level (i.e., the most shared opinion at the end of the simulation) and in the number of opinion clusters. Thus, scale transformation appears to be a third degree of freedom of opinion dynamics models. This result deeply impacts both theoretical research on models' properties and on the application of models on real-world data.

Keywords: degrees of freedom, empirical validation, opinion scale, opinion dynamics

Procedia PDF Downloads 103

Authors: R Palson Kennedy, P.Kiran Sai

Abstract:

This article introduces machine learning (ML) researchers to the hurdles that geoscience problems present, as well as the opportunities for improvement in both ML and geosciences. This article presents a review from the data life cycle perspective to meet that need. Numerous facets of geosciences present unique difficulties for the study of intelligent systems. Geosciences data is notoriously difficult to analyze since it is frequently unpredictable, intermittent, sparse, multi-resolution, and multi-scale. The first half addresses data science’s essential concepts and theoretical underpinnings, while the second section contains key themes and sharing experiences from current publications focused on each stage of the data life cycle. Finally, themes such as open science, smart data, and team science are considered.

Keywords: Data science, intelligent system, machine learning, big data, data life cycle, recent development, geo science

Procedia PDF Downloads 116

Authors: Yuan Zheng

Abstract:

3D model-based vehicle matching provides a new way for vehicle recognition, localization and tracking. Its key is to construct an evaluation function, also called fitness function, to measure the degree of vehicle matching. The existing fitness functions often poorly perform when the clutter and occlusion exist in traffic scenarios. In this paper, we present a practical and efficient fitness function. Unlike the existing evaluation functions, the proposed fitness function is to study the vehicle matching problem from both local and global perspectives, which exploits the pixel gradient information as well as the silhouette information. In view of the discrepancy between 3D vehicle model and real vehicle, a weighting strategy is introduced to differently treat the fitting of the model’s wireframes. Additionally, a normalization operation for the model’s projection is performed to improve the accuracy of the matching. Experimental results on real traffic videos reveal that the proposed fitness function is efficient and robust to the cluttered background and partial occlusion.

Keywords: 3D-2D matching, fitness function, 3D vehicle model, local image gradient, silhouette information

Procedia PDF Downloads 367

Authors: Shrawan Baghel, Helen Cathcart, Biall J. O'Reilly

Abstract:

Amorphous drug formulations have great potential to enhance solubility and thus bioavailability of BCS class II drugs. However, the higher free energy and molecular mobility of the amorphous form lowers the activation energy barrier for crystallization and thermodynamically drives it towards the crystalline state which makes them unstable. Accurate determination of the crystallization tendency/kinetics is the key to the successful design and development of such systems. In this study, dipyridamole (DPM) and cinnarizine (CNZ) has been selected as model compounds. Thermodynamic fragility (m_T) is measured from the heat capacity change at the glass transition temperature (Tg) whereas dynamic fragility (m_D) is evaluated using methods based on extrapolation of configurational entropy to zero 〖(m〗_(D_CE )), and heating rate dependence of Tg 〖(m〗_(D_Tg)). The mean relaxation time of amorphous drugs was calculated from Vogel-Tammann-Fulcher (VTF) equation. Furthermore, the correlation between fragility and glass forming ability (GFA) of model drugs has been established and the relevance of these parameters to crystallization of amorphous drugs is also assessed. Moreover, the crystallization kinetics of model drugs under isothermal conditions has been studied using Johnson-Mehl-Avrami (JMA) approach to determine the Avrami constant ‘n’ which provides an insight into the mechanism of crystallization. To further probe into the crystallization mechanism, the non-isothermal crystallization kinetics of model systems was also analysed by statistically fitting the crystallization data to 15 different kinetic models and the relevance of model-free kinetic approach has been established. In addition, the crystallization mechanism for DPM and CNZ at each extent of transformation has been predicted. The calculated fragility, glass forming ability (GFA) and crystallization kinetics is found to be in good correlation with the stability prediction of amorphous solid dispersions. Thus, this research work involves a multidisciplinary approach to establish fragility, GFA and crystallization kinetics as stability predictors for amorphous drug formulations.

