Search results for: data combining

Authors: Serkan Aras, Emrah Gulay

Abstract:

Combining various forecasting models is an important tool for researchers to attain more accurate forecasts. A great number of papers have shown that selecting single models as dissimilar models, or methods based on different information as possible leads to better forecasting performances. However, there is not a certain rule regarding the number of single models to be used in any combining methods. This study focuses on determining the optimal or near optimal number for single models with the help of statistical tests. An extensive experiment is carried out by utilizing some well-known time series data sets from diverse fields. Furthermore, many rival forecasting methods and some of the commonly used combining methods are employed. The obtained results indicate that some statistically significant performance differences can be found regarding the number of the single models in the combining methods under investigation.

Keywords: combined forecast, forecasting, M-competition, time series

Procedia PDF Downloads 334

Authors: Yuzhi Cai

Abstract:

In this paper, we discuss a Bayesian approach to quantile autoregressive (QAR) time series model estimation and forecasting. Together with a combining forecasts technique, we then predict USD to GBP currency exchange rates. Combined forecasts contain all the information captured by the fitted QAR models at different quantile levels and are therefore better than those obtained from individual models. Our results show that an unequally weighted combining method performs better than other forecasting methodology. We found that a median AR model can perform well in point forecasting when the predictive density functions are symmetric. However, in practice, using the median AR model alone may involve the loss of information about the data captured by other QAR models. We recommend that combined forecasts should be used whenever possible.

Keywords: combining forecasts, MCMC, predictive density functions, quantile forecasting, quantile modelling

Procedia PDF Downloads 231

Authors: Shahzad Yousaf, Imran Shafi

Abstract:

This paper presents a new method for calculation of pressure loss coefficients by use of the artificial neural network (ANN) in tee junctions. Geometry and flow parameters are feed into ANN as the inputs for purpose of training the network. Efficacy of the network is demonstrated by comparison of the experimental and ANN based calculated data of pressure loss coefficients for combining flows in a tee junction. Reynolds numbers ranging from 200 to 14000 and discharge ratios varying from minimum to maximum flow for calculation of pressure loss coefficients have been used. Pressure loss coefficients calculated using ANN are compared to the models from literature used in junction flows. The results achieved after the application of ANN agrees reasonably to the experimental values.

Keywords: artificial neural networks, combining flow, pressure loss coefficients, solar collector tee junctions

Procedia PDF Downloads 366

Authors: N. R. N. Idris, S. Baharom

Abstract:

A meta-analysis may be performed using aggregate data (AD) or an individual patient data (IPD). In practice, studies may be available at both IPD and AD level. In this situation, both the IPD and AD should be utilised in order to maximize the available information. Statistical advantages of combining the studies from different level have not been fully explored. This study aims to quantify the statistical benefits of including available IPD when conducting a conventional summary-level meta-analysis. Simulated meta-analysis were used to assess the influence of the levels of data on overall meta-analysis estimates based on IPD-only, AD-only and the combination of IPD and AD (mixed data, MD), under different study scenario. The percentage relative bias (PRB), root mean-square-error (RMSE) and coverage probability were used to assess the efficiency of the overall estimates. The results demonstrate that available IPD should always be included in a conventional meta-analysis using summary level data as they would significantly increased the accuracy of the estimates. On the other hand, if more than 80% of the available data are at IPD level, including the AD does not provide significant differences in terms of accuracy of the estimates. Additionally, combining the IPD and AD has moderating effects on the biasness of the estimates of the treatment effects as the IPD tends to overestimate the treatment effects, while the AD has the tendency to produce underestimated effect estimates. These results may provide some guide in deciding if significant benefit is gained by pooling the two levels of data when conducting meta-analysis.

