Search results for: cluster model approach

Authors: Amartya Hatua, Trung Nguyen, Andrew Sung

Abstract:

In this paper, we study the information diffusion process on Twitter as a multivariate time series problem. Our model concerns three measures (volume, network influence, and sentiment of tweets) based on 10 features, and we collected 27 million tweets to build our information diffusion time series dataset for analysis. Then, different time series clustering techniques with Dynamic Time Warping (DTW) distance were used to identify different patterns of information diffusion. Finally, we built the information diffusion prediction models for new hashtags which comprise two phrases: The first phrase is recognizing the pattern using k-NN with DTW distance; the second phrase is building the forecasting model using the traditional Autoregressive Integrated Moving Average (ARIMA) model and the non-linear recurrent neural network of Long Short-Term Memory (LSTM). Preliminary results of performance evaluation between different forecasting models show that LSTM with clustering information notably outperforms other models. Therefore, our approach can be applied in real-world applications to analyze and predict the information diffusion characteristics of selected topics or memes (hashtags) in Twitter.

Keywords: ARIMA, DTW, information diffusion, LSTM, RNN, time series clustering, time series forecasting, Twitter

Procedia PDF Downloads 391

Authors: Andras Gyorfi, Andras Torma, Adrienn Buruzs

Abstract:

Mobility has been evolved to a determining field of science. The continuously developing segment involves a variety of affected issues such as public and economic sectors. Beside the changes in mobility the state of environment had also changed in the last period. Alternative mobility as a separate category and the idea of its widespread appliance is such a new field that needs to be studied deeper. Alternative mobility implies finding new types of propulsion, using innovative kinds of power and energy resources, revolutionizing the approach to vehicular control. Including new resources and excluding others has such a complex effect which cannot be unequivocally confirmed by today’s scientific achievements. Changes in specific parameters will most likely reduce the environmental impacts, however, the production of new substances or even their subtraction of the system will cause probably energy deficit as well. The aim of this research is to elaborate the environmental impact matrix of alternative mobility and cognize the factors that are yet unknown, analyse them, look for alternative solutions and conclude all the above in a coherent system. In order to this, we analyse it with a method called ‘the system of systems (SoS) method’ to model the effects and the dynamics of the system. A part of the research process is to examine its impacts on the environment, and to decide whether the newly developed versions of alternative mobility are affecting the environmental state. As a final result, a complex approach will be used which can supplement the current scientific studies. By using the SoS approach, we create a framework of reference containing elements in which we examine the interactions as well. In such a way, a flexible and modular model can be established which supports the prioritizing of effects and the deeper analysis of the complex system.

Keywords: environment, alternative mobility, complex model, element analysis, multidimensional map

Procedia PDF Downloads 324

Authors: Nur Aidya Hanum Aizam, Vikneswary Uvaraja

Abstract:

The agenda of showing the scheduled time for performing certain tasks is known as timetabling. It widely used in many departments such as transportation, education, and production. Some difficulties arise to ensure all tasks happen in the time and place allocated. Therefore, many researchers invented various programming model to solve the scheduling problems from several fields. However, the studies in developing the general integer programming model for many timetabling problems are still questionable. Meanwhile, this thesis describe about creating a general model which solve different types of timetabling problems by considering the basic constraints. Initially, the common basic constraints from five different fields are selected and analyzed. A general basic integer programming model was created and then verified by using the medium set of data obtained randomly which is much similar to realistic data. The mathematical software, AIMMS with CPLEX as a solver has been used to solve the model. The model obtained is significant in solving many timetabling problems easily since it is modifiable to all types of scheduling problems which have same basic constraints.

Keywords: AIMMS mathematical software, integer linear programming, scheduling problems, timetabling

Procedia PDF Downloads 436

Authors: Strahinja Kovačević, Sanja Podunavac-Kuzmanović, Lidija Jevrić, Milica Karadžić

Abstract:

The present study is based on the quantitative structure-activity relationship (QSAR) analysis of eighteen compounds with anti-prion activity. The structures and anti-prion activities (expressed in response units, RU%) of the analyzed compounds are taken from CHEMBL database. In the first step of analysis 85 molecular descriptors were calculated and based on them the hierarchical cluster analysis (HCA) and principal component analysis (PCA) were carried out in order to detect potential significant similarities or dissimilarities among the studied compounds. The calculated molecular descriptors were physicochemical, lipophilicity and ADMET (absorption, distribution, metabolism, excretion and toxicity) descriptors. The first stage of the QSAR analysis was simple linear regression modeling. It resulted in one acceptable model that correlates Henry's law constant with RU% units. The obtained 2D-QSAR model was validated by cross-validation as an internal validation method. The validation procedure confirmed the model’s quality and therefore it can be used for prediction of anti-prion activity. The next stage of the analysis of anti-prion activity will include 3D-QSAR and molecular docking approaches in order to select the most promising compounds in treatment of prion diseases. These results are the part of the project No. 114-451-268/2016-02 financially supported by the Provincial Secretariat for Science and Technological Development of AP Vojvodina.