Keywords: amorphous, fragility, glass forming ability, molecular mobility, mean relaxation time, crystallization kinetics, stability

Procedia PDF Downloads 324

Authors: Zhifeng Kong

Abstract:

Over-parameterized neural networks have attracted a great deal of attention in recent deep learning theory research, as they challenge the classic perspective of over-fitting when the model has excessive parameters and have gained empirical success in various settings. While a number of theoretical works have been presented to demystify properties of such models, the convergence properties of such models are still far from being thoroughly understood. In this work, we study the convergence properties of training two-hidden-layer partially over-parameterized fully connected networks with the Rectified Linear Unit activation via gradient descent. To our knowledge, this is the first theoretical work to understand convergence properties of deep over-parameterized networks without the equally-wide-hidden-layer assumption and other unrealistic assumptions. We provide a probabilistic lower bound of the widths of hidden layers and proved linear convergence rate of gradient descent. We also conducted experiments on synthetic and real-world datasets to validate our theory.

Keywords: over-parameterization, rectified linear units ReLU, convergence, gradient descent, neural networks

Procedia PDF Downloads 111

Authors: Marc Bachelet, Abhijit Kumar Chatterjee, José Manuel Avila

Abstract:

Data quality is a key component of any data-driven organization. Without data quality, organizations cannot effectively make data-driven decisions, which often leads to poor business performance. Therefore, it is important for an organization to ensure that the data they use is of high quality. This is where the concept of data mesh comes in. Data mesh is an organizational and architectural decentralized approach to data management that can help organizations improve the quality of data. The concept of data mesh was first introduced in 2020. Its purpose is to decentralize data ownership, making it easier for domain experts to manage the data. This can help organizations improve data quality by reducing the reliance on centralized data teams and allowing domain experts to take charge of their data. This paper intends to discuss how a set of elements, including data mesh, are tools capable of increasing data quality. One of the key benefits of data mesh is improved metadata management. In a traditional data architecture, metadata management is typically centralized, which can lead to data silos and poor data quality. With data mesh, metadata is managed in a decentralized manner, ensuring accurate and up-to-date metadata, thereby improving data quality. Another benefit of data mesh is the clarification of roles and responsibilities. In a traditional data architecture, data teams are responsible for managing all aspects of data, which can lead to confusion and ambiguity in responsibilities. With data mesh, domain experts are responsible for managing their own data, which can help provide clarity in roles and responsibilities and improve data quality. Additionally, data mesh can also contribute to a new form of organization that is more agile and adaptable. By decentralizing data ownership, organizations can respond more quickly to changes in their business environment, which in turn can help improve overall performance by allowing better insights into business as an effect of better reports and visualization tools. Monitoring and analytics are also important aspects of data quality. With data mesh, monitoring, and analytics are decentralized, allowing domain experts to monitor and analyze their own data. This will help in identifying and addressing data quality problems in quick time, leading to improved data quality. Data culture is another major aspect of data quality. With data mesh, domain experts are encouraged to take ownership of their data, which can help create a data-driven culture within the organization. This can lead to improved data quality and better business outcomes. Finally, the paper explores the contribution of AI in the coming years. AI can help enhance data quality by automating many data-related tasks, like data cleaning and data validation. By integrating AI into data mesh, organizations can further enhance the quality of their data. The concepts mentioned above are illustrated by AEKIDEN experience feedback. AEKIDEN is an international data-driven consultancy that has successfully implemented a data mesh approach. By sharing their experience, AEKIDEN can help other organizations understand the benefits and challenges of implementing data mesh and improving data quality.

Keywords: data culture, data-driven organization, data mesh, data quality for business success

Procedia PDF Downloads 99

Authors: Pramika J. Muthya

Abstract:

This paper investigates the impact of web graphics and interface design on building visitor trust in websites. A quantitative survey approach was used to examine how aesthetic and usability elements of website design influence user perceptions of trustworthiness. 133 participants aged 18-25 who live in urban Bangalore and engage in online transactions were recruited via convenience sampling. Data was collected through an online survey measuring trust levels based on website design, using validated constructs like the Visual Aesthetic of Websites Inventory (VisAWI). Statistical analysis, including ordinal regression, was conducted to analyze the results. The findings show a statistically significant relationship between web graphics and interface design and the level of trust visitors place in a website. The goodness-of-fit statistics and highly significant model fitting information provide strong evidence for rejecting the null hypothesis of no relationship. Well-designed visual aesthetics like simplicity, diversity, colorfulness, and craftsmanship are key drivers of perceived credibility. Intuitive navigation and usability also increase trust. The results emphasize the strategic importance for companies to invest in appealing graphic design, consistent with existing theoretical frameworks. There are also implications for taking a user-centric approach to web design and acknowledging the reciprocal link between pre-existing user trust and perception of visuals. While generalizable, limitations include possible sampling and self-report biases. Further research can build on these findings to deepen understanding of nuanced cultural and temporal factors influencing online trust. Overall, this study makes a significant contribution by providing empirical evidence that reinforces the crucial impact of thoughtful graphic design in fostering lasting user trust in websites.

