Search results for: entropy weights analysis

Authors: Niranjana Devi N.

Abstract:

Test case selection is to select the subset of only the fit test cases and remove the unfit, ambiguous, redundant, unnecessary test cases which in turn improve the quality and reduce the cost of software testing. Test cases optimization is the problem of finding the best subset of test cases from a pool of the test cases to be audited. It will meet all the objectives of testing concurrently. But most of the research have evaluated the fitness of test cases only on single parameter fault detecting capability and optimize the test cases using a single objective. In the proposed approach, nine parameters are considered for test case selection and the best subset of parameters for test case selection is obtained using Interval Type-2 Fuzzy Rough Set. Test case selection is done in two stages. The first stage is the fuzzy entropy-based filtration technique, used for estimating and reducing the ambiguity in test case fitness evaluation and selection. The second stage is the ant colony optimization-based wrapper technique with a forward search strategy, employed to select test cases from the reduced test suite of the first stage. The results are evaluated using the Coverage parameters, Precision, Recall, F-Measure, APSC, APDC, and SSR. The experimental evaluation demonstrates that by this approach considerable computational effort can be avoided.

Keywords: ant colony optimization, fuzzy entropy, interval type-2 fuzzy rough set, test case selection

Procedia PDF Downloads 630

Authors: Ramona Zharfpeykan, Paul Rouse

Abstract:

Organizational sustainability is important to both organizations themselves and their stakeholders. Despite its increasing popularity and increasing numbers of organizations reporting sustainability, research on evaluating and comparing the sustainability performance of companies is limited. The aim of this study was to develop models to measure sustainability performance for both cross-sectional and longitudinal comparisons across companies in the same or different industries. A secondary aim was to see if sustainability reports can be used to evaluate sustainability performance. The study used both a content analysis of Australian sustainability reports in mining and metals and financial services for 2011-2014 and a survey of Australian and New Zealand organizations. Two methods ranging from a composite index using uniform weights to data envelopment analysis (DEA) were employed to analyze the data and develop the models. The results show strong statistically significant relationships between the developed models, which suggests that each model provides a consistent, systematic and reasonably robust analysis. The results of the models show that for both industries, companies that had sustainability scores above or below the industry average stayed almost the same during the study period. These indices and models can be used by companies to evaluate their sustainability performance and compare it with previous years, or with other companies in the same or different industries. These methods can also be used by various stakeholders and sustainability ranking companies such as the Global Reporting Initiative (GRI).

Keywords: data envelopment analysis, sustainability, sustainability performance measurement system, sustainability performance index, global reporting initiative

Procedia PDF Downloads 143

Authors: Denis Jordan, Daniel Golkowski, Mathias Lukas, Katharina Merz, Caroline Mlynarcik, Max Maurer, Valentin Riedl, Stefan Foerster, Eberhard F. Kochs, Andreas Bender, Ruediger Ilg

Abstract:

Introduction: So far, clinical assessments that rely on behavioral responses to differentiate coma states or even predict outcome in coma patients are unreliable, e.g. because of some patients’ motor disabilities. The present study was aimed to provide prognosis in coma patients using markers from electroencephalogram (EEG), blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) and [18F]-fluorodeoxyglucose (FDG) positron emission tomography (PET). Unsuperwised principal component analysis (PCA) was used for multimodal integration of markers. Methods: Approved by the local ethics committee of the Technical University of Munich (Germany) 20 patients (aged 18-89) with severe brain damage were acquired through intensive care units at the Klinikum rechts der Isar in Munich and at the Therapiezentrum Burgau (Germany). At the day of EEG/fMRI/PET measurement (date I) patients (<3.5 month in coma) were grouped in the minimal conscious state (MCS) or vegetative state (VS) on the basis of their clinical presentation (coma recovery scale-revised, CRS-R). Follow-up assessment (date II) was also based on CRS-R in a period of 8 to 24 month after date I. At date I, 63 channel EEG (Brain Products, Gilching, Germany) was recorded outside the scanner, and subsequently simultaneous FDG-PET/fMRI was acquired on an integrated Siemens Biograph mMR 3T scanner (Siemens Healthineers, Erlangen Germany). Power spectral densities, permutation entropy (PE) and symbolic transfer entropy (STE) were calculated in/between frontal, temporal, parietal and occipital EEG channels. PE and STE are based on symbolic time series analysis and were already introduced as robust markers separating wakefulness from unconsciousness in EEG during general anesthesia. While PE quantifies the regularity structure of the neighboring order of signal values (a surrogate of cortical information processing), STE reflects information transfer between two signals (a surrogate of directed connectivity in cortical networks). fMRI was carried out using SPM12 (Wellcome Trust Center for Neuroimaging, University of London, UK). Functional images were realigned, segmented, normalized and smoothed. PET was acquired for 45 minutes in list-mode. For absolute quantification of brain’s glucose consumption rate in FDG-PET, kinetic modelling was performed with Patlak’s plot method. BOLD signal intensity in fMRI and glucose uptake in PET was calculated in 8 distinct cortical areas. PCA was performed over all markers from EEG/fMRI/PET. Prognosis (persistent VS and deceased patients vs. recovery to MCS/awake from date I to date II) was evaluated using the area under the curve (AUC) including bootstrap confidence intervals (CI, *: p<0.05). Results: Prognosis was reliably indicated by the first component of PCA (AUC=0.99*, CI=0.92-1.00) showing a higher AUC when compared to the best single markers (EEG: AUC<0.96*, fMRI: AUC<0.86*, PET: AUC<0.60). CRS-R did not show prediction (AUC=0.51, CI=0.29-0.78). Conclusion: In a multimodal analysis of EEG/fMRI/PET in coma patients, PCA lead to a reliable prognosis. The impact of this result is evident, as clinical estimates of prognosis are inapt at time and could be supported by quantitative biomarkers from EEG, fMRI and PET. Due to the small sample size, further investigations are required, in particular allowing superwised learning instead of the basic approach of unsuperwised PCA.