Keywords: anti-prion activity, chemometrics, molecular modeling, QSAR

Procedia PDF Downloads 304

Authors: Levent Basayigit, Sinan Demir, Yusuf Ucar, Burhan Kara

Abstract:

Many research projects require accurate delineation of the different land cover type of the agricultural area. Especially it is critically important for the definition of specific plants like cannabis. However, the complexity of vegetation stands structure, abundant vegetation species, and the smooth transition between different seconder section stages make vegetation classification difficult when using traditional approaches such as the maximum likelihood classifier. Most of the time, classification distinguishes only between trees/annual or grain. It has been difficult to accurately determine the cannabis mixed with other plants. In this paper, a mixed distribution models approach is applied to classify pure and mix cannabis parcels using Worldview-2 imagery in the Lakes region of Turkey. Five different land use types (i.e. sunflower, maize, bare soil, and cannabis) were identified in the image. A constrained Gaussian mixture discriminant analysis (GMDA) was used to unmix the image. In the study, 255 reflectance ratios derived from spectral signatures of seven bands (Blue-Green-Yellow-Red-Rededge-NIR1-NIR2) were randomly arranged as 80% for training and 20% for test data. Gaussian mixed distribution model approach is proved to be an effective and convenient way to combine very high spatial resolution imagery for distinguishing cannabis vegetation. Based on the overall accuracies of the classification, the Gaussian mixed distribution model was found to be very successful to achieve image classification tasks. This approach is sensitive to capture the illegal cannabis planting areas in the large plain. This approach can also be used for monitoring and determination with spectral reflections in illegal cannabis planting areas.

Keywords: Gaussian mixture discriminant analysis, spectral mixture model, Worldview-2, land parcels

Procedia PDF Downloads 197

Authors: Noor Ismael, Lisa Mische Lawson, Lauren Little, Murad Moqbel

Abstract:

Background/Significance: Caregivers of children with Autism Spectrum Disorders (ASD) develop coping mechanisms to overcome daily challenges to successfully parent their child. There is variability in coping strategies used among caregivers of children with ASD. Capturing homogeneity among such variable groups may help elucidate targeted intervention approaches for caregivers of children with ASD. Study Purpose: This study aimed to identify groups of caregivers of children with ASD based on coping mechanisms, and to examine whether there are differences among these groups in terms of strain level. Methods: This study utilized a secondary data analysis, and included survey responses of 273 caregivers of children with ASD. Measures consisted of the COPE Inventory and the Caregiver Strain Questionnaire. Data analyses consisted of cluster analysis to group caregiver coping strategies, and analysis of variance to compare the caregiver coping groups on strain level. Results: Cluster analysis results showed four distinct groups with different combinations of coping strategies: Social-Supported/Planning (group one), Spontaneous/Reactive (group two), Self-Supporting/Reappraisal (group three), and Religious/Expressive (group four). Caregivers in group one (Social-Supported/Planning) demonstrated significantly higher levels than the remaining three groups in the use of the following coping strategies: planning, use of instrumental social support, and use of emotional social support, relative to the other three groups. Caregivers in group two (Spontaneous/Reactive) used less restraint relative to the other three groups, and less suppression of competing activities relative to the other three groups as coping strategies. Also, group two showed significantly lower levels of religious coping as compared to the other three groups. In contrast to group one, caregivers in group three (Self-Supporting/Reappraisal) demonstrated significantly lower levels of the use of instrumental social support and the use of emotional social support relative to the other three groups. Additionally, caregivers in group three showed more acceptance, positive reinterpretation and growth coping strategies. Caregivers in group four (Religious/Expressive) demonstrated significantly higher levels of religious coping relative to the other three groups and utilized more venting of emotions strategies. Analysis of Variance results showed no significant differences between the four groups on the strain scores. Conclusions: There are four distinct groups with different combinations of coping strategies: Social-Supported/Planning, Spontaneous/Reactive, Self-Supporting/Reappraisal, and Religious/Expressive. Each caregiver group engaged in a combination of coping strategies to overcome the strain of caregiving.

Keywords: autism, caregivers, cluster analysis, coping strategies

Procedia PDF Downloads 282

Authors: Riyadh Jamegh, Alla Eldin Kassam, Sawsan Sabih

Abstract:

In order to overcome uncertainty conditions and inability to meet customers' requests due to these conditions, organizations tend to reserve a certain safety stock level (SSL). This level must be chosen carefully in order to avoid the increase in holding cost due to excess in SSL or shortage cost due to too low SSL. This paper used soft computing fuzzy logic to identify optimal SSL; this fuzzy model uses the dynamic concept to cope with high complexity environment status. The proposed model can deal with three input variables, i.e., demand stability level, raw material availability level, and on hand inventory level by using dynamic fuzzy logic to obtain the best SSL as an output. In this model, demand stability, raw material, and on hand inventory levels are described linguistically and then treated by inference rules of the fuzzy model to extract the best level of safety stock. The aim of this research is to provide dynamic approach which is used to identify safety stock level, and it can be implanted in different industries. Numerical case study in the dairy industry with Yogurt 200 gm cup product is explained to approve the validity of the proposed model. The obtained results are compared with the current level of safety stock which is calculated by using the traditional approach. The importance of the proposed model has been demonstrated by the significant reduction in safety stock level.