Keywords: web graphics, interface design, visitor trust, website design, aesthetics, user experience, online trust, visual design, graphic design, user perceptions, user expectations

Procedia PDF Downloads 25

Authors: Ji Eun Shin, Byung Ho Jung, Dong Hoon Lim

Abstract:

It is almost impossible to store or analyze big data increasing exponentially with traditional technologies. Hadoop is a new technology to make that possible. R programming language is by far the most popular statistical tool for big data analysis based on distributed processing with Hadoop technology. With RHadoop that integrates R and Hadoop environment, we implemented parallel multiple regression analysis with different sizes of actual data. Experimental results showed our RHadoop system was much faster as the number of data nodes increases. We also compared the performance of our RHadoop with lm function and big lm packages available on big memory. The results showed that our RHadoop was faster than other packages owing to paralleling processing with increasing the number of map tasks as the size of data increases.

Keywords: big data, Hadoop, parallel regression analysis, R, RHadoop

Procedia PDF Downloads 411

Authors: Haojie Wang, Xun Li, Rui Yin

Abstract:

In order to solve the memorization overfitting in the meta-learning MAML algorithm, a method of generating mutually exclusive tasks based on data augmentation is proposed. This method generates a mutex task by corresponding one feature of the data to multiple labels, so that the generated mutex task is inconsistent with the data distribution in the initial dataset. Because generating mutex tasks for all data will produce a large number of invalid data and, in the worst case, lead to exponential growth of computation, this paper also proposes a key data extraction method, that only extracts part of the data to generate the mutex task. The experiments show that the method of generating mutually exclusive tasks can effectively solve the memorization overfitting in the meta-learning MAML algorithm.

Keywords: data augmentation, mutex task generation, meta-learning, text classification.

Procedia PDF Downloads 68

Authors: Sifat Rahman Ahona, Rifat Tasnim, Naima Hassan

Abstract:

Data aggregation is a helpful technique for reducing the data communication overhead in wireless sensor network. One of the important tasks of data aggregation is positioning of the aggregator points. There are a lot of works done on data aggregation. But, efficient positioning of the aggregators points is not focused so much. In this paper, authors are focusing on the positioning or the placement of the aggregation points in wireless sensor network. Authors proposed an algorithm to select the aggregators positions for a scenario where aggregator nodes are more powerful than sensor nodes.

Keywords: aggregation point, data communication, data aggregation, wireless sensor network

Procedia PDF Downloads 129

Authors: Paula Simões, Isabel Natário

Abstract:

This paper reviews a number of theoretical aspects for implementing an explicit spatial perspective in econometrics for modelling non-continuous data, in general, and count data, in particular. It provides an overview of the several spatial econometric approaches that are available to model data that are collected with reference to location in space, from the classical spatial econometrics approaches to the recent developments on spatial econometrics to model count data, in a Bayesian hierarchical setting. Considerable attention is paid to the inferential framework, necessary for structural consistent spatial econometric count models, incorporating spatial lag autocorrelation, to the corresponding estimation and testing procedures for different assumptions, to the constrains and implications embedded in the various specifications in the literature. This review combines insights from the classical spatial econometrics literature as well as from hierarchical modeling and analysis of spatial data, in order to look for new possible directions on the processing of count data, in a spatial hierarchical Bayesian econometric context.