Keywords: coma states and prognosis, electroencephalogram, entropy, functional magnetic resonance imaging, machine learning, positron emission tomography, principal component analysis

Procedia PDF Downloads 310

Authors: R. Stevens, W.L. Bryden

Abstract:

The objective of this study was to determine the efficacy of zeolite in preventing the adverse effects of aflatoxin B1 (AFB1) in broilers. A total of 540 one-day-old Ross chicks were randomly divided into nine treatments, with four replicate pens per treatment and 15 chicks per pen. The treatments included 3 Levels of AFB1 (0,1and 2 mg/kg diet) and 3 levels of zeolite (0, 1.5 and 3 %) in a 3 ×3 factorial arrangement. The experimental treatments commenced on d 7 post-hatch. A starter diet was provided from d 1 to 14, a grower diet from d 15 to 28 and a finisher diet from d 29 to d 49. Diets were based on corn and soybeans and formulated to meet the bird's requirements. The evaluated parameters were as follows: Bodyweight, daily gain, feed intake (FI), feed conversion (FC), relative weights of organs (carcass, liver, heart and abdominal fat) and clinical biochemistry parameters: alanine aminotransferase (ALT) and aspartate aminotransferase (AST). Bodyweight, daily gain and FC were significantly (P<0.05) impaired by aflatoxin. Relative weights of the liver and heart were also affected. The addition of zeolite (1.5 and 3 %) to the contaminated diets ameliorated the effects of aflatoxin, especially at the higher level of inclusion. These data demonstrate that this specific sorbent (zeolite) can protect against the toxicity of AFB1in young broiler chicks.

Keywords: aflatoxin, broiler, toxicity, zeolite

Procedia PDF Downloads 124

Authors: Zohreh Asadi-Shekari, Mehdi Moeinaddini, Muhammad Zaly Shah, Amran Hamzah

Abstract:

Bicycle level of service (BLOS) is a measure for evaluating street conditions for cyclists. Currently, various methods are proposed for BLOS. These analytical methods however have some drawbacks: they usually assume cyclists as users that can share street facilities with motorized vehicles, it is not easy to link them to design process and they are not easy to follow. In addition, they only support a narrow range of cycling facilities and may not be applicable for all situations. Along this, the current paper introduces various effective design factors for bicycle-friendly streets. This study considers cyclists as users of streets who have special needs and facilities. Therefore, the key factors that influence BLOS based on different cycling facilities that are proposed by developed guidelines and literature are identified. The combination of these factors presents a complete set of effective design factors for bicycle-friendly streets. In addition, the weight of each factor in existing BLOS models is estimated and these effective factors are ranked based on these weights. These factors and their weights can be used in further studies to propose special bicycle-friendly street design model.

Keywords: bicycle level of service, bicycle-friendly streets, cycling facilities, rating system, urban streets

Procedia PDF Downloads 462

Authors: Mohammad Saber Eslamlou, Manouchehr Tabibian, Mahta Mirmoghtadaei

Abstract:

The history of urban development and the increasing complexities of urban life have long been intertwined with different natural and man-made disasters. Sometimes, these unpleasant events have destroyed the cities forever. The growth of the urban population and the increase of social and economic resources in the cities increased the importance of developing a holistic approach to dealing with unknown urban disasters. As a result, the interest in resilience has increased in most of the scientific fields, and the urban planning literature has been enriched with the studies of the social, economic, infrastructural, and physical abilities of the cities. In this regard, different conceptual frameworks and patterns have been developed focusing on dimensions of resilience and different kinds of disasters. As the most frequent and likely natural disaster in Iran is flooding, the present study aims to develop a cause-effect model of urban resilience against flood in Karaj City. In this theoretical study, desk research and documentary studies were used to find the elements and dimensions of urban resilience. In this regard, 6 dimensions and 32 elements were found for urban resilience and a questionnaire was made by considering the requirements of TOPSIS techniques (pairwise comparison). The sample of the research consisted of 10 participants who were faculty members, academicians, board members of research centers, managers of the Ministry of Road and Urban Development, board members of New Towns Development Company, experts, and practitioners of consulting companies who had scientific and research backgrounds. The gathered data in this survey were analyzed using TOPSIS and Shannon Entropy techniques. The results show that Infrastructure/Physical, Social, Organizational/ Institutional, Structural/Physical, Economic, and Environmental dimensions are the most effective factors in urban resilience against floods in Karaj, respectively. Finally, a comprehensive model and a systematic framework of factors that affect the urban resilience of Karaj against floods was developed. This cause – effect model shows how different factors are related and influence each other, based on their connected structure and preferences.

