Search results for: K-nearest neighbors algorithm

Authors: H. Barati, M. Wu, A. Kharicha, A. Ludwig

Abstract:

A transient model for nozzle clogging has been developed and successfully validated against a laboratory experiment. Key steps of clogging are considered: transport of particles by turbulent flow towards the nozzle wall; interactions between fluid flow and nozzle wall, and the adhesion of the particle on the wall; the growth of the clog layer and its interaction with the flow. The current paper is to investigate the mesh (size and type) sensitivity of the model in both two and three dimensions. It is found that the algorithm for clog growth alone excluding the flow effect is insensitive to the mesh type and size, but the calculation including flow becomes sensitive to the mesh quality. The use of 2D meshes leads to overestimation of the clog growth because the 3D nature of flow in the boundary layer cannot be properly solved by 2D calculation. 3D simulation with tetrahedron mesh can also lead to an error estimation of the clog growth. A mesh-independent result can be achieved with hexahedral mesh, or at least with triangular prism (inflation layer) for near-wall regions.

Keywords: clogging, continuous casting, inclusion, simulation, submerged entry nozzle

Procedia PDF Downloads 283

Authors: N. Ploysook, R. Rugsaj, C. Suvanjumrat

Abstract:

The characteristic requirement for producing rectangular shape bottles was a uniform thickness of the plastic bottle wall. Die shaping was a good technique which controlled the wall thickness of bottles. An advance technology which was the finite element method (FEM) for blowing parison to be a rectangular shape bottle was conducted to reduce waste plastic from a trial and error method of a die shaping and parison control method. The artificial intelligent (AI) comprised of artificial neural network and genetic algorithm was selected to optimize the die gap shape from the FEM results. The application of AI technique could optimize the suitable die gap shape for the parison blow molding which did not depend on the parison control method to produce rectangular bottles with the uniform wall. Particularly, this application can be used with cheap blow molding machines without a parison controller therefore it will reduce cost of production in the bottle blow molding process.

Keywords: AI, bottle, die shaping, FEM

Procedia PDF Downloads 238

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 372

Authors: Qiao Qiao

Abstract:

With the development of The Times, the elderly demand for pension service facilities is increasing. Taking 26 street towns in Jiangjin District of Chongqing as examples, ArcGIS spatial analysis method was used to analyze the distribution status of the elderly population, the core density of the elderly population, and the spatial layout characteristics of institutional elderly care facilities in Jiangjin District of Chongqing. The results showed that there were differences in the structure and aging degree of the elderly population in each street town. There is a certain imbalance between the spatial distribution of the elderly population and the planning and construction of elderly care facilities. The accessibility of elderly care facilities is uneven. Therefore, a genetic algorithm is used to optimize the spatial layout of institutional elderly care facilities, improve the accessibility of facilities, strengthen the participation of multiple subjects, and provide a reference for the future construction planning of elderly care facilities.

Keywords: institutional pension facilities, spatial layout, accessibility, ArcGIS

Procedia PDF Downloads 9

Authors: Berengere Vire, Nicolas Salvetat, Yoann Lannay, Guillaume Marcellin, Siem Van Der Laan, Franck Molina, Dinah Weissmann

Abstract:

Predicting suicidal behaviors is one of the most complex challenges of daily psychiatric practices. Today, suicide risk prediction using biological tools is not validated and is only based on subjective clinical reports of the at-risk individual. Therefore, there is a great need to identify biomarkers that would allow early identification of individuals at risk of suicide. Alterations of adenosine-to-inosine (A-to-I) RNA editing of neurotransmitter receptors and other proteins have been shown to be involved in etiology of different psychiatric disorders and linked to suicidal behavior. RNA editing is a co- or post-transcriptional process leading to a site-specific alteration in RNA sequences. It plays an important role in the epi transcriptomic regulation of RNA metabolism. On postmortem human brain tissue (prefrontal cortex) of depressed suicide victims, Alcediag found specific alterations of RNA editing activity on the mRNA coding for the serotonin 2C receptor (5-HT2cR). Additionally, an increase in expression levels of ADARs, the RNA editing enzymes, and modifications of RNA editing profiles of prime targets, such as phosphodiesterase 8A (PDE8A) mRNA, have also been observed. Interestingly, the PDE8A gene is located on chromosome 15q25.3, a genomic region that has recurrently been associated with the early-onset major depressive disorder (MDD). In the current study, we examined whether modifications in RNA editing profile of prime targets allow identifying disease-relevant blood biomarkers and evaluating suicide risk in patients. To address this question, we performed a clinical study to identify an RNA editing signature in blood of depressed patients with and without the history of suicide attempts. Patient’s samples were drawn in PAXgene tubes and analyzed on Alcediag’s proprietary RNA editing platform using next generation sequencing technology. In addition, gene expression analysis by quantitative PCR was performed. We generated a multivariate algorithm comprising various selected biomarkers to detect patients with a high risk to attempt suicide. We evaluated the diagnostic performance using the relative proportion of PDE8A mRNA editing at different sites and/or isoforms as well as the expression of PDE8A and the ADARs. The significance of these biomarkers for suicidality was evaluated using the area under the receiver-operating characteristic curve (AUC). The generated algorithm comprising the biomarkers was found to have strong diagnostic performances with high specificity and sensitivity. In conclusion, we developed tools to measure disease-specific biomarkers in blood samples of patients for identifying individuals at the greatest risk for future suicide attempts. This technology not only fosters patient management but is also suitable to predict the risk of drug-induced psychiatric side effects such as iatrogenic increase of suicidal ideas/behaviors.

Keywords: blood biomarker, next-generation-sequencing, RNA editing, suicide

Procedia PDF Downloads 259

Authors: Norm Josephy, Lucy Kimball, Victoria Steblovskaya

Abstract:

Hedging of a portfolio of European options under proportional transaction costs is considered. Our discrete time financial market model extends the binomial market model with transaction costs to the case where the underlying stock price ratios are distributed over a bounded interval rather than over a two-point set. An optimal hedging strategy is chosen from a set of admissible non-self-financing hedging strategies. Our approach to optimal hedging of a portfolio of options is based on theoretical foundation that includes determination of a no-arbitrage option price interval as well as on properties of the non-self-financing strategies and their residuals. A computational algorithm for optimizing an investor relevant criterion over the set of admissible non-self-financing hedging strategies is developed. Applicability of our approach is demonstrated using both simulated data and real market data.

Keywords: extended binomial model, non-self-financing hedging, optimization, proportional transaction costs

Procedia PDF Downloads 252

Authors: Thinh Ngo, Paul Rad, Brian Kelley

Abstract:

Noise estimation is essential in today wireless systems for power control, adaptive modulation, interference suppression and quality of service. Deep learning (DL) has already been applied in the physical layer for modulation and signal classifications. Unacceptably low accuracy of less than 50% is found to undermine traditional application of DL classification for SNR prediction. In this paper, we use divide-and-conquer algorithm and classifier fusion method to simplify SNR classification and therefore enhances DL learning and prediction. Specifically, multiple CNNs are used for classification rather than a single CNN. Each CNN performs a binary classification of a single SNR with two labels: less than, greater than or equal. Together, multiple CNNs are combined to effectively classify over a range of SNR values from −20 ≤ SNR ≤ 32 dB.We use pre-trained CNNs to predict SNR over a wide range of joint channel parameters including multiple Doppler shifts (0, 60, 120 Hz), power-delay profiles, and signal-modulation types (QPSK,16QAM,64-QAM). The approach achieves individual SNR prediction accuracy of 92%, composite accuracy of 70% and prediction convergence one order of magnitude faster than that of traditional estimation.

Keywords: classification, CNN, deep learning, prediction, SNR

Procedia PDF Downloads 134

Authors: B. Chandra

Abstract:

In Association Rule Mining, much attention has been paid for developing algorithms for large (frequent/closed/maximal) itemsets but very little attention has been paid to improve the performance of rule generation algorithms. Rule generation is an important part of Association Rule Mining. In this paper, a novel approach named NARG (Association Rule using Antecedent Support) has been proposed for rule generation that uses memory resident data structure named FCET (Frequent Closed Enumeration Tree) to find frequent/closed itemsets. In addition, the computational speed of NARG is enhanced by giving importance to the rules that have lower antecedent support. Comparative performance evaluation of NARG with fast association rule mining algorithm for rule generation has been done on synthetic datasets and real life datasets (taken from UCI Machine Learning Repository). Performance analysis shows that NARG is computationally faster in comparison to the existing algorithms for rule generation.