Keywords: spatial data analysis, spatial econometrics, Bayesian hierarchical models, count data

Procedia PDF Downloads 561

Authors: Gueidi Afef, Gharsellaoui Hamza, Ben Ahmed Samir

Abstract:

This paper deals with the secret of the continual progression data that new data management solutions have been emerged: The NoSQL databases. They crossed several areas like personalization, profile management, big data in real-time, content management, catalog, view of customers, mobile applications, internet of things, digital communication and fraud detection. Nowadays, these database management systems are increasing. These systems store data very well and with the trend of big data, a new challenge’s store demands new structures and methods for managing enterprise data. The new intelligent machine in the e-learning sector, thrives on more data, so smart machines can learn more and faster. The robotics are our use case to focus on our test. The implementation of NoSQL for Robotics wrestle all the data they acquire into usable form because with the ordinary type of robotics; we are facing very big limits to manage and find the exact information in real-time. Our original proposed approach was demonstrated by experimental studies and running example used as a use case.

Keywords: NoSQL databases, database management systems, robotics, big data

Procedia PDF Downloads 324

Authors: C. B. Le, V. N. Pham

Abstract:

In modern data analysis, multi-source data appears more and more in real applications. Multi-source data clustering has emerged as a important issue in the data mining and machine learning community. Different data sources provide information about different data. Therefore, multi-source data linking is essential to improve clustering performance. However, in practice multi-source data is often heterogeneous, uncertain, and large. This issue is considered a major challenge from multi-source data. Ensemble is a versatile machine learning model in which learning techniques can work in parallel, with big data. Clustering ensemble has been shown to outperform any standard clustering algorithm in terms of accuracy and robustness. However, most of the traditional clustering ensemble approaches are based on single-objective function and single-source data. This paper proposes a new clustering ensemble method for multi-source data analysis. The fuzzy optimized multi-objective clustering ensemble method is called FOMOCE. Firstly, a clustering ensemble mathematical model based on the structure of multi-objective clustering function, multi-source data, and dark knowledge is introduced. Then, rules for extracting dark knowledge from the input data, clustering algorithms, and base clusterings are designed and applied. Finally, a clustering ensemble algorithm is proposed for multi-source data analysis. The experiments were performed on the standard sample data set. The experimental results demonstrate the superior performance of the FOMOCE method compared to the existing clustering ensemble methods and multi-source clustering methods.

Keywords: clustering ensemble, multi-source, multi-objective, fuzzy clustering

Procedia PDF Downloads 146

Authors: M. Yushalify Misro, Ahmad Ramli, Jamaludin M. Ali

Abstract:

Bezier curves have useful properties for path generation problem, for instance, it can generate the reference trajectory for vehicles to satisfy the path constraints. Both algorithms join cubic Bezier curve segment smoothly to generate the path. Some of the useful properties of Bezier are curvature. In mathematics, the curvature is the amount by which a geometric object deviates from being flat, or straight in the case of a line. Another extrinsic example of curvature is a circle, where the curvature is equal to the reciprocal of its radius at any point on the circle. The smaller the radius, the higher the curvature thus the vehicle needs to bend sharply. In this study, we use Bezier curve to fit highway-like curve. We use the different approach to finding the best approximation for the curve so that it will resemble highway-like curve. We compute curvature value by analytical differentiation of the Bezier Curve. We will then compute the maximum speed for driving using the curvature information obtained. Our research works on some assumptions; first the Bezier curve estimates the real shape of the curve which can be verified visually. Even, though, the fitting process of Bezier curve does not interpolate exactly on the curve of interest, we believe that the estimation of speed is acceptable. We verified our result with the manual calculation of the curvature from the map.

Keywords: speed estimation, path constraints, reference trajectory, Bezier curve

Procedia PDF Downloads 349

Authors: Haojie Wang, Xun Li, Rui Yin

Abstract:

In order to solve the memorization overfitting in the model-agnostic meta-learning MAML algorithm, a method of generating mutually exclusive tasks based on data augmentation is proposed. This method generates a mutex task by corresponding one feature of the data to multiple labels so that the generated mutex task is inconsistent with the data distribution in the initial dataset. Because generating mutex tasks for all data will produce a large number of invalid data and, in the worst case, lead to an exponential growth of computation, this paper also proposes a key data extraction method that only extract part of the data to generate the mutex task. The experiments show that the method of generating mutually exclusive tasks can effectively solve the memorization overfitting in the meta-learning MAML algorithm.

Keywords: mutex task generation, data augmentation, meta-learning, text classification.

Procedia PDF Downloads 103