Keywords: urban resilience, TOPSIS, Shannon entropy, cause-effect model of resilience, flood

Procedia PDF Downloads 29

Authors: Mohammed Abdullah Alshahrani

Abstract:

The research focuses on investigating the use of partial least squares (PLS) methodology for addressing challenges associated with high-dimensional correlated data. Recent technological advancements have led to experiments producing data characterized by a large number of variables compared to observations, with substantial inter-variable correlations. Such data patterns are common in chemometrics, where near-infrared (NIR) spectrometer calibrations record chemical absorbance levels across hundreds of wavelengths, and in genomics, where thousands of genomic regions' copy number alterations (CNA) are recorded from cancer patients. PLS serves as a widely used method for analyzing high-dimensional data, functioning as a regression tool in chemometrics and a classification method in genomics. It handles data complexity by creating latent variables (components) from original variables. However, applying PLS can present challenges. The study investigates key areas to address these challenges, including unifying interpretations across three main PLS algorithms and exploring unusual negative shrinkage factors encountered during model fitting. The research presents an alternative approach to addressing the interpretation challenge of predictor weights associated with PLS. Sparse estimation of predictor weights is employed using a penalty function combining a lasso penalty for sparsity and a Cauchy distribution-based penalty to account for variable dependencies. The results demonstrate sparse and grouped weight estimates, aiding interpretation and prediction tasks in genomic data analysis. High-dimensional data scenarios, where predictors outnumber observations, are common in regression analysis applications. Ordinary least squares regression (OLS), the standard method, performs inadequately with high-dimensional and highly correlated data. Copy number alterations (CNA) in key genes have been linked to disease phenotypes, highlighting the importance of accurate classification of gene expression data in bioinformatics and biology using regularized methods like PLS for regression and classification.

Keywords: partial least square regression, genetics data, negative filter factors, high dimensional data, high correlated data

Procedia PDF Downloads 9

Authors: Hua Cao, Laurent Peyrodie, Olivier Agnani, Cecile Donze

Abstract:

Multiple sclerosis (MS) is a disease, which affects the central nervous system, and causes balance problem. In clinical, this disorder is usually evaluated using static posturography. Some linear or nonlinear measures, extracted from the posturographic data (i.e. center of pressure, COP) recorded during a balance test, has been used to analyze postural control of MS patients. In this study, the trend (TREND) and the sample entropy (SampEn), two nonlinear parameters were chosen to investigate their relationships with the expanded disability status scale (EDSS) score. Forty volunteers with different EDSS scores participated in our experiments with eyes open (EO) and closed (EC). TREND and two types of SampEn (SampEn1 and SampEn2) were calculated for each combined COP’s position signal. The results have shown that TREND had a weak negative correlation to EDSS while SampEn2 had a strong positive correlation to EDSS. Compared to TREND and SampEn1, SampEn2 showed a better significant correlation to EDSS and an ability to discriminate the MS patients in the EC case. In addition, the outcome of the study suggests that the multi-dimensional nonlinear analysis could provide some information about the impact of disability progression in MS on dynamics of the COP data.

Keywords: balance, multiple sclerosis, nonlinear analysis, postural sway

Procedia PDF Downloads 309

Authors: Mohanad Alhabo, Li Zhang, Naveed Nawaz

Abstract:

Small cell deployment in 5G networks is a promising technology to enhance capacity and coverage. However, unplanned deployment may cause high interference levels and high number of unnecessary handovers, which in turn will result in an increase in the signalling overhead. To guarantee service continuity, minimize unnecessary handovers, and reduce signalling overhead in heterogeneous networks, it is essential to properly model the handover decision problem. In this paper, we model the handover decision according to Multiple Attribute Decision Making (MADM) method, specifically Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS). In this paper, we propose a hybrid TOPSIS method to control the handover in heterogeneous network. The proposed method adopts a hybrid weighting, which is a combination of entropy and standard deviation. A hybrid weighting control parameter is introduced to balance the impact of the standard deviation and entropy weighting on the network selection process and the overall performance. Our proposed method shows better performance, in terms of the number of frequent handovers and the mean user throughput, compared to the existing methods.

Keywords: handover, HetNets, interference, MADM, small cells, TOPSIS, weight

Procedia PDF Downloads 113

Authors: Obaid AlOtaibi, Salman Hussain

Abstract:

Wind energy is associated with many geographical factors including wind speed, climate change, surface topography, environmental impacts, and several economic factors, most notably the advancement of wind technology and energy prices. It is the fastest-growing and least economically expensive method for generating electricity. Wind energy generation is directly related to the characteristics of spatial wind. Therefore, the feasibility study for the wind energy conversion system is based on the value of the energy obtained relative to the initial investment and the cost of operation and maintenance. In Kuwait, wind energy is an appropriate choice as a source of energy generation. It can be used in groundwater extraction in agricultural areas such as Al-Abdali in the north and Al-Wafra in the south, or in fresh and brackish groundwater fields or remote and isolated locations such as border areas and projects away from conventional power electricity services, to take advantage of alternative energy, reduce pollutants, and reduce energy production costs. The study covers the State of Kuwait with an exception of metropolitan area. Climatic data were attained through the readings of eight distributed monitoring stations affiliated with Kuwait Institute for Scientific Research (KISR). The data were used to assess the daily, monthly, quarterly, and annual available wind energy accessible for utilization. The researchers applied the Suitability Model to analyze the study by using the ArcGIS program. It is a model of spatial analysis that compares more than one location based on grading weights to choose the most suitable one. The study criteria are: the average annual wind speed, land use, topography of land, distance from the main road networks, urban areas. According to the previous criteria, the four proposed locations to establish wind farm projects are selected based on the weights of the degree of suitability (excellent, good, average, and poor). The percentage of areas that represents the most suitable locations with an excellent rank (4) is 8% of Kuwait’s area. It is relatively distributed as follows: Al-Shqaya, Al-Dabdeba, Al-Salmi (5.22%), Al-Abdali (1.22%), Umm al-Hayman (0.70%), North Wafra and Al-Shaqeeq (0.86%). The study recommends to decision-makers to consider the proposed location (No.1), (Al-Shqaya, Al-Dabdaba, and Al-Salmi) as the most suitable location for future development of wind farms in Kuwait, this location is economically feasible.