Keywords: knowledge discovery, association rule mining, antecedent support, rule generation

Procedia PDF Downloads 324

Authors: Md. Aminul Islam, Adel Merabet, Rachid Beguenane, Hussein Ibrahim

Abstract:

This paper presents a simulation and mathematical model of stand-alone solar-wind-diesel based hybrid energy system (HES). A power management system is designed for multiple energy resources in a stand-alone hybrid energy system. Both Solar photovoltaic and wind energy conversion system consists of maximum power point tracking (MPPT), voltage regulation, and basic power electronic interfaces. An additional diesel generator is included to support and improve the reliability of stand-alone system when renewable energy sources are not available. A power management strategy is introduced to distribute the generated power among resistive load banks. The frequency regulation is developed with conventional phase locked loop (PLL) system. The power management algorithm was applied in Matlab®/Simulink® to simulate the results.

Keywords: solar photovoltaic, wind energy, diesel engine, hybrid energy system, power management, frequency and voltage regulation

Procedia PDF Downloads 454

Authors: Gebremeskel Hagos Gebremedhin, Feng Chong, Heyan Huang

Abstract:

Software system is the combination of architecture and organized components to accomplish a specific function or set of functions. A good software architecture facilitates application system development, promotes achievement of functional requirements, and supports system reconfiguration. We describe three studies demonstrating the utility of our architecture in the subdomain of mobile office robots and identify software engineering principles embodied in the architecture. The main aim of this paper is to analyze prove architecture design and automatic re-architecting using machine learning. Intelligence software architecture and automatic re-architecting process is reorganizing in to more suitable one of the software organizational structure system using the user access dataset for creating relationship among the components of the system. The 3-step approach of data mining was used to analyze effective recovery, transformation and implantation with the use of clustering algorithm. Therefore, automatic re-architecting without changing the source code is possible to solve the software complexity problem and system software reuse.

Keywords: intelligence, software architecture, re-architecting, software reuse, High level design

Procedia PDF Downloads 119

Authors: Andrii Natochii

Abstract:

The UA9 collaboration in the last two years has installed two different types of detectors to investigate the channeling effect in the bent silicon crystals with high-energy particles beam on the CERN accelerator facilities: Cherenkov detector CpFM and silicon pixel detector Timepix. In the current work, we describe the main performances of the Timepix detector operation at the SPS and H8 extracted beamline at CERN. We are presenting some detector calibration results and tuning. Our research topics also cover a cluster analysis algorithm for the particle hits reconstruction. We describe the optimal acquisition setup for the Timepix device and the edges of its functionality for the high energy and flux beam monitoring. The measurements of the crystal parameters are very important for the future bent crystal applications and needs a track reconstruction apparatus. Thus, it was decided to construct a short range (1.2 m long) particle telescope based on the Timepix sensors and test it at H8 SPS extraction beamline. The obtained results will be shown as well.

Keywords: beam monitoring, channeling, particle tracking, Timepix detector

Procedia PDF Downloads 180

Authors: Leidy M. Aldana, Jorge E. Camargo

Abstract:

Adversarial Machine Learning has gained importance in recent years as Cybersecurity has gained too, especially malware, it has affected different entities and people in recent years. This paper shows a literature review about defense methods created to prevent adversarial machine learning attacks, firstable it shows an introduction about the context and the description of some terms, in the results section some of the attacks are described, focusing on detecting adversarial examples before coming to the machine learning algorithm and showing other categories that exist in defense. A method with five steps is proposed in the method section in order to define a way to make the literature review; in addition, this paper summarizes the contributions in this research field in the last seven years to identify research directions in this area. About the findings, the category with least quantity of challenges in defense is the Detection of adversarial examples being this one a viable research route with the adaptive approach in attack and defense.