Keywords: inventory optimization, soft computing, safety stock optimization, dairy industries inventory optimization

Procedia PDF Downloads 125

Authors: Liudmyla Koliechkina, Olena Dvirna

Abstract:

The statement of the multi-objective optimization problem on combinatorial configurations is formulated, and the approach to its solution is proposed. The problem is of interest as a combinatorial optimization one with many criteria, which is a model of many applied tasks. The approach to solving the multi-objective optimization problem on combinatorial configurations consists of two stages; the first is the reduction of the multi-objective problem to the single criterion based on existing multi-objective optimization methods, the second stage solves the directly replaced single criterion combinatorial optimization problem by the horizontal combinatorial method. This approach provides the optimal solution to the multi-objective optimization problem on combinatorial configurations, taking into account additional restrictions for a finite number of steps.

Keywords: discrete set, linear combinatorial optimization, multi-objective optimization, Pareto solutions, partial permutation set, structural graph

Procedia PDF Downloads 167

Authors: Nkongho Ayuketang Arreyndip, Ebobenow Joseph

Abstract:

In this paper, to model a real life wind turbine, a probabilistic approach is proposed to model the dynamics of the blade elements of a small axial wind turbine under extreme stochastic wind speeds conditions. It was found that the power and the torque probability density functions even though decreases at these extreme wind speeds but are not infinite. Moreover, we also found that it is possible to stabilize the power coefficient (stabilizing the output power) above rated wind speeds by turning some control parameters. This method helps to explain the effect of turbulence on the quality and quantity of the harness power and aerodynamic torque.

Keywords: probability, probability density function, stochastic, turbulence

Procedia PDF Downloads 586

Authors: Peiming Li

Abstract:

This paper presents an innovative blockchain-based approach to enhance the reliability and efficiency of federated learning systems. By integrating a decentralized oracle consensus mechanism into the federated learning framework, we address key challenges of data and model integrity. Our approach utilizes a network of redundant oracles, functioning as independent validators within an epoch-based training system in the federated learning model. In federated learning, data is decentralized, residing on various participants' devices. This scenario often leads to concerns about data integrity and model quality. Our solution employs blockchain technology to establish a transparent and tamper-proof environment, ensuring secure data sharing and aggregation. The decentralized oracles, a concept borrowed from blockchain systems, act as unbiased validators. They assess the contributions of each participant using a Hidden Markov Model (HMM), which is crucial for evaluating the consistency of participant inputs and safeguarding against model poisoning and malicious activities. Our methodology's distinct feature is its epoch-based training. An epoch here refers to a specific training phase where data is updated and assessed for quality and relevance. The redundant oracles work in concert to validate data updates during these epochs, enhancing the system's resilience to security threats and data corruption. The effectiveness of this system was tested using the Mnist dataset, a standard in machine learning for benchmarking. Results demonstrate that our blockchain-oriented federated learning approach significantly boosts system resilience, addressing the common challenges of federated environments. This paper aims to make these advanced concepts accessible, even to those with a limited background in blockchain or federated learning. We provide a foundational understanding of how blockchain technology can revolutionize data integrity in decentralized systems and explain the role of oracles in maintaining model accuracy and reliability.

Keywords: federated learning system, block chain, decentralized oracles, hidden markov model

Procedia PDF Downloads 63

Authors: Hycham Aboutaleb, Bruno Monsuez

Abstract:

Current systems complexity has reached a degree that requires addressing conception and design issues while taking into account all the necessary aspects. Therefore, one of the main challenges is the way complex systems are specified and designed. The exponential growing effort, cost and time investment of complex systems in modeling phase emphasize the need for a paradigm, a framework and a environment to handle the system model complexity. For that, it is necessary to understand the expectations of the human user of the model and his limits. This paper presents a generic framework for designing complex systems, highlights the requirements a system model needs to fulfill to meet human user expectations, and defines the refined functional as well as non functional requirements modeling tools needs to meet to be useful in model-based system engineering.