Keywords: aggregate data, combined-level data, individual patient data, meta-analysis

Procedia PDF Downloads 352

Authors: Yuzhi Cai

Abstract:

This talk presents a Bayesian approach to quantile autoregressive (QAR) time series model estimation and forecasting. We establish that the joint posterior distribution of the model parameters and future values is well defined. The associated MCMC algorithm for parameter estimation and forecasting converges to the posterior distribution quickly. We also present a combining forecasts technique to produce more accurate out-of-sample forecasts by using a weighted sequence of fitted QAR models. A moving window method to check the quality of the estimated conditional quantiles is developed. We verify our methodology using simulation studies and then apply it to currency exchange rate data. An application of the method to the USD to GBP daily currency exchange rates will also be discussed. The results obtained show that an unequally weighted combining method performs better than other forecasting methodology.

Keywords: combining forecasts, MCMC, quantile modelling, quantile forecasting, predictive density functions

Procedia PDF Downloads 324

Authors: Rais Yusoh, Rosli Saad, Mokhtar Saidin, Fauzi Andika, Sabiu Bala Muhammad

Abstract:

Resistivity and seismic refraction profiling have become a common method in pre-investigations for visualizing subsurface structure. The integration of the methods could reduce an interpretation ambiguity. Both methods have their individual software packages for data inversion, but potential to combine certain geophysical methods are restricted; however, the research algorithms that have this functionality was existed and are evaluated personally. The interpretation of subsurface were improve by combining inversion data from both methods by influence each other models using closure coupling; thus, by implementing both methods to support each other which could improve the subsurface interpretation. These methods were applied on a field dataset from a pre-investigation for archeology in finding the ancient river. There were no major changes in the inverted model by combining data inversion for this archetype which probably due to complex geology. The combine data analysis provides an additional technique for interpretation such as an alluvium, which can have strong influence on the ancient river findings.

Keywords: ancient river, combine inversion, resistivity, seismic refraction

Procedia PDF Downloads 307

Authors: Meng Su

Abstract:

High-dimensional data analysis often presents challenges in capturing the complex, nonlinear relationships and manifold structures inherent to the data. This article presents a novel approach that leverages the strengths of two powerful techniques, Diffusion Maps and Diffusion Probabilistic Models (DPMs), to address these challenges. By integrating the dimensionality reduction capability of Diffusion Maps with the data modeling ability of DPMs, the proposed method aims to provide a comprehensive solution for analyzing and generating high-dimensional data. The Diffusion Map technique preserves the nonlinear relationships and manifold structure of the data by mapping it to a lower-dimensional space using the eigenvectors of the graph Laplacian matrix. Meanwhile, DPMs capture the dependencies within the data, enabling effective modeling and generation of new data points in the low-dimensional space. The generated data points can then be mapped back to the original high-dimensional space, ensuring consistency with the underlying manifold structure. Through a detailed example implementation, the article demonstrates the potential of the proposed hybrid approach to achieve more accurate and effective modeling and generation of complex, high-dimensional data. Furthermore, it discusses possible applications in various domains, such as image synthesis, time-series forecasting, and anomaly detection, and outlines future research directions for enhancing the scalability, performance, and integration with other machine learning techniques. By combining the strengths of Diffusion Maps and DPMs, this work paves the way for more advanced and robust data analysis methods.

Keywords: diffusion maps, diffusion probabilistic models (DPMs), manifold learning, high-dimensional data analysis

Procedia PDF Downloads 76

Authors: Rohit Mittal, Bright Keswani, Amit Mithal

Abstract:

This paper provides a sense about the application of computational intelligence techniques used to develop computer games, especially car racing. For the deep sense and knowledge of artificial intelligence, this paper is divided into various sections that is optimization, imitation, innovation and combining approach of optimization and imitation. This paper is mainly concerned with combining approach which tells different aspects of using fitness measures and supervised learning techniques used to imitate aspects of behavior. The main achievement of this paper is based on modelling player behaviour and evolving new game content such as racing tracks as single car racing on single track.