Keywords: Malware, adversarial, machine learning, defense, attack

Procedia PDF Downloads 63

Authors: Lei Zhang, Tingting Xu, Jun He, Zhenyu Yan

Abstract:

Email marketing is one of the most important segments of online marketing. It has been proved to be the most effective way to acquire and retain customers. The email content is vital to customers. Different customers may have different familiarity with a product, so a successful marketing strategy must personalize email content based on individual customers’ product affinity. In this study, we build our personalized email marketing strategy with three types of emails: nurture, promotion, and conversion. Each type of email has a different influence on customers. We investigate this difference by analyzing customers’ open rates, click rates and opt-out rates. Feature importance from response models is also analyzed. The goal of the marketing strategy is to improve the click rate on conversion-type emails. To build the personalized strategy, we formulate the problem as a reinforcement learning problem and adopt a Q-learning algorithm with variations. The simulation results show that our model-based strategy outperforms the current marketer’s strategy.

Keywords: email marketing, email content, reinforcement learning, machine learning, Q-learning

Procedia PDF Downloads 194

Authors: Ali Asheibi

Abstract:

The rapid increase in the size of databases required to store power quality monitoring data has demanded new techniques for analysing and understanding the data. One suggested technique to assist in analysis is data mining. Preparing raw data to be ready for data mining exploration take up most of the effort and time spent in the whole data mining process. Clustering is an important technique in data mining and machine learning in which underlying and meaningful groups of data are discovered. Large amounts of harmonic data have been collected from an actual harmonic monitoring system in a distribution system in Australia for three years. This amount of acquired data makes it difficult to identify operational events that significantly impact the harmonics generated on the system. In this paper, harmonic data preparation processes to better understanding of the data have been presented. Underlying classes in this data has then been identified using clustering technique based on the Minimum Message Length (MML) method. The underlying operational information contained within the clusters can be rapidly visualised by the engineers. The C5.0 algorithm was used for classification and interpretation of the generated clusters.

Keywords: data mining, harmonic data, clustering, classification

Procedia PDF Downloads 248

Authors: Mohamad Khairi Ishak

Abstract:

Conventional public key crypto systems such as RSA (Ron Rivest, Adi Shamir and Leonard Adleman), DSA (Digital Signature Algorithm), and Elgamal are no longer efficient to be implemented in the small, memory constrained devices. Elliptic Curve Cryptography (ECC), which allows smaller key length as compared to conventional public key crypto systems, has thus become a very attractive choice for many applications. This paper describes implementation of an elliptic curve cryptography (ECC) encryption engine on a FPGA. The system has been implemented in 2 different key sizes, which are 131 bits and 163 bits. Area and timing analysis are provided for both key sizes for comparison. The crypto system, which has been implemented on Altera’s EPF10K200SBC600-1, has a hardware size of 5945/9984 and 6913/9984 of logic cells for 131 bits implementation and 163 bits implementation respectively. The crypto system operates up to 43 MHz, and performs point multiplication operation in 11.3 ms for 131 bits implementation and 14.9 ms for 163 bits implementation. In terms of speed, our crypto system is about 8 times faster than the software implementation of the same system.

Keywords: elliptic curve cryptography, FPGA, key sizes, memory

Procedia PDF Downloads 321

Authors: Yu T. Tsai, Jin H. Huang

Abstract:

In this paper, the specific sound transmission loss (TL) of the laminated composite plate (LCP) with different material properties in each layer is investigated. The numerical method to obtain the TL of the LCP is proposed by using elastic plate theory. The transfer matrix approach is novelty presented for computational efficiency in solving the numerous layers of dynamic stiffness matrix (D-matrix) of the LCP. Besides the numerical simulations for calculating the TL of the LCP, the material properties inverse method is presented for the design of a laminated composite plate analogous to a metallic plate with a specified TL. As a result, it demonstrates that the proposed computational algorithm exhibits high efficiency with a small number of iterations for achieving the goal. This method can be effectively employed to design and develop tailor-made materials for various applications.