Keywords: system modeling, modeling language, modeling requirements, framework

Procedia PDF Downloads 531

Authors: William J. Crowther, Conor Marsh

Abstract:

Object-based flow simulation allows fast computation of arbitrarily complex aerodynamic models made up of simple objects with limited flow interactions. The proposed approach is universally applicable to objects made from arbitrarily scaled ellipsoid primitives at arbitrary aerodynamic attitude and angular rate. The use of a component-based aerodynamic modelling approach increases efficiency by allowing selective inclusion of different physics models at run-time and allows extensibility through the development of new models. Insight into the numerical stability of the model under first order fixed-time step integration schemes is provided by stability analysis of the drag component. The compute cost of model components and functions is evaluated and compared against numerical benchmarks. Model static outputs are verified against theoretical expectations and dynamic behaviour using falling plate data from the literature. The model is applied to a range of case studies to demonstrate the efficacy of its application in extensibility, ease of use, and low computational cost. Dynamically complex multi-body systems can be implemented in a transparent and efficient manner, and we successfully demonstrate large scenes with hundreds of objects interacting with diverse flow fields.

Keywords: aerodynamics, real-time simulation, low-order model, flight dynamics

Procedia PDF Downloads 102

Authors: Jia-Lien Hsu, Chiung-Wen Chang

Abstract:

Research articles may follow a simple and succinct structure of organizational patterns, called move. For example, considering extended abstracts, we observe that an extended abstract usually consists of five moves, including Background, Aim, Method, Results, and Conclusion. As another example, when publishing articles in PubMed, authors are encouraged to provide a structured abstract, which is an abstract with distinct and labeled sections (e.g., Introduction, Methods, Results, Discussions) for rapid comprehension. This paper introduces a method for computational analysis of move structures (i.e., Background-Purpose-Method-Result-Conclusion) in abstracts and introductions of research documents, instead of manually time-consuming and labor-intensive analysis process. In our approach, sentences in a given abstract and introduction are automatically analyzed and labeled with a specific move (i.e., B-P-M-R-C in this paper) to reveal various rhetorical status. As a result, it is expected that the automatic analytical tool for move structures will facilitate non-native speakers or novice writers to be aware of appropriate move structures and internalize relevant knowledge to improve their writing. In this paper, we propose a Bayesian approach to determine move tags for research articles. The approach consists of two phases, training phase and testing phase. In the training phase, we build a Bayesian model based on a couple of given initial patterns and the corpus, a subset of CiteSeerX. In the beginning, the priori probability of Bayesian model solely relies on initial patterns. Subsequently, with respect to the corpus, we process each document one by one: extract features, determine tags, and update the Bayesian model iteratively. In the testing phase, we compare our results with tags which are manually assigned by the experts. In our experiments, the promising accuracy of the proposed approach reaches 56%.

Keywords: academic English writing, assisted writing, move tag analysis, Bayesian approach

Procedia PDF Downloads 330

Authors: Ka C. Lam, Oluwafunmibi S. Idowu

Abstract:

Building girth is important in building economics and mostly used in quantities take-off of various cost items. Literature suggests that the use of conceptual quantities can improve the accuracy of cost models. Girth or perimeter of a building can be used to estimate conceptual quantities. Hence, the current paper aims to model the perimeter-area function of buildings shapes for use at the conceptual design stage. A detailed literature review on existing building shape indexes was carried out. An empirical approach was used to study the relationship between area and the shortest length of a four-sided orthogonal polygon. Finally, a mathematical approach was used to establish the observed relationships. The empirical results obtained were in agreement with the mathematical model developed. A new equation termed “conceptual perimeter equation” is proposed. The equation can be used to estimate building envelope quantities such as external wall area, external finishing area and scaffolding area before sketch or detailed drawings are prepared.

Keywords: building envelope, building shape index, conceptual quantities, cost modelling, girth

Procedia PDF Downloads 342

Authors: Delphine Ribes Lemay, Valentine Bernasconi, André Andrade, Lara DéFayes, Mathieu Salzmann, FréDéRic Kaplan, Nicolas Henchoz

Abstract:

Digitization and preservation of large heritage induce high maintenance costs to keep up with the technical standards and ensure sustainable access. Creating impactful usage is instrumental to justify the resources for long-term preservation. The Museum für Gestaltung of Zurich holds one of the biggest poster collections of the world from which 52’000 were digitised. In the process of building a digital installation to valorize the collection, one objective was to develop an algorithm capable of predicting the next poster to show according to the ones already displayed. The work presented here describes the steps to build an algorithm able to automatically create sequences of posters reflecting associations performed by curator and professional designers. The exposed challenge finds similarities with the domain of song playlist algorithms. Recently, artificial intelligence techniques and more specifically, deep-learning algorithms have been used to facilitate their generations. Promising results were found thanks to Recurrent Neural Networks (RNN) trained on manually generated playlist and paired with clusters of extracted features from songs. We used the same principles to create the proposed algorithm but applied to a challenging medium, posters. First, a convolutional autoencoder was trained to extract features of the posters. The 52’000 digital posters were used as a training set. Poster features were then clustered. Next, an RNN learned to predict the next cluster according to the previous ones. RNN training set was composed of poster sequences extracted from a collection of books from the Gestaltung Museum of Zurich dedicated to displaying posters. Finally, within the predicted cluster, the poster with the best proximity compared to the previous poster is selected. The mean square distance between features of posters was used to compute the proximity. To validate the predictive model, we compared sequences of 15 posters produced by our model to randomly and manually generated sequences. Manual sequences were created by a professional graphic designer. We asked 21 participants working as professional graphic designers to sort the sequences from the one with the strongest graphic line to the one with the weakest and to motivate their answer with a short description. The sequences produced by the designer were ranked first 60%, second 25% and third 15% of the time. The sequences produced by our predictive model were ranked first 25%, second 45% and third 30% of the time. The sequences produced randomly were ranked first 15%, second 29%, and third 55% of the time. Compared to designer sequences, and as reported by participants, model and random sequences lacked thematic continuity. According to the results, the proposed model is able to generate better poster sequencing compared to random sampling. Eventually, our algorithm is sometimes able to outperform a professional designer. As a next step, the proposed algorithm should include a possibility to create sequences according to a selected theme. To conclude, this work shows the potentiality of artificial intelligence techniques to learn from existing content and provide a tool to curate large sets of data, with a permanent renewal of the presented content.