Keywords: evolution algorithm, genetic, optimization, imitation, racing, innovation, gaming

Procedia PDF Downloads 618

Authors: M. Pepe, S. Ackermann, L. Fregonese, C. Achille

Abstract:

3D models obtained with advanced survey techniques such as close-range photogrammetry and laser scanner are nowadays particularly appreciated in Cultural Heritage and Archaeology fields. In order to produce high quality models representing archaeological evidences and anthropological artifacts, the appearance of the model (i.e. color) beyond the geometric accuracy, is not a negligible aspect. The integration of the close-range photogrammetry survey techniques with the laser scanner is still a topic of study and research. By combining point cloud data sets of the same object generated with both technologies, or with the same technology but registered in different moment and/or natural light condition, could construct a final point cloud with accentuated color dissimilarities. In this paper, a methodology to uniform the different data sets, to improve the chromatic quality and to highlight further details by balancing the point color will be presented.

Keywords: color models, cultural heritage, laser scanner, photogrammetry

Procedia PDF Downloads 260

Authors: Bharat Singh Deora, Sushma Satpute

Abstract:

Cloud computing is based on resource sharing. Like other resources which can be shareable, storage is a resource which can be shared. We can use collective resources of storage from different locations and maintain a central index table for storage details. The storage combining of different places can form a suitable data storage which is operated from one location and is very economical. Proper storage of data should improve data reliability & availability and bandwidth utilization. Also, we are moving the contents of one storage to other according to our need.

Keywords: cloud computing, cloud storage, Iaas, PaaS, SaaS

Procedia PDF Downloads 314

Authors: Nikita Valluri, Vatcharaporn Esichaikul

Abstract:

With the popularity of data-driven decision making on the rise, this study focuses on providing an alternative outlook towards the process of decision-making. Combining quantitative and qualitative methods rooted in the social sciences, an integrated framework is presented with a focus on delivering a much more robust and efficient approach towards the concept of data-driven decision-making with respect to not only Big data but also 'Thick data', a new form of qualitative data. In support of this, an example from the retail sector has been illustrated where the framework is put into action to yield insights and leverage business intelligence. An interpretive approach to analyze findings from both kinds of quantitative and qualitative data has been used to glean insights. Using traditional Point-of-sale data as well as an understanding of customer psychographics and preferences, techniques of data mining along with qualitative methods (such as grounded theory, ethnomethodology, etc.) are applied. This study’s final goal is to establish the framework as a basis for providing a holistic solution encompassing both the Big and Thick aspects of any business need. The proposed framework is a modified enhancement in lieu of traditional data-driven decision-making approach, which is mainly dependent on quantitative data for decision-making.

Keywords: big data, customer behavior, customer experience, data mining, qualitative methods, quantitative methods, thick data

Procedia PDF Downloads 136

Authors: Pan Hongxia, Wang Zhenhua

Abstract:

In this paper, the EMD decomposition in the process of endpoint effect adopted data based on AR model to predict the continuation and window function method of combining the two effective inhibition. Proven by simulation of the simulation signal obtained the ideal effect, then, apply this method to the gearbox test data is also achieved good effect in the process, for the analysis of the subsequent data processing to improve the calculation accuracy. In the end, under various working conditions for the gearbox fault diagnosis laid a good foundation.

Keywords: gearbox, fault diagnosis, ar model, end effect

Procedia PDF Downloads 343

Authors: Jolie M. Martin

Abstract:

Running experiments while logging detailed user actions has become the standard way of testing product features at Pinterest, as at many other Internet companies. While this technique offers plenty of statistical power to assess the effects of product changes on behavioral metrics, it does not often give us much insight into why users respond the way they do. By combining at-scale experiments with smaller surveys of users in each experimental condition, we have developed a unique approach for measuring the impact of our product and communication treatments on user sentiment, attitudes, and comprehension.