Keywords: sound transmission loss, laminated composite plate, transfer matrix approach, inverse problem, elastic plate theory, material properties

Procedia PDF Downloads 388

Authors: Gaurav Sancheti

Abstract:

This study presents an impacting approach of Artificial Neural Networks (ANNs) in determining the quantity of High Tensile Steel (HTS) strands required in post-tensioned (PT) slabs. Various PT slab configurations were generated by varying the span and depth of the slab. For each of these slab configurations, quantity of required HTS strands were recorded. ANNs with backpropagation algorithm and varying architectures were developed and their performance was evaluated in terms of Mean Square Error (MSE). The recorded data for the quantity of HTS strands was used as a feeder database for training the developed ANNs. The networks were validated using various validation techniques. The results show that the proposed ANNs have a great potential with good prediction and generalization capability.

Keywords: artificial neural networks, back propagation, conceptual design, high tensile steel strands, post tensioned slabs, validation techniques

Procedia PDF Downloads 221

Authors: Ryan Messina, Mehedi Hasan

Abstract:

This research is to examine the impacts of using data to generate performance requirements for automation in visual inspections using machine vision. These situations are intended for design and how projects can smooth the transfer of tacit knowledge to using an algorithm. We have proposed a framework when specifying machine vision systems. This framework utilizes varying levels of automation as contingency planning to reduce data processing complexity. Using data assists in extracting tacit knowledge from those who can perform the manual tasks to assist design the system; this means that real data from the system is always referenced and minimizes errors between participating parties. We propose using three indicators to know if the project has a high risk of failing to meet requirements related to accuracy and reliability. All systems tested achieved a better integration into operations after applying the framework.

Keywords: automation, contingency planning, continuous engineering, control theory, machine vision, system requirements, system thinking

Procedia PDF Downloads 204

Authors: Amin Ranjbar, Amir Arsalan Jalili Zolfaghari, Amir Abolfazl Suratgar, Mehrdad Khajavi

Abstract:

This paper describes two stages of learning-based fault detection procedure in alternators. The procedure consists of three states of machine condition namely shortened brush, high impedance relay and maintaining a healthy condition in the alternator. The fault detection algorithm uses Wigner-Ville distribution as a feature extractor and also appropriate feature classifier. In this work, ANN (Artificial Neural Network) and also SVM (support vector machine) were compared to determine more suitable performance evaluated by the mean squared of errors criteria. Modules work together to detect possible faulty conditions of machines working. To test the method performance, a signal database is prepared by making different conditions on a laboratory setup. Therefore, it seems by implementing this method, satisfactory results are achieved.

Keywords: alternator, artificial neural network, support vector machine, time-frequency analysis, Wigner-Ville distribution

Procedia PDF Downloads 374

Authors: Ammar Alali, Mahmoud Abughaban

Abstract:

Cost reduction and drilling optimization is the goal of many drilling operators. Historically, stuck pipe incidents were a major segment of non-productive time (NPT) associated costs. Traditionally, stuck pipe problems are part of the operations and solved post-sticking. However, the real key to savings and success is in predicting the stuck pipe incidents and avoiding the conditions leading to its occurrences. Previous attempts in stuck-pipe predictions have neglected the local geology of the problem. The proposed predictive tool utilizes geophysical data processing techniques and Machine Learning (ML) algorithms to predict drilling activities events in real-time using surface drilling data with minimum computational power. The method combines two types of analysis: (1) real-time prediction, and (2) cause analysis. Real-time prediction aggregates the input data, including historical drilling surface data, geological formation tops, and petrophysical data, from wells within the same field. The input data are then flattened per the geological formation and stacked per stuck-pipe incidents. The algorithm uses two physical methods (stacking and flattening) to filter any noise in the signature and create a robust pre-determined pilot that adheres to the local geology. Once the drilling operation starts, the Wellsite Information Transfer Standard Markup Language (WITSML) live surface data are fed into a matrix and aggregated in a similar frequency as the pre-determined signature. Then, the matrix is correlated with the pre-determined stuck-pipe signature for this field, in real-time. The correlation used is a machine learning Correlation-based Feature Selection (CFS) algorithm, which selects relevant features from the class and identifying redundant features. The correlation output is interpreted as a probability curve of stuck pipe incidents prediction in real-time. Once this probability passes a fixed-threshold defined by the user, the other component, cause analysis, alerts the user of the expected incident based on set pre-determined signatures. A set of recommendations will be provided to reduce the associated risk. The validation process involved feeding of historical drilling data as live-stream, mimicking actual drilling conditions, of an onshore oil field. Pre-determined signatures were created for three problematic geological formations in this field prior. Three wells were processed as case studies, and the stuck-pipe incidents were predicted successfully, with an accuracy of 76%. This accuracy of detection could have resulted in around 50% reduction in NPT, equivalent to 9% cost saving in comparison with offset wells. The prediction of stuck pipe problem requires a method to capture geological, geophysical and drilling data, and recognize the indicators of this issue at a field and geological formation level. This paper illustrates the efficiency and the robustness of the proposed cross-disciplinary approach in its ability to produce such signatures and predicting this NPT event.