Keywords: Kuwait, renewable energy, spatial analysis, wind energy

Procedia PDF Downloads 105

Authors: R. Shamsi, F. Sharifi

Abstract:

In this paper, we obtain the projection of inefficient units in data envelopment analysis (DEA) in the case of stochastic inputs and outputs using the multi-objective programming (MOP) structure. In some problems, the inputs might be stochastic while the outputs are deterministic, and vice versa. In such cases, we propose a multi-objective DEA-R model because in some cases (e.g., when unnecessary and irrational weights by the BCC model reduce the efficiency score), an efficient decision-making unit (DMU) is introduced as inefficient by the BCC model, whereas the DMU is considered efficient by the DEA-R model. In some other cases, only the ratio of stochastic data may be available (e.g., the ratio of stochastic inputs to stochastic outputs). Thus, we provide a multi-objective DEA model without explicit outputs and prove that the input-oriented MOP DEA-R model in the invariable return to scale case can be replaced by the MOP-DEA model without explicit outputs in the variable return to scale and vice versa. Using the interactive methods for solving the proposed model yields a projection corresponding to the viewpoint of the DM and the analyst, which is nearer to reality and more practical. Finally, an application is provided.

Keywords: DEA-R, multi-objective programming, stochastic data, data envelopment analysis

Procedia PDF Downloads 79

Authors: Yu-Mi Song, Sung-Ah Kim, Dongyoun Shin

Abstract:

Cities are complex systems of diverse and inter-tangled activities. These activities and their complex interrelationships create diverse urban phenomena. And such urban phenomena have considerable influences on the lives of citizens. This research aimed to develop a method to reveal the causes and effects among diverse urban elements in order to enable better understanding of urban activities and, therefrom, to make better urban planning strategies. Specifically, this study was conducted to solve a data-recommendation problem found on a Korean public data homepage. First, a correlation analysis was conducted to find the correlations among random urban data. Then, based on the results of that correlation analysis, the weighted data network of each urban data was provided to people. It is expected that the weights of urban data thereby obtained will provide us with insights into cities and show us how diverse urban activities influence each other and induce feedback.

Keywords: big data, machine learning, ontology model, urban data model

Procedia PDF Downloads 384

Authors: Chukiat Chaiboonsri, Satawat Wannapan

Abstract:

This paper used an asymmetric informative concept to apply in the macroeconomic model estimation of the tourism sector in Thailand. The variables used to statistically analyze are Thailand international and domestic tourism revenues, the expenditures of foreign and domestic tourists, service investments by private sectors, service investments by the government of Thailand, Thailand service imports and exports, and net service income transfers. All of data is a time-series index which was observed between 2002 and 2015. Empirically, the tourism multiplier and accelerator were estimated by two statistical approaches. The first was the result of the Generalized Method of Moments model (GMM) based on the assumption which the tourism market in Thailand had perfect information (Symmetrical data). The second was the result of the Maximum Entropy Bootstrapping approach (MEboot) based on the process that attempted to deal with imperfect information and reduced uncertainty in data observations (Asymmetrical data). In addition, the tourism leakages were investigated by a simple model based on the injections and leakages concept. The empirical findings represented the parameters computed from the MEboot approach which is different from the GMM method. However, both of the MEboot estimation and GMM model suggests that Thailand’s tourism sectors are in a period capable of stimulating the economy.

Keywords: TThailand tourism, Maximum Entropy Bootstrapping approach, macroeconomic model, asymmetric information

Procedia PDF Downloads 269

Authors: Sajjad Baghernezhad

Abstract:

Meta-heuristic and hybrid algorithms have high adeer in modeling medical problems. In this study, a neural network was used to predict covid-19 among high-risk and low-risk patients. This study was conducted to collect the applied method and its target population consisting of 550 high-risk and low-risk patients from the Kerman University of medical sciences medical center to predict the coronavirus. In this study, the memetic algorithm, which is a combination of a genetic algorithm and a local search algorithm, has been used to update the weights of the neural network and develop the accuracy of the neural network. The initial study showed that the accuracy of the neural network was 88%. After updating the weights, the memetic algorithm increased by 93%. For the proposed model, sensitivity, specificity, positive predictivity value, value/accuracy to 97.4, 92.3, 95.8, 96.2, and 0.918, respectively; for the genetic algorithm model, 87.05, 9.20 7, 89.45, 97.30 and 0.967 and for logistic regression model were 87.40, 95.20, 93.79, 0.87 and 0.916. Based on the findings of this study, neural network models have a lower error rate in the diagnosis of patients based on individual variables and vital signs compared to the regression model. The findings of this study can help planners and health care providers in signing programs and early diagnosis of COVID-19 or Corona.