Keywords: experiments, methodology, surveys, user experience

Procedia PDF Downloads 297

Authors: Mohd Harun, Cini Varghese, Eldho Varghese, Seema Jaggi

Abstract:

Hybridization crosses find a vital role in breeding experiments to evaluate the combining abilities of individual parental lines or crosses for creation of lines with desirable qualities. There are various ways of obtaining progenies and further studying the combining ability effects of the lines taken in a breeding programme. Some of the most common methods are diallel or two-way cross, triallel or three-way cross, tetra-allele or four-way cross. These techniques help the breeders to improve the quantitative traits which are of economical as well as nutritional importance in crops and animals. Amongst these methods, tetra-allele cross provides extra information in terms of the higher specific combining ability (sca) effects and the hybrids thus produced exhibit individual as well as population buffering mechanism because of the broad genetic base. Most of the common commercial hybrids in corn are either three-way or four-way cross hybrids. Tetra-allele cross came out as the most practical and acceptable scheme for the production of slaughter pigs having fast growth rate, good feed efficiency, and carcass quality. Tetra-allele crosses are mostly used for exploitation of heterosis in case of commercial silkworm production. Experimental designs involving tetra-allele crosses have been studied extensively in literature. Optimality of designs has also been considered as a researchable issue. In practical situations, it is advisable to include sca effects in the model as this information is needed by the breeder to improve economically and nutritionally important quantitative traits. Thus, a model that provides information regarding the specific traits by utilizing sca effects along with general combining ability (gca) effects may help the breeders to deal with the problem of various stresses. In this paper, a model for experimental designs involving tetra-allele crosses that incorporates both gca and sca has been defined. Optimality aspects of such designs have been discussed incorporating sca effects in the model. Orthogonality conditions have been derived for block designs ensuring estimation of contrasts among the gca effects, after eliminating the nuisance factors, independently from sca effects. User friendly SAS macro and web solution (webPTC) have been developed for the generation and analysis of such designs.

Keywords: general combining ability, optimality, specific combining ability, tetra-allele cross, webPTC

Procedia PDF Downloads 110

Authors: Rashmi Raghunandan, Sanjana Shankar, Rakshitha K. Bhat

Abstract:

Social media contributes a vast amount of data and information about individuals to the internet. This project will greatly reduce the need for unnecessary manual analysis of large and diverse social media profiles by filtering out and combining the useful information from various social media profiles, eliminating irrelevant data. It differs from the existing social media aggregators in that it does not provide a consolidated view of various profiles. Instead, it provides consolidated INFORMATION derived from the subject’s posts and other activities. It also allows analysis over multiple profiles and analytics based on several profiles. We strive to provide a query system to provide a natural language answer to questions when a user does not wish to go through the entire profile. The information provided can be filtered according to the different use cases it is used for.

Keywords: social network, analysis, Facebook, Linkedin, git, big data

Procedia PDF Downloads 420

Authors: José A. Menezes, Giordano Cabral, Bruno T. Gomes

Abstract:

Choosing good features is an essential part of machine learning. Recent techniques aim to automate this process. For instance, feature learning intends to learn the transformation of raw data into a useful representation to machine learning tasks. In automatic audio classification tasks, this is interesting since the audio, usually complex information, needs to be transformed into a computationally convenient input to process. Another technique tries to generate features by searching a feature space. Genetic algorithms, for instance, have being used to generate audio features by combining or modifying them. We find this approach particularly interesting and, despite the undeniable advances of feature learning approaches, we wanted to take a step forward in the use of genetic algorithms to find audio features, combining them with more conventional methods, like PCA, and inserting search control mechanisms, such as constraints over a confusion matrix. This work presents the results obtained on particular audio classification problems.

Keywords: feature generation, feature learning, genetic algorithm, music information retrieval

Procedia PDF Downloads 404

Authors: Karine B. de Oliveira, Carina F. Dorneles

Abstract:

The structure level matching is the problem of combining elements of a structure, which can be represented as entities, classes, XML elements, web forms, and so on. This is a challenge due to large number of distinct representations of semantically similar structures. This paper describes a structure-based matching method applied to search for different representations in data sources, considering the similarity between elements of two structures and the data source context. Using real data sources, we have conducted an experimental study comparing our approach with our baseline implementation and with another important schema matching approach. We demonstrate that our proposal reaches higher precision than the baseline.