Keywords: drilling optimization, hazard prediction, machine learning, stuck pipe

Procedia PDF Downloads 228

Authors: Shiuh-Jer Huang, Su-Shean Chen, Sheam-Chyun Lin

Abstract:

Two-wheel inverted pendulum robot (TWIPR) is designed with two-hub DC motors for human riding and motion control evaluation. In order to measure the tilt angle and angular velocity of the inverted pendulum robot, accelerometer and gyroscope sensors are chosen. The mobile robot’s moving position and velocity were estimated based on DC motor built in hall sensors. The control kernel of this electric mobile robot is designed with embedded Arduino Nano microprocessor. A handle bar was designed to work as steering mechanism. The intelligent model-free fuzzy sliding mode control (FSMC) was employed as the main control algorithm for this mobile robot motion monitoring with different control purpose adjustment. The intelligent controllers were designed for balance control, and moving speed control purposes of this robot under different operation conditions and the control performance were evaluated based on experimental results.

Keywords: balance control, speed control, intelligent controller, two wheel inverted pendulum

Procedia PDF Downloads 224

Authors: Gyoutae Park, Seungho Han, Byungduk Kim, Youngdo Jo, Yongsop Shim, Yeonjae Lee, Sangguk Ahn, Hiesik Kim, Jungil Park

Abstract:

In this paper, we presented not only development technology of an explosion proof type and portable combustible gas leak detector but also algorithm to improve accuracy for measuring gas concentrations. The presented techniques are to apply the flame-proof enclosure and intrinsic safe explosion proof to an infrared gas leak detector at first in Korea and to improve accuracy using linearization recursion equation and Lagrange interpolation polynomial. Together, we tested sensor characteristics and calibrated suitable input gases and output voltages. Then, we advanced the performances of combustible gaseous detectors through reflecting demands of gas safety management fields. To check performances of two company's detectors, we achieved the measurement tests with eight standard gases made by Korea Gas Safety Corporation. We demonstrated our instruments better in detecting accuracy other than detectors through experimental results.

Keywords: accuracy improvement, IR gas sensor, gas leak, detector

Procedia PDF Downloads 391

Authors: Shahbaz G. Hassan

Abstract:

Water quality models are very important to predict the changes in surface water quality for environmental management. The aim of this paper is to give an overview of the water qualities, and to provide directions for selecting models in specific situation. Water quality models include one kind of model based on a mechanistic approach, while other models simulate water quality without considering a mechanism. Mechanistic models can be widely applied and have capabilities for long-time simulation, with highly complexity. Therefore, more spaces are provided to explain the principle and application experience of mechanistic models. Mechanism models have certain assumptions on rivers, lakes and estuaries, which limits the application range of the model, this paper introduces the principles and applications of water quality model based on the above three scenarios. On the other hand, mechanistic models are more easily to compute, and with no limit to the geographical conditions, but they cannot be used with confidence to simulate long term changes. This paper divides the empirical models into two broad categories according to the difference of mathematical algorithm, models based on artificial intelligence and models based on statistical methods.