Keywords: Artificial Intelligence, Digital Humanities, serendipity, design research

Procedia PDF Downloads 184

Authors: Imran Hashmi, Sipkaduwa Arachchige Sashika Sureni Wickramasooriya

Abstract:

Gene-drive technology has emerged as a promising tool for disease control by influencing the population dynamics of disease-carrying organisms. Various gene drive mechanisms, derived from global laboratory experiments, aim to strategically manage and prevent the spread of targeted diseases. One prominent strategy involves population replacement, wherein genetically modified mosquitoes are introduced to replace the existing local wild population. To enhance our understanding and aid in the design of effective release strategies, we employ a comprehensive mathematical model. The utilized approach employs agent-based modeling, enabling the consideration of individual mosquito attributes and flexibility in parameter manipulation. Through the integration of an agent-based model and a meta-population spatial approach, the dynamics of gene drive mosquito spreading in a released site are simulated. The model's outcomes offer valuable insights into future population dynamics, providing guidance for the development of informed release strategies. This research significantly contributes to the ongoing discourse on the responsible and effective implementation of gene drive technology for disease vector control.

Keywords: gene drive, agent-based modeling, disease-carrying organisms, malaria

Procedia PDF Downloads 65

Authors: G. Shabib, Esam H. Abd-Elhameed, G. Magdy

Abstract:

In this paper, a comparison of discrete time PID, PSS controllers is presented through small signal stability of power system comprising of one machine connected to infinite bus system. This comparison achieved by using a new approach of discretization which converts the S-domain model of analog controllers to a Z-domain model to enhance the damping of a single machine power system. The new method utilizes the Plant Input Mapping (PIM) algorithm. The proposed algorithm is stable for any sampling rate, as well as it takes the closed loop characteristic into consideration. On the other hand, the traditional discretization methods such as Tustin’s method is produce satisfactory results only; when the sampling period is sufficiently low.

Keywords: PSS, power system stabilizer PID, proportional-integral-derivative PIM, plant input mapping

Procedia PDF Downloads 505

Authors: A. Benbouali, H. Saidi, A. Derrouazin, T. Bessaad

Abstract:

Aerial robotics is a very exciting research field dealing with a variety of subjects, including the attitude control. This paper deals with the control of a four rotor vertical take-off and landing (VTOL) Unmanned Aerial Vehicle. The paper presents a mathematical model based on the approach of Lagrange for the flight control of an autonomous quad-rotor. It also describes the controller architecture which is based on PID regulators. The control method has been simulated in closed loop in different situations. All the calculation stages and the simulation results have been detailed.

Keywords: quad-rotor, lagrange approach, proportional integral derivate (PID) controller, Matlab/Simulink

Procedia PDF Downloads 400

Authors: Jorge Anibal Restrepo Morales, Sonia Martín Gómez

Abstract:

A model based on a logical framework was developed to quantify SMEs' internationalization capacity. To do so, linguistic variables, such as human talent, infrastructure, innovation strategies, FTAs, marketing strategies, finance, etc. were integrated. It is argued that a company’s management of international markets depends on internal factors, especially capabilities and resources available. This study considers internal factors as the biggest business challenge because they force companies to develop an adequate set of capabilities. At this stage, importance and strategic relevance have to be defined in order to build competitive advantages. A fuzzy inference system is proposed to model the resources, skills, and capabilities that determine the success of internationalization. Data: 157 linguistic variables were used. These variables were defined by international trade entrepreneurs, experts, consultants, and researchers. Using expert judgment, the variables were condensed into18 factors that explain SMEs’ export capacity. The proposed model is applied by means of a case study of the textile and clothing cluster in Medellin, Colombia. In the model implementation, a general index of 28.2 was obtained for internationalization capabilities. The result confirms that the sector’s current capabilities and resources are not sufficient for a successful integration into the international market. The model specifies the factors and variables, which need to be worked on in order to improve export capability. In the case of textile companies, the lack of a continuous recording of information stands out. Likewise, there are very few studies directed towards developing long-term plans, and., there is little consistency in exports criteria. This method emerges as an innovative management tool linked to internal organizational spheres and their different abilities.