Keywords: context, data source, index, matching, search, similarity, structure

Procedia PDF Downloads 337

Authors: Prathiksha Kamath, Pratibha Jamkhandi, Prateek Ghanti, Priyanshu Gupta, M. Lakshmi Neelima

Abstract:

Combining Image Captioning and Visual Question Answering (VQA) tasks have emerged as a new and exciting research area. The image captioning task involves generating a textual description that summarizes the content of the image. VQA aims to answer a natural language question about the image. Both these tasks include computer vision and natural language processing (NLP) and require a deep understanding of the content of the image and semantic relationship within the image and the ability to generate a response in natural language. There has been remarkable growth in both these tasks with rapid advancement in deep learning. In this paper, we present a comprehensive review of recent progress in combining image captioning and visual question-answering (VQA) tasks. We first discuss both image captioning and VQA tasks individually and then the various ways in which both these tasks can be integrated. We also analyze the challenges associated with these tasks and ways to overcome them. We finally discuss the various datasets and evaluation metrics used in these tasks. This paper concludes with the need for generating captions based on the context and captions that are able to answer the most likely asked questions about the image so as to aid the VQA task. Overall, this review highlights the significant progress made in combining image captioning and VQA, as well as the ongoing challenges and opportunities for further research in this exciting and rapidly evolving field, which has the potential to improve the performance of real-world applications such as autonomous vehicles, robotics, and image search.

Keywords: image captioning, visual question answering, deep learning, natural language processing

Procedia PDF Downloads 53

Authors: Terkimbi Vange, Obed Abimiku, Lateef Lekan Bello, Lucky Omoigui

Abstract:

In 2014 and 2015, eight maize inbred lines resistant to Striga hermonthica (Del) Benth were crossed in 8 x 8 half diallel (Griffing method 11, model 1). The eight parent inbred lines were planted out in a Randomized Complete Block Design (RCBD) with three replications at two different Striga infested environments (Lafia and Makurdi) during the late cropping season. The objectives were to determine the combining ability of Striga resistant maize inbred lines and identify suitable inbreds for hybrids development. The lines were used to estimate general combining ability (GCA), and specific combining ability (SCA) effects for Striga related parameters such as Striga shoot counts, Striga damage rating (SDR), plant height and grain yield and other agronomic traits. The result of combined ANOVA revealed that mean squares were highly significant for all traits except Striga damage rating (SDR1) at 8WAS and Striga emergence count (STECOI) at 8WAS. Mean squares for SCA were significantly low for all traits. TZSTR190 was the highest yielding parent, and TZSTR166xTZST190 was the highest yielding hybrid (cross). Parent TZSTR166, TZEI188, TZSTR190 and TZSTR193 shows significant (p < 0.05) positive GCA effects for grain yield while the rest had negative GCA effects for grain yield. Parent TZSTR166, TZEI188, TZSTR190, and TZSTR193 could be used for initiating hybrid development. Also, TZSTR166xTZSTR190 cross was the best specific combiner followed by TZEI188xTZSTR193, TZEI80xTZSTR193, and TZSTR190xTZSTR193. TZSTR166xTZSTR190 and TZSTR190xTZSTR193 had the highest SCA effects. However, TZEI80 and TZSTR190 manifested a high positive SCA effect with TZSTR166 indicating that these two inbreds combined better with TZSTR166.

Keywords: combining ability, Striga hermonthica, resistance, grain yield

Procedia PDF Downloads 222

Authors: Xulu Yao, Moi Hoon Yap, Yanlong Zhang

Abstract:

As a branch of artificial neural network, deep learning is widely used in the field of image recognition, but the lack of its dataset leads to imperfect model learning. By analysing the data scale requirements of deep learning and aiming at the application in GUI generation, it is found that the collection of GUI dataset is a time-consuming and labor-consuming project, which is difficult to meet the needs of current deep learning network. To solve this problem, this paper proposes a semi-supervised deep learning model that relies on the original small-scale datasets to produce a large number of reliable data sets. By combining the cyclic neural network with the generated countermeasure network, the cyclic neural network can learn the sequence relationship and characteristics of data, make the generated countermeasure network generate reasonable data, and then expand the Rico dataset. Relying on the network structure, the characteristics of collected data can be well analysed, and a large number of reasonable data can be generated according to these characteristics. After data processing, a reliable dataset for model training can be formed, which alleviates the problem of dataset shortage in deep learning.