Keywords: COVID-19, decision support technique, neural network, genetic algorithm, memetic algorithm

Procedia PDF Downloads 47

Authors: Hasan Heydari

Abstract:

In recent decades, locating urban services is one of the significant discussions in urban planning. Among these considerations, cities require more accurate planning in order to supply citizen needs, especially part of urban safety. In order to gain this goal, one of the main tasks of urban planners and managers is specifying suitable sites to locate firefighting stations. This study has been done to reach this purpose. Therefore effective criteria consist of coverage radius, population density, proximity to pathway network, land use (compatible and incompatible neighborhood) have been specified. After that, descriptive and local information of the criteria was provided and their layers were created in ArcGIS 9.3. Using Analytic Hierarchy Process (AHP) these criteria and their sub-criteria got the weights. These layers were classified regarding their weights and finally were overlaid by Index Overlay Model and provided the final site selection map for firefighting stations of Arak city. The results gained by analyzing in GIS environment indicate the existing fire station don’t cover the whole city sufficiently and some of the stations have established on the unsuitable sites. The output map indicates the best sites to locate firefighting stations of Arak.

Keywords: site-selection, firefighting stations, analytic hierarchy process (AHP), GIS, index overlay model

Procedia PDF Downloads 326

Authors: Xudong He, Jian Wang, Jiqiang Liu, Lei Han, Yang Yu, Shaohua Lv

Abstract:

Nowadays, the threats of the internet are enormous and increasing; however, the classification of huge alert messages generated in this environment is relatively monotonous. It affects the accuracy of the network situation assessment, and also brings inconvenience to the security managers to deal with the emergency. In order to deal with potential network threats effectively and provide more effective data to improve the network situation awareness. It is essential to build a hierarchical filtering method to prevent the threats. In this paper, it establishes a model for data monitoring, which can filter systematically from the original data to get the grade of threats and be stored for using again. Firstly, it filters the vulnerable resources, open ports of host devices and services. Then use the entropy theory to calculate the performance changes of the host devices at the time of the threat occurring and filter again. At last, sort the changes of the performance value at the time of threat occurring. Use the alerts and performance data collected in the real network environment to evaluate and analyze. The comparative experimental analysis shows that the threat filtering method can effectively filter the threat alerts effectively.

Keywords: correlation analysis, hierarchical filtering, multisource data, network security

Procedia PDF Downloads 176

Authors: Banu Aytül Ekmekçi

Abstract:

Salinity stress has negative effects on agricultural yield throughout the world, affecting production whether it is for subsistence or economic gain. This study investigates the inductive role of vitamin C and its application mode in mitigating the detrimental effects of irrigation with diluted (10, 20 and 30 %) NaCl + water on carthamus tinctorius plants. The results show that 10% of salt water exhibited insignificant changes, while the higher levels impaired growth by reducing seed germination, dry weights of shoot and root, water status and chlorophyll contents. However, irrigation with salt water enhanced carotenoids and antioxidant enzyme activities. The detrimental effects of salt water were ameliorated by application of 100 ppm ascorbic acid (vitamin C). The inductive role of vitamin was associated with the improvement of seed germination, growth, plant water status, carotenoids, endogenous ascorbic acid and antioxidant enzyme activities. Moreover, vitamin C alone or in combination with 30% NaCl water increased the intensity of protein bands as well as synthesized additional new proteins with molecular weights of 205, 87, 84, 65 and 45 kDa. This could increase tolerance mechanisms of treated plants towards water salinity.

Keywords: salinity, stress, vitamin c, antioxidant, NaCl, enzyme

Procedia PDF Downloads 491

Authors: Jannes Quer, Amir Niknejad, Marcus Weber

Abstract:

Computational Drug Design is often based on Molecular Dynamics simulations of molecular systems. Molecular Dynamics can be used to simulate, e.g., the binding and unbinding event of a small drug-like molecule with regard to the active site of an enzyme or a receptor. However, the time-scale of the overall binding event is many orders of magnitude longer than the time-scale of simulation. Thus, there is a need to speed-up molecular simulations. In order to speed up simulations, the molecular dynamics trajectories have to be ”steared” out of local minimizers of the potential energy surface – the so-called metastabilities – of the molecular system. Increasing the kinetic energy (temperature) is one possibility to accelerate simulated processes. However, with temperature the entropy of the molecular system increases, too. But this kind ”stearing” is not directed enough to stear the molecule out of the minimum toward the saddle point. In this article, we give a new mathematical idea, how a potential energy surface can be changed in such a way, that entropy is kept under control while the trajectories are still steared out of the metastabilities. In order to compute the unsteared transition behaviour based on a steared simulation, we propose to use extrapolation methods. In the end we mathematically show, that our method accelerates the simulations along the direction, in which the curvature of the potential energy surface changes the most, i.e., from local minimizers towards saddle points.

Keywords: extrapolation, Eyring-Kramers, metastability, multilevel sampling

Procedia PDF Downloads 296

Authors: Reza Mokhtari Malek Abadi, Mohsen Saghaei, Fatemeh Iman

Abstract:

This paper investigates the indices of a creative city in Isfahan. Its main aim is to evaluate quantitative status of the creative city indices in Isfahan city, analyze the dispersion and distribution of these indices in Isfahan city. Concerning these, this study tries to analyze the creative city indices in fifteen area of Isfahan through secondary data, questionnaire, TOPSIS model, Shannon entropy and SPSS. Based on this, the fifteen areas of Isfahan city have been ranked with 12 factors of creative city indices. The results of studies show that fifteen areas of Isfahan city are not equally benefiting from creative indices and there is much difference between the areas of Isfahan city.