Keywords: business strategy, exports, internationalization, fuzzy set methods

Procedia PDF Downloads 294

Authors: Kazuyoshi Mori

Abstract:

In this paper, we consider the parametrization of Anantharam’s model within the framework of the factorization approach. In the parametrization, we investigate the number of required parameters of Anantharam’s model. We consider single-input single-output systems in this paper. By the investigation, we find three cases that are (1) there exist plants which require only one parameter and (2) two parameters, and (3) the number of parameters is at most three.

Keywords: linear systems, parametrization, coprime factorization, number of parameters

Procedia PDF Downloads 213

Authors: Ayliana Dharmawan, Heng Yong Sheng, Zhang Xiaojin, Tan Thai Lian

Abstract:

To facilitate discharge planning, doctors are presently required to assign an Estimated Discharge Date (EDD) for each patient admitted to the hospital. This assignment of the EDD is largely based on the doctor’s judgment. This can be difficult for cases which are complex or relatively new to the doctor. It is hypothesized that a data-driven approach would be able to facilitate the doctors to make accurate estimations of the discharge date. Making use of routinely collected data on inpatient discharges between January 2013 and May 2016, a predictive model was developed using machine learning techniques to predict the Length of Stay (and hence the EDD) of inpatients, at the point of admission. The predictive performance of the model was compared to that of the clinicians using accuracy measures. Overall, the best performing model was found to be able to predict EDD with an accuracy improvement in Average Squared Error (ASE) by -38% as compared to the first EDD determined by the present method. It was found that important predictors of the EDD include the provisional diagnosis code, patient’s age, attending doctor at admission, medical specialty at admission, accommodation type, and the mean length of stay of the patient in the past year. The predictive model can be used as a tool to accurately predict the EDD.

Keywords: inpatient, estimated discharge date, EDD, prediction, data-driven

Procedia PDF Downloads 174

Authors: Khenfer Koummich Fatma, Hendel Fatiha, Mesbahi Larbi

Abstract:

This paper presents an approach based on Hidden Markov Models (HMM: Hidden Markov Model) using HTK tools. The goal is to create a human-machine interface with a voice recognition system that allows the operator to teleoperate a mentor robot to execute specific tasks as rotate, raise, close, etc. This system should take into account different levels of environmental noise. This approach has been applied to isolated words representing the robot commands pronounced in two languages: French and Arabic. The obtained recognition rate is the same in both speeches, Arabic and French in the neutral words. However, there is a slight difference in favor of the Arabic speech when Gaussian white noise is added with a Signal to Noise Ratio (SNR) equals 30 dB, in this case; the Arabic speech recognition rate is 69%, and the French speech recognition rate is 80%. This can be explained by the ability of phonetic context of each speech when the noise is added.

Keywords: Arabic speech recognition, Hidden Markov Model (HMM), HTK, noise, TIMIT, voice command

Procedia PDF Downloads 385

Authors: A. Amar

Abstract:

A new model, namely the crystal model, has been modified to calculate the radius and density distribution of light nuclei up to ⁸Be. The crystal model has been modified according to solid-state physics, which uses the analogy between nucleon distribution and atoms distribution in the crystal. The model has analytical analysis to calculate the radius where the density distribution of light nuclei has obtained from analogy of crystal lattice. The distribution of nucleons over crystal has been discussed in a general form. The equation that has been used to calculate binding energy was taken from the solid-state model of repulsive and attractive force. The numbers of the protons were taken to control repulsive force, where the atomic number was responsible for the attractive force. The parameter has been calculated from the crystal model was found to be proportional to the radius of the nucleus. The density distribution of light nuclei was taken as a summation of two clusters distribution as in ⁶Li=alpha+deuteron configuration. A test has been done on the data obtained for radius and density distribution using double folding for d+⁶,⁷Li with M3Y nucleon-nucleon interaction. Good agreement has been obtained for both the radius and density distribution of light nuclei. The model failed to calculate the radius of ⁹Be, so modifications should be done to overcome discrepancy.