Keywords: GUI, deep learning, GAN, data augmentation

Procedia PDF Downloads 156

Authors: Victor Breux, Jérôme Boutet, Alain Goret, Viviane Cattin

Abstract:

Early detection of anomalies in data centers is important to reduce downtimes and the costs of periodic maintenance. However, there is little research on this topic and even fewer on the fusion of sensor data for the detection of abnormal events. The goal of this paper is to propose a method for anomaly detection in data centers by combining sensor data (temperature, humidity, power) and deep learning models. The model described in the paper uses one autoencoder per sensor to reconstruct the inputs. The auto-encoders contain Long-Short Term Memory (LSTM) layers and are trained using the normal samples of the relevant sensors selected by correlation analysis. The difference signal between the input and its reconstruction is then used to classify the samples using feature extraction and a random forest classifier. The data measured by the sensors of a data center between January 2019 and May 2020 are used to train the model, while the data between June 2020 and May 2021 are used to assess it. Performances of the model are assessed a posteriori through F1-score by comparing detected anomalies with the data center’s history. The proposed model outperforms the state-of-the-art reconstruction method, which uses only one autoencoder taking multivariate sequences and detects an anomaly with a threshold on the reconstruction error, with an F1-score of 83.60% compared to 24.16%.

Keywords: anomaly detection, autoencoder, data centers, deep learning

Procedia PDF Downloads 165

Authors: Nashrawan Naser Taha, Andrew M. Cox

Abstract:

In a multi-cultural learning context, where ties are weak and dynamic, combining qualitative with quantitative research methods may be more effective. Such a combination may also allow us to answer different types of question, such as about people’s perception of the network. In this study the use of observation, interviews and photos were explored as ways of enhancing data from social network questionnaires. Integrating all of these methods was found to enhance the quality of data collected and its accuracy, also providing a richer story of the network dynamics and the factors that shaped these changes over time.

Keywords: mixed methods, social network analysis, multi-cultural learning, social network dynamics

Procedia PDF Downloads 491

Authors: Jaime Carpio-García, Elena Blanco-Fernández, Jorge Rodríguez-Hernández, Daniel Castro-Fresno

Abstract:

Geosynthetic mechanical behavior vs temperature or vs aging has been widely studied independently during the last years, both in laboratory and in outdoor conditions. This paper studies this behavior deeper, considering that geosynthetics have to perform adequately at different outdoor temperatures once they have been subjected to a certain degree of aging, and also considering the different geosynthetic structures made of the same material. This combining effect has been not considered so far, and it is important to ensure the performance of geosynthetics, especially where high temperatures are expected. In order to fill this gap, six commercial geosynthetics with different internal structures made of polypropylene (PP), high density polyethylene (HDPE), bitumen and polyvinyl chloride (PVC), or even a combination of some of them have been mechanically tested at mild temperature (20ºC or 23ºC) and at warm temperature (45ºC) before and after specific exposition to air at standardized high temperature in order to simulate 25 years of aging due to oxidation. Besides, for 45ºC tests, an innovative heating system during test for high deformable specimens is proposed. The influence of the combining effect of aging, structure and temperature in the product behavior have been analyzed and discussed, concluding that internal structure is more influential than aging in the mechanical behavior of a geosynthetic versus temperature.