Keywords: grading, creative city, creative city evaluation indicators, regional planning model

Procedia PDF Downloads 434

Authors: Ja-Keoung Koo, Kensuke Nakamura, Hyohun Kim, Dongwha Shin, Yeonseok Kim, Ji-Su Ahn, Byung-Woo Hong

Abstract:

The machine learning techniques based on a convolutional neural network (CNN) have been actively developed and successfully applied to a variety of image analysis tasks including reconstruction, noise reduction, resolution enhancement, segmentation, motion estimation, object recognition. The classical visual information processing that ranges from low level tasks to high level ones has been widely developed in the deep learning framework. It is generally considered as a challenging problem to derive visual interpretation from high dimensional imagery data. A CNN is a class of feed-forward artificial neural network that usually consists of deep layers the connections of which are established by a series of non-linear operations. The CNN architecture is known to be shift invariant due to its shared weights and translation invariance characteristics. However, it is often computationally intractable to optimize the network in particular with a large number of convolution layers due to a large number of unknowns to be optimized with respect to the training set that is generally required to be large enough to effectively generalize the model under consideration. It is also necessary to limit the size of convolution kernels due to the computational expense despite of the recent development of effective parallel processing machinery, which leads to the use of the constantly small size of the convolution kernels throughout the deep CNN architecture. However, it is often desired to consider different scales in the analysis of visual features at different layers in the network. Thus, we propose a CNN model where different sizes of the convolution kernels are applied at each layer based on the random projection. We apply random filters with varying sizes and associate the filter responses with scalar weights that correspond to the standard deviation of the random filters. We are allowed to use large number of random filters with the cost of one scalar unknown for each filter. The computational cost in the back-propagation procedure does not increase with the larger size of the filters even though the additional computational cost is required in the computation of convolution in the feed-forward procedure. The use of random kernels with varying sizes allows to effectively analyze image features at multiple scales leading to a better generalization. The robustness and effectiveness of the proposed CNN based on random kernels are demonstrated by numerical experiments where the quantitative comparison of the well-known CNN architectures and our models that simply replace the convolution kernels with the random filters is performed. The experimental results indicate that our model achieves better performance with less number of unknown weights. The proposed algorithm has a high potential in the application of a variety of visual tasks based on the CNN framework. Acknowledgement—This work was supported by the MISP (Ministry of Science and ICT), Korea, under the National Program for Excellence in SW (20170001000011001) supervised by IITP, and NRF-2014R1A2A1A11051941, NRF2017R1A2B4006023.

Keywords: deep learning, convolutional neural network, random kernel, random projection, dimensionality reduction, object recognition

Procedia PDF Downloads 259

Authors: Sinarwati Mohamad Suhaili, Naomie Salim, Mohamad Nazim Jambli

Abstract:

Sequence-to-sequence (seq2seq) models augmented with attention mechanisms are playing an increasingly important role in automated customer service. These models, which are able to recognize complex relationships between input and output sequences, are crucial for optimizing chatbot responses. Central to these mechanisms are neural attention weights that determine the focus of the model during sequence generation. Despite their widespread use, there remains a gap in the comparative analysis of different attention weighting functions within seq2seq models, particularly in the domain of chatbots using the customer support Twitter (CST) dataset. This study addresses this gap by evaluating four distinct attention-scoring functions -dot, multiplicative/general, additive, and an extended multiplicative function with a tanh activation parameter- in neural generative seq2seq models. Utilizing the CST dataset, these models were trained and evaluated over 10 epochs with the AdamW optimizer. Evaluation criteria included validation loss and BLEU scores implemented under both greedy and beam search strategies with a beam size of k=3. Results indicate that the model with the tanh-augmented multiplicative function significantly outperforms its counterparts, achieving the lowest validation loss (1.136484) and the highest BLEU scores (0.438926 under greedy search, 0.443000 under beam search, k=3). These results emphasize the crucial influence of selecting an appropriate attention-scoring function in improving the performance of seq2seq models for chatbots. Particularly, the model that integrates tanh activation proves to be a promising approach to improve the quality of chatbots in the customer support context.

Keywords: attention weight, chatbot, encoder-decoder, neural generative attention, score function, sequence-to-sequence

Procedia PDF Downloads 48

Authors: Shahin Mirshekari, Mohammadreza Moradi, Hossein Jafari, Mehdi Jafari, Mohammad Ensaf

Abstract:

This research employs Gaussian Process Regression (GPR) with an ensemble kernel, integrating Exponential Squared, Revised Matern, and Rational Quadratic kernels to analyze pharmaceutical sales data. Bayesian optimization was used to identify optimal kernel weights: 0.76 for Exponential Squared, 0.21 for Revised Matern, and 0.13 for Rational Quadratic. The ensemble kernel demonstrated superior performance in predictive accuracy, achieving an R² score near 1.0, and significantly lower values in MSE, MAE, and RMSE. These findings highlight the efficacy of ensemble kernels in GPR for predictive analytics in complex pharmaceutical sales datasets.