Keywords: nuclear physics, nuclear lattice, study nucleus as crystal, light nuclei till to ⁸Be

Procedia PDF Downloads 176

Authors: Abdoullah Nyassi, Golam Sarwar, Sarra Baldeh, Mamadou S. K. Jallow, Bai Lamin Dondeh, Isaac Osei, Grant A. Mackenzie

Abstract:

A randomized controlled trial is the "gold standard" for evaluating the efficacy of an intervention. Large-scale, cluster-randomized trials are expensive and difficult to conduct, though. To guarantee the validity and generalizability of findings, high-quality, dependable, and accurate data management systems are necessary. Robust data management systems are crucial for optimizing and validating the quality, accuracy, and dependability of trial data. Regarding the difficulties of data gathering in clinical trials in low-resource areas, there is a scarcity of literature on this subject, which may raise concerns. Effective data management systems and implementation goals should be part of trial procedures. Publicizing the creative clinical data management techniques used in clinical trials should boost public confidence in the study's conclusions and encourage further replication. In the ongoing pneumococcal vaccine schedule study in rural Gambia, this report details the development and deployment of multi-data management systems and methodologies. We implemented six different data management, synchronization, and reporting systems using Microsoft Access, RedCap, SQL, Visual Basic, Ruby, and ASP.NET. Additionally, data synchronization tools were developed to integrate data from these systems into the central server for reporting systems. Clinician, lab, and field data validation systems and methodologies are the main topics of this report. Our process development efforts across all domains were driven by the complexity of research project data collected in real-time data, online reporting, data synchronization, and ways for cleaning and verifying data. Consequently, we effectively used multi-data management systems, demonstrating the value of creative approaches in enhancing the consistency, accuracy, and reporting of trial data in a poor resource setting.

Keywords: data management, data collection, data cleaning, cluster-randomized trial

Procedia PDF Downloads 27

Authors: Prakash Singh

Abstract:

The focus of this study is on the use of the SWEAT analysis approach to determine how effectively a school, as an organization, has implemented its curriculum. To gauge the feelings of the teaching staff, unstructured interviews were employed in this study, asking the participants for their ideas and opinions on each of the three identified aspects of the school: instructional materials, media and technology; teachers’ professional competencies; and the curriculum. This investigation was based on the five key components of the SWEAT model: strengths, weaknesses, expectations, abilities, and tensions. The findings of this exploratory study evoke the significance of the SWEAT achievement model as a tool for strategic analysis to be undertaken in any organization. The findings further affirm the usefulness of this analytical tool for human resource development. Employees have expectations, but competency gaps in their professional abilities may hinder them from fulfilling their tasks in terms of their job description. Also, tensions in the working environment can contribute to their experiences of tobephobia (fear of failure). The SWEAT analysis approach detects such shortcomings in any organization and can therefore culminate in the development of programmes to address such concerns. The strategic SWEAT analysis process can provide a clear distinction between success and failure, and between mediocrity and excellence in organizations. However, more research needs to be done on the effectiveness of the SWEAT analysis approach as a strategic analytical tool.

Keywords: SWEAT analysis, strategic analysis, tobephobia, competency gaps

Procedia PDF Downloads 507

Authors: Saad M. Darwish, Mohamed A. El-Iskandarani, Guitar M. Shawkat

Abstract:

Nowadays, the amount of available multimedia data is continuously on the rise. The need to find a required image for an ordinary user is a challenging task. Content based image retrieval (CBIR) computes relevance based on the visual similarity of low-level image features such as color, textures, etc. However, there is a gap between low-level visual features and semantic meanings required by applications. The typical method of bridging the semantic gap is through the automatic image annotation (AIA) that extracts semantic features using machine learning techniques. In this paper, a multi-label image annotation system guided by Firefly and Bayesian method is proposed. Firstly, images are segmented using the maximum variance intra cluster and Firefly algorithm, which is a swarm-based approach with high convergence speed, less computation rate and search for the optimal multiple threshold. Feature extraction techniques based on color features and region properties are applied to obtain the representative features. After that, the images are annotated using translation model based on the Net Bayes system, which is efficient for multi-label learning with high precision and less complexity. Experiments are performed using Corel Database. The results show that the proposed system is better than traditional ones for automatic image annotation and retrieval.

Keywords: feature extraction, feature selection, image annotation, classification

Procedia PDF Downloads 586

Authors: E. Asadollahi-Yazdi, J. Gardan, P. Lafon

Abstract:

Nowadays, manufactures are encountered with production of different version of products due to quality, cost and time constraints. On the other hand, Additive Manufacturing (AM) as a production method based on CAD model disrupts the design and manufacturing cycle with new parameters. To consider these issues, the researchers utilized Design For Manufacturing (DFM) approach for AM but until now there is no integrated approach for design and manufacturing of product through the AM. So, this paper aims to provide a general methodology for managing the different production issues, as well as, support the interoperability with AM process and different Product Life Cycle Management tools. The problem is that the models of System Engineering which is used for managing complex systems cannot support the product evolution and its impact on the product life cycle. Therefore, it seems necessary to provide a general methodology for managing the product’s diversities which is created by using AM. This methodology must consider manufacture and assembly during product design as early as possible in the design stage. The latest approach of DFM, as a methodology to analyze the system comprehensively, integrates manufacturing constraints in the numerical model in upstream. So, DFM for AM is used to import the characteristics of AM into the design and manufacturing process of a hybrid product to manage the criteria coming from AM. Also, the research presents an integrated design method in order to take into account the knowledge of layers manufacturing technologies. For this purpose, the interface model based on the skin and skeleton concepts is provided, the usage and manufacturing skins are used to show the functional surface of the product. Also, the material flow and link between the skins are demonstrated by usage and manufacturing skeletons. Therefore, this integrated approach is a helpful methodology for designer and manufacturer in different decisions like material and process selection as well as, evaluation of product manufacturability.