Keywords: geosynthetics, mechanical behavior, temperature, aging, internal structure

Procedia PDF Downloads 46

Authors: Cassandra A. Skinner, Linda R. Chamberlain

Abstract:

The nonprofit and public sectors are increasingly turning to Geographic Information Systems for data visualizations which can better inform programmatic and policy decisions. Additionally, the private and nonprofit sectors are turning to systems mapping to better understand the ecosystems within which they operate. This study explores the potential which combining these data visualization methods—a method which is called geographic systems mapping—to create an exhaustive and comprehensive understanding of a social problem’s ecosystem may have in social innovation efforts. Researchers with Grand Valley State University collaborated with Talent 2025 of West Michigan to conduct a mixed-methods research study to paint a comprehensive picture of the workforce development ecosystem in West Michigan. Using semi-structured interviewing, observation, secondary research, and quantitative analysis, data were compiled on workforce development organizations’ locations, programming, metrics for success, partnerships, funding sources, and service language. To best visualize and disseminate the data, a geographic system map was created which identifies programmatic, operational, and geographic gaps in workforce development services of West Michigan. By combining geographic and systems mapping methods, the geographic system map provides insight into the cross-sector relationships, collaboration, and competition which exists among and between workforce development organizations. These insights identify opportunities for and constraints around cross-sectoral social innovation in the West Michigan workforce development ecosystem. This paper will discuss the process utilized to prepare the geographic systems map, explain the results and outcomes, and demonstrate how geographic systems mapping illuminated the needs of the community and opportunities for social innovation. As complicated social problems like unemployment often require cross-sectoral and multi-stakeholder solutions, there is potential for geographic systems mapping to be a tool which informs, enables, and inspires these solutions.

Keywords: cross-sector collaboration, data visualization, geographic systems mapping, social innovation, workforce development

Procedia PDF Downloads 273

Authors: Li Qingjian, Li Ke, He Chun, Huang Yong

Abstract:

In this paper, the method of combining the Pohl Seidman's deep belief network with the self-organizing neural network is proposed to classify the target. This method is mainly aimed at the high nonlinearity of the hyperspectral image, the high sample dimension and the difficulty in designing the classifier. The main feature of original data is extracted by deep belief network. In the process of extracting features, adding known labels samples to fine tune the network, enriching the main characteristics. Then, the extracted feature vectors are classified into the self-organizing neural network. This method can effectively reduce the dimensions of data in the spectrum dimension in the preservation of large amounts of raw data information, to solve the traditional clustering and the long training time when labeled samples less deep learning algorithm for training problems, improve the classification accuracy and robustness. Through the data simulation, the results show that the proposed network structure can get a higher classification precision in the case of a small number of known label samples.

Keywords: DBN, SOM, pattern classification, hyperspectral, data compression

Procedia PDF Downloads 317

Authors: Jun Ming Moey, Zhiyaun Chen, David Nicholson

Abstract:

Bad actors are often hard to detect in data that imprints their behaviour patterns because they are comparatively rare events embedded in non-bad actor data. An unsupervised machine learning framework is applied here to detect bad actors in financial crime datasets that record millions of transactions undertaken by hundreds of actors (<0.01% bad). Specifically, the framework combines ‘shallow’ (PCA, Isolation Forest) and ‘deep’ (Autoencoder) methods to detect outlier patterns. Detection performance analysis for both the individual methods and their combination is reported.

Keywords: detection, machine learning, deep learning, unsupervised, outlier analysis, data science, fraud, financial crime

Procedia PDF Downloads 71

Authors: Srijanani Anurag Prasad

Abstract:

The Coalescence Hidden-variable Fractal Interpolation Surface (CHFIS) was built by combining interpolation data from the Iterated Function System (IFS). The interpolation data in a CHFIS comprises a row and/or column of uncertain values when a single point is entered. Alternatively, a row and/or column of additional points are placed in the given interpolation data to demonstrate the node added CHFIS. There are three techniques for inserting new points that correspond to the row and/or column of nodes inserted, and each method is further classified into four types based on the values of the inserted nodes. As a result, numerous forms of node insertion can be found in a CHFIS.