Keywords: Gaussian process regression, ensemble kernels, bayesian optimization, pharmaceutical sales analysis, time series forecasting, data analysis

Procedia PDF Downloads 29

Authors: Layla Farhat, Mathieu Bardoux, Stéphane Longuemart, Ziad Herro, Abdelhak Hadj Sahraoui

Abstract:

Electrocaloric (EC) effect refers to the isothermal entropy or adiabatic temperature changes of a dielectric material induced by an external electric field. This phenomenon has been largely ignored for application because only modest EC effects (2.6

Keywords: electrocaloric effect, thermoreflectance, ferroelectricity, cooling system

Procedia PDF Downloads 153

Authors: Guem-Jae Chung, Jin-Hui Lee, Myung-Min Oh

Abstract:

Environmental control of in vitro-propagated apple plantlets is required for successful acclimation to ex vitro due to its low survival rate. This study aimed to determine the proper lighting condition for ex vitro acclimation of the apple plantlets in plant factories. In vitro-propagated M9 apple plantlets treated with pre-acclimatization for 1 week were exposed to following light treatments for additional 6 weeks; 60 μmol·m⁻²·s⁻¹ (A), 100 μmol·m⁻²·s⁻¹ (B), 140 μmol·m⁻²·s⁻¹ (C), 180 μmol·m⁻²·s⁻¹ (D), 60 μmol·m⁻²·s⁻¹ → 100 μmol·m⁻²·s⁻¹ at 2 weeks (E) or 4 weeks (F), 60 μmol·m⁻²·s⁻¹ → 100 μmol·m⁻²·s⁻¹ at 2 weeks → 140 μmol·m⁻²·s⁻¹ at 4 weeks (G) and 60 μmol·m⁻²·s⁻¹ → 140 μmol·m⁻²·s⁻¹ at 4 weeks (H). Shoot height, total leaf area, soil-plant analysis development (SPAD) value, root length, fresh and dry weights of shoots and roots were measured every 2 weeks after transplanting. In addition, the photosynthetic rate was measured at 5 weeks after transplanting. At 6 weeks after transplanting, shoot height of B was significantly higher than the other treatments. SPAD value, total leaf area and root length of B and F were relatively higher than the other treatments. Root fresh weights of B, D, F, and G were relatively higher than those in the other treatments. D induced the highest value in shoot fresh weight probably due to stem hardening, but it also resulted in shoot damage in the early stage of acclimation. Photosynthetic rate at 5 weeks after the transplanting was significantly increased as the light intensity increased. These results suggest that 100 μmol·m⁻²·s⁻¹ for 6 weeks (B) or gradually increased treatment from 60 μmol·m⁻²·s⁻¹ to 140 μmol·m⁻²·s⁻¹ at 2 weeks interval (F) were the proper lighting conditions for successful acclimation of in vitro-propagated apple plantlets. Acknowledgment: This work was supported by Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries (IPET) through Agri-Bio industry Technology Development Program, funded by Ministry of Agriculture, Food and Rural Affairs (MAFRA) (315003051SB020).

Keywords: acclimation, in vitro-propagated apple plantlets, light intensity, plant factory

Procedia PDF Downloads 109

Authors: Selim M. Khan

Abstract:

Indoor air quality is strongly influenced by the presence of radioactive radon (222Rn) gas. Indeed, exposure to high 222Rn concentrations is unequivocally linked to DNA damage and lung cancer and is a worsening issue in North American and European built environments, having increased over time within newer housing stocks as a function of as yet unclear variables. Indoor air radon concentration can be influenced by a wide range of environmental, structural, and behavioral factors. As some of these factors are quantitative while others are qualitative, no single statistical model can determine indoor radon level precisely while simultaneously considering all these variables across a complex and highly diverse dataset. The ability of AI- machine (deep) learning to simultaneously analyze multiple quantitative and qualitative features makes it suitable to predict radon with a high degree of precision. Using Canadian and Swedish long-term indoor air radon exposure data, we are using artificial deep neural network models with random weights and polynomial statistical models in MATLAB to assess and predict radon health risk to human as a function of geospatial, human behavioral, and built environmental metrics. Our initial artificial neural network with random weights model run by sigmoid activation tested different combinations of variables and showed the highest prediction accuracy (>96%) within the reasonable iterations. Here, we present details of these emerging methods and discuss strengths and weaknesses compared to the traditional artificial neural network and statistical methods commonly used to predict indoor air quality in different countries. We propose an artificial deep neural network with random weights as a highly effective method for assessing and predicting indoor radon.

Keywords: radon, radiation protection, lung cancer, aI-machine deep learnng, risk assessment, risk prediction, Europe, North America

Procedia PDF Downloads 73

Authors: Khalid Kaddoura, Tarek Zayed

Abstract:

Infrastructure assets are essential in urban cities; their purpose is to facilitate the public needs. As a result, their conditions and states shall always be monitored to avoid any sudden malfunction. Sewer systems, one of the assets, are an essential part of the underground infrastructure as they transfer sewer medium to designated areas. However, their conditions are subject to deterioration due to ageing. Therefore, it is of great significance to assess the conditions of pipelines to avoid sudden collapses. Current practices of sewer pipeline assessment rely on industrial protocols that consider distinct defects and grades to conclude the limited average or peak score of the assessed assets. This research aims to enhance the evaluation by integrating the Quality Function Deployment (QFD) and the Decision-Making Trial and Evaluation Laboratory (DEMATEL) methods in assessing the condition of sewer pipelines. The methodology shall study the cause and effect relationship of the systems’ defects to deduce the relative influence weights of each defect. Subsequently, the overall grade is calculated by aggregating the WHAT’s and HOW’s of the House of Quality (HOQ) using the computed relative weights. Thus, this study shall enhance the evaluation of the assets to conclude informative rehabilitation and maintenance plans for decision makers.