Keywords: additive manufacturing, 3D printing, design for manufacturing, integrated design, interoperability

Procedia PDF Downloads 316

Authors: Hsiao-Lan Wang, Chun-Han Chiang, I-Chen Chen

Abstract:

In Chinese Mandarin, lexical tones play an important role to provide contrasts in word meaning. They are pitch patterns and can be quantified as the fundamental frequency (F0), expressed in Hertz (Hz). In this study, we aim to investigate the influence of frequency discrimination on Chinese children’s performance of reading abilities. Fifty participants from 3rd to 4th grades, including 24 children with reading difficulties and 26 age-matched children, were examined. A serial of cognitive, language, reading and psychoacoustic tests were administrated. Magnetoencephalography (MEG) was also employed to study children’s auditory sensitivity. In the present study, auditory frequency was measured through slide-up pitch, slide-down pitch and frequency-modulated tone. The results showed that children with Chinese reading difficulties were significantly poor at phonological awareness and auditory discrimination for the identification of frequency-modulated tone. Chinese children’s character reading performance was significantly related to lexical tone awareness and auditory perception of frequency-modulated tone. In our MEG measure, we compared the mismatch negativity (MMNm), from 100 to 200 ms, in two groups. There were no significant differences between groups during the perceptual discrimination of standard sounds, fast-up and fast-down frequencies. However, the data revealed significant cluster differences between groups in the slow-up and slow-down frequencies discrimination. In the slow-up stimulus, the cluster demonstrated an upward field map at 106-151 ms (p < .001) with a strong peak time at 127ms. The source analyses of two dipole model and localization resolution model (CLARA) from 100 to 200 ms both indicated a strong source from the left temporal area with 45.845% residual variance. Similar results were found in the slow-down stimulus with a larger upward current at 110-142 ms (p < 0.05) and a peak time at 117 ms in the left temporal area (47.857% residual variance). In short, we found a significant group difference in the MMNm while children processed frequency-modulated tones with slow temporal changes. The findings may imply that perception of sound frequency signals with slower temporal modulations was related to reading and language development in Chinese. Our study may also support the recent hypothesis of underlying non-verbal auditory temporal deficits accounting for the difficulties in literacy development seen developmental dyslexia.

Keywords: Chinese Mandarin, frequency modulation sweeps, magnetoencephalography, mismatch negativity, reading difficulties

Procedia PDF Downloads 576

Authors: Fabrice Setephin Atedjio, Jean-Pierre Lienou, Frederica F. Nelson, Sachin S. Shetty

Abstract:

Malicious users exploit vulnerabilities in computer systems, significantly disrupting their performance and revealing the inadequacies of existing protective solutions. Even machine learning-based approaches, designed to ensure reliability, can be compromised by adversarial attacks that undermine their robustness. This paper addresses two critical aspects of enhancing model reliability. First, we focus on improving model performance and robustness against adversarial threats. To achieve this, we propose a strategy by harnessing deep reinforcement learning. Second, we introduce an approach leveraging generative adversarial networks to counter adversarial attacks effectively. Our results demonstrate substantial improvements over previous works in the literature, with classifiers exhibiting enhanced accuracy in classification tasks, even in the presence of adversarial perturbations. These findings underscore the efficacy of the proposed model in mitigating intrusions and adversarial attacks within the machine learning landscape.

Keywords: machine learning, reliability, adversarial attacks, deep-reinforcement learning, robustness

Procedia PDF Downloads 8

Authors: Odd Ivar Haugen

Abstract:

Society increasingly relies on complex systems whose behaviour is determined, not by the properties of each part, but by the interaction between them. The behaviour of such systems is emergent. Modelling emergent system behaviour requires a systems approach that incorporates the necessary concepts that are capable of determining such behaviour. The CESM metamodel is a model of system models. A set of system models needs to address the elements of CESM at different levels of abstraction to be able to model the behaviour of a complex system. Modern ships contain numerous sophisticated equipment, often accompanied by a local safety system to protect its integrity. These control systems are then connected into a larger integrated system in order to achieve the ship’s objective or mission. The integrated system becomes what is commonly known as a system of systems, which can be termed a complex system. Examples of such complex systems are the dynamic positioning system and the power management system. Three ship accidents are provided as examples of how system complexity may contribute to accidents. Then, the three accidents are discussed in terms of how the Multi-Level/Multi-Model Safety Analysis might catch scenarios such as those leading to the accidents described.

Keywords: emergent properties, CESM metamodel, multi-level/multi-model safety analysis, safety, system complexity, system models, systems thinking

Procedia PDF Downloads 5