Keywords: fractal, interpolation, iterated function system, coalescence, node insertion, knot insertion

Procedia PDF Downloads 77

Authors: Hamed Homaei, Seyed Reza Hejazi, Iraj Mahdavi

Abstract:

Return policy (RP) is a strategy for supply chain coordination, whereby the retailer returns the unsold products to the manufacturer or the manufacturer offers a credit on unsold products to the retailer at the end of selling season. Early order commitment (EOC) is another efficient mechanism for channel coordination wherein the retailer commits to purchasing from the manufacturer a fixed order quantity a few periods in advance of the regular delivery lead time. This paper studies the coordination issue of a two-level supply chain with one retailer and one manufacturer through combining two mentioned contracts. The main purpose of this paper is to present an analytical model to show that how the contract which is created by combining RP and EOC can improve supply chain performance. Numerical analyses show that the supply chain coordination through mentioned contract in compare with EOC mechanism, can improve supply chain performance under certain ranges of model parameters. Furthermore, some numerical analyses are done to determine the best buyback price in order to achieve maximum cost saving in the supply chain. Finally, a revenue sharing scheme is presented in order to achieve a win-win condition in the supply chain.

Keywords: supply chain coordination, early order commitment, return policy, revenue sharing

Procedia PDF Downloads 265

Authors: Hanane Belayachi, Sarra Bourahla, Amel Belayachi, Fadela Nemchi, Mostefa Belhakem

Abstract:

The most used techniques to remove pollutants from wastewater are adsorption onto activated carbon (AC) and oxidation using a photocatalyst slurry. The aim of this work is to eliminate such drawbacks by combining AC and titanium dioxide (TiO₂) in a photocatalytically Regenerative Activated Carbon. Anatase titania was deposited on powder-activated carbon made from grape seeds by the impregnation method, and then the composite photocatalyst was employed for the removal of reactive black 5, which is an anionic azo dye, from water. The AGS/TiO₂ was characterized by BET, MEB, RDX and optical absorption spectroscopy. The BET surface area and the pore structure of composite photocatalysts (AGS/TiO₂) and activated grape seeds (AGS) were evaluated from nitrogen adsorption data at 77 K in relation to process conditions. Our results indicate that the photocatalytic activity of AGS/TiO₂ was much higher than single-phase titania. The adsorption equilibrium of reactive black 5 from aqueous solutions on the examined materials was investigated. Langmuir, Freundlich, and Redlich–Petersen models were fitted to experimental equilibrium data, and their goodness of fit is compared. The degradation kinetics fitted well to the Langmuir-Hinselwood pseudo first order rate low. The photocatalytic activity of AGS/TiO₂ was much higher than virgin TiO₂. Chemical oxygen demand (COD) removal was measured at regular intervals to quantify the mineralization of the dye. Above 96% mineralization was observed. These results suggest that UV-irradiated TiO₂ immobilized on activated carbon may be considered an adequate process for the treatment of diluted colored textile wastewater.

Keywords: activated carbon, pollutant, catalysis, TiO₂

Procedia PDF Downloads 16

Authors: Yong Wee Foo, Lai Meng Tang

Abstract:

Artificial Intelligence (AI), particularly the usage of deep neural networks for hierarchical representations from data, has found numerous complex applications across various domains, including computer vision, robotics, autonomous vehicles, and other scientific fields. However, their inherent “black box” nature can sometimes make it challenging for early researchers or school students of various levels to comprehend and trust the results they produce. Consequently, there has been a growing demand for reliable visualization tools in engineering and science education to help learners understand, trust, and explain a deep learning network. This has led to a notable emphasis on the visualization of AI in the research community in recent years. AI visualization tools are increasingly being adopted to significantly improve the comprehension of complex topics in deep learning. This paper proposes a novel approach to empower students to actively explore the inner workings of deep neural networks by combining the student-centered learning approach of flipped classroom models with the investigative power of AI visualization tools, namely, the TensorFlow Playground, the Local Interpretable Model-agnostic Explanations (LIME), and the SHapley Additive exPlanations (SHAP), for delivering an AI education curriculum. Combining the two factors is vital in fostering ownership, responsibility, and critical thinking skills in the age of AI.

Keywords: deep learning, explainable AI, AI visualization, representation learning

Procedia PDF Downloads 40