Keywords: condition assessment, DEMATEL, QFD, sewer pipelines

Procedia PDF Downloads 398

Authors: Smriti Bhatnagar

Abstract:

This study aims to find if the Transit-oriented planning and development approach benefit the quality of air in neighborhoods of New Delhi. Two methodologies, namely the land use regression analysis and the Transit-oriented development index analysis, are being used to explore this relationship. Land Use Regression Analysis makes use of urban form characteristics as obtained for 33 neighborhoods in Delhi. These comprise road lengths, land use areas, population and household densities, number of amenities and distance between amenities. Regressions are run to establish the relationship between urban form variables and air quality parameters (dependent variables). For the Transit-oriented development index analysis, the Transit-oriented Development index is developed as a composite index comprising 29 urban form indicators. This index is developed by assigning weights to each of the 29 urban form data points. Regressions are run to establish the relationship between the Transit-oriented development index and air quality parameters. The thesis finds that elements of Transit-oriented development if incorporated in planning approach, have a positive effect on air quality. Roads suited for non-motorized transport, well connected civic amenities in neighbourhoods, for instance, have a directly proportional relationship with air quality. Transit-oriented development index, however, is not found to have a consistent relationship with air quality parameters. The reason could this, however, be in the way that the index has been constructed.

Keywords: air quality, land use regression, mixed-use planning, transit-oriented development index, New Delhi

Procedia PDF Downloads 238

Authors: Forrest Kaatz, Adhemar Bultheel

Abstract:

We use an informational statistical mechanics approach to study the catalytic thermodynamics of platinum and palladium cuboctahedral nanoclusters. Nanoclusters and their adatoms are viewed as chemical graphs with a nearest neighbor adjacency matrix. We use the Morse potential to determine bond energies between cluster atoms in a coordination type calculation. We use adsorbate energies calculated from density functional theory (DFT) to study the adatom effects on the thermodynamic quantities, which are derived from a Hamiltonian. Oxygen radical and molecular adsorbates are studied on platinum clusters and hydrogen on palladium clusters. We calculate the entropy, free energy, and total energy as the coverage of adsorbates increases from bridge and hollow sites on the surface. Thermodynamic behavior versus adatom coverage is related to the structural distribution of adatoms on the nanocluster surfaces. The thermodynamic functions are characterized using a simple adsorption model, with linear trends as the coverage of adatoms increases. The data exhibits size effects for the measured thermodynamic properties with cluster diameters between 2 and 5 nm. Entropy and enthalpy calculations of Pt-O2 compare well with previous theoretical data for Pt(111)-O2, and our Pd-H results show similar trends as experimental measurements for Pd-H2 nanoclusters. Our methods are general and may be applied to wide variety of nanocluster adsorbate systems.

Keywords: catalytic thermodynamics, palladium nanocluster absorbates, platinum nanocluster absorbates, statistical mechanics

Procedia PDF Downloads 125

Authors: Nihal Murali, Kunal Gupta, Surekha Bhanot

Abstract:

Training of artificial neural networks (ANNs) using reinforcement learning (RL) techniques is being widely discussed in the robot learning literature. The high model complexity of ANNs along with the model-free nature of RL algorithms provides a desirable combination for many robotics applications. There is a huge need for algorithms that generalize using raw sensory inputs, such as vision, without any hand-engineered features or domain heuristics. In this paper, the standard control problem of line following robot was used as a test-bed, and an ANN controller for the robot was trained on images from a live video feed using Q-learning. A virtual agent was first trained in simulation environment and then deployed onto a robot’s hardware. The robot successfully learns to traverse a wide range of curves and displays excellent generalization ability. Qualitative analysis of the evolution of policies, performance and weights of the network provide insights into the nature and convergence of the learning algorithm.

Keywords: artificial neural networks, q-learning, reinforcement learning, robot learning

Procedia PDF Downloads 349

Authors: Todd Zhou, Mikhail Yurochkin

Abstract:

Out-of-distribution (OOD) detection is receiving increasing amounts of attention in the machine learning research community, boosted by recent technologies, such as autonomous driving and image processing. This newly-burgeoning field has called for the need for more effective and efficient methods for out-of-distribution generalization methods. Without accessing the label information, deploying machine learning models to out-of-distribution domains becomes extremely challenging since it is impossible to evaluate model performance on unseen domains. To tackle this out-of-distribution detection difficulty, we designed a model selection pipeline algorithm and developed a model selection framework with different impurity and randomness measurements to evaluate and choose the best-performing models for out-of-distribution data. By exploring different randomness scores based on predicted probabilities, we adopted the out-of-distribution entropy and developed a custom-designed score, ”CombinedScore,” as the evaluation criterion. This proposed score was created by adding labeled source information into the judging space of the uncertainty entropy score using harmonic mean. Furthermore, the prediction bias was explored through the equality of opportunity violation measurement. We also improved machine learning model performance through model calibration. The effectiveness of the framework with the proposed evaluation criteria was validated on the Folktables American Community Survey (ACS) datasets.

Keywords: model selection, domain generalization, model fairness, randomness measurements, bias index

Procedia PDF Downloads 101