Keywords: empirical models, mathematical, statistical, water quality

Procedia PDF Downloads 264

Authors: Sıdıka Kaya, Gülay Sain Güven, Seda Karsavuran, Onur Toka

Abstract:

This study is part of an ongoing research project supported by the Scientific and Technological Research Council of Turkey (TUBITAK) under Project Number 114K404, and participation to this conference was supported by Hacettepe University Scientific Research Coordination Unit under Project Number 10243. Evaluation of hospital readmissions is gaining importance in terms of quality and cost, and is becoming the target of national policies. In Turkey, the topic of hospital readmission is relatively new on agenda and very few studies have been conducted on this topic. The aim of this study was to determine 30-day readmission rates and risk factors for readmission. Whether readmission was planned, related to the prior admission and avoidable or not was also assessed. The study was designed as a ‘prospective cohort study.’ 472 patients hospitalized in internal medicine departments of a university hospital in Turkey between February 1, 2015 and April 30, 2015 were followed up. Analyses were conducted using IBM SPSS Statistics version 22.0 and SPSS Modeler 16.0. Average age of the patients was 56 and 56% of the patients were female. Among these patients 95 were readmitted. Overall readmission rate was calculated as 20% (95/472). However, only 31 readmissions were unplanned. Unplanned readmission rate was 6.5% (31/472). Out of 31 unplanned readmission, 24 was related to the prior admission. Only 6 related readmission was avoidable. To determine risk factors for readmission we constructed Chi-square automatic interaction detector (CHAID) decision tree algorithm. CHAID decision trees are nonparametric procedures that make no assumptions of the underlying data. This algorithm determines how independent variables best combine to predict a binary outcome based on ‘if-then’ logic by portioning each independent variable into mutually exclusive subsets based on homogeneity of the data. Independent variables we included in the analysis were: clinic of the department, occupied beds/total number of beds in the clinic at the time of discharge, age, gender, marital status, educational level, distance to residence (km), number of people living with the patient, any person to help his/her care at home after discharge (yes/no), regular source (physician) of care (yes/no), day of discharge, length of stay, ICU utilization (yes/no), total comorbidity score, means for each 3 dimensions of Readiness for Hospital Discharge Scale (patient’s personal status, patient’s knowledge, and patient’s coping ability) and number of daycare admissions within 30 days of discharge. In the analysis, we included all 95 readmitted patients (46.12%), but only 111 (53.88%) non-readmitted patients, although we had 377 non-readmitted patients, to balance data. The risk factors for readmission were found as total comorbidity score, gender, patient’s coping ability, and patient’s knowledge. The strongest identifying factor for readmission was comorbidity score. If patients’ comorbidity score was higher than 1, the risk for readmission increased. The results of this study needs to be validated by other data–sets with more patients. However, we believe that this study will guide further studies of readmission and CHAID is a useful tool for identifying risk factors for readmission.

Keywords: decision tree, hospital, internal medicine, readmission

Procedia PDF Downloads 256

Authors: Abdelghani Alidra, Mohamed Tahar Kimour

Abstract:

Genetic algorithm must adapt themselves at design time to cope with the search problem specific requirements and at runtime to balance exploration and convergence objectives. In a previous article, we have shown that modeling and implementing Genetic Algorithms (GA) using the software product line (SPL) paradigm is very appreciable because they constitute a product family sharing a common base of code. In the present article we propose to extend the use of the feature model of the genetic algorithms family to model the potential states of the GA in what is called a Dynamic Software Product Line. The objective of this paper is the systematic generation of a reconfigurable architecture that supports the dynamic of the GA and which is easily deduced from the feature model. The resultant GA is able to perform dynamic reconfiguration autonomously to fasten the convergence process while producing better solutions. Another important advantage of our approach is the exploitation of recent advances in the domain of dynamic SPLs to enhance the performance of the GAs.

Keywords: self-adaptive genetic algorithms, software engineering, dynamic software product lines, reconfigurable architecture

Procedia PDF Downloads 285

Authors: Amir Sharifi Miavaghi

Abstract:

It is believed recombination mechnisms in graphene-bonded perovskite solar cells based on numerical model in which doped-graphene structures are employed as anode/cathode bonding semiconductor. Moreover, th‌‌‌‌e da‌‌‌‌‌rk-li‌‌‌‌‌ght c‌‌‌‌urrent d‌‌‌‌ens‌‌‌‌ity-vo‌‌‌‌‌‌‌ltage density-voltage cu‌‌‌‌‌‌‌‌‌‌‌rves are investigated by regression analysis. L‌‌‌oss m‌‌‌‌echa‌‌‌‌nisms suc‌‌‌h a‌‌‌‌‌‌s ba‌‌‌‌ck c‌‌‌ontact b‌‌‌‌‌arrier, d‌‌‌‌eep surface defect i‌‌‌‌n t‌‌‌‌‌‌‌he adsorbent la‌‌‌yer is det‌‌‌‌‌ermined b‌‌‌y adapting th‌‌‌e sim‌‌‌‌‌ulated ce‌‌‌‌‌ll perfor‌‌‌‌‌mance to t‌‌‌‌he measure‌‌‌‌ments us‌‌‌‌ing the diffe‌‌‌‌‌‌rential evolu‌‌‌‌‌tion of th‌‌‌‌e global optimization algorithm. T‌‌‌‌he performance of t‌‌‌he c‌‌‌‌ell i‌‌‌‌n the connection proc‌‌‌‌‌ess incl‌‌‌‌‌‌udes J-V cur‌‌‌‌‌‌ves that are examined at di‌‌‌‌‌fferent tempe‌‌‌‌‌‌‌ratures an‌‌‌d op‌‌‌‌en cir‌‌‌‌cuit vol‌‌‌‌tage (V) und‌‌‌‌er differ‌‌‌‌‌ent light intensities as a function of temperature. Ba‌‌‌‌sed o‌‌‌n t‌‌‌he prop‌‌‌‌osed nu‌‌‌‌‌merical mod‌‌‌‌el a‌‌‌‌nd the acquired lo‌‌‌‌ss mecha‌‌‌‌‌‌nisms, our approach can be used to improve the efficiency of the solar cell further. Due to the high demand for alternative energy sources, solar cells are good alternatives for energy storage using the photovoltaic phenomenon.

Keywords: numerical model, recombination mechanism, graphen, perovskite solarcell

Procedia PDF Downloads 69

Authors: V. Rashtchi, H. Bizhani, F. R. Tatari

Abstract:

This paper presents a new online loss minimization for an induction motor drive. Among the many loss minimization algorithms (LMAs) for an induction motor, a particle swarm optimization (PSO) has the advantages of fast response and high accuracy. However, the performance of the PSO and other optimization algorithms depend on the accuracy of the modeling of the motor drive and losses. In the development of the loss model, there is always a trade off between accuracy and complexity. This paper presents a new online optimization to determine an optimum flux level for the efficiency optimization of the vector-controlled induction motor drive. An induction motor (IM) model in d-q coordinates is referenced to the rotor magnetizing current. This transformation results in no leakage inductance on the rotor side, thus the decomposition into d-q components in the steady-state motor model can be utilized in deriving the motor loss model. The suggested algorithm is simple for implementation.

Keywords: induction machine, loss minimization, magnetizing current, particle swarm optimization

Procedia PDF Downloads 633

Authors: Thomas L. McCullough, John D. Hague, Marylesa M. Howard, Matthew K. Kiser, Michael A. Mazur, Lance K. McLean, Johanna L. Turk

Abstract:

Radiological search and mapping depends on the successful recognition of anomalies in large data sets which contain varied and dynamic backgrounds. We present a new algorithmic approach for real-time anomaly detection which is resistant to common detector imperfections, avoids the limitations of a source template library and provides immediate, and easily interpretable, user feedback. This algorithm is based on a continuous wavelet transform for variance reduction and evaluates the deviation between a foreground measurement and a local background expectation using methods from linear algebra. We also present a technique for recognizing and visualizing spectrally similar clusters of data. This technique uses Laplacian Eigenmap Manifold Learning to perform dimensional reduction which preserves the geometric "closeness" of the data while maintaining sensitivity to outlying data. We illustrate the utility of both techniques on real-world data sets.

Keywords: radiological search, radiological mapping, radioactivity, radiation protection

Procedia PDF Downloads 693

Authors: Ashish James, Anusha James

Abstract:

The impressive predictive performance of deep learning techniques on a wide range of tasks has led to its widespread use. Estimating the confidence of these predictions is paramount for improving the safety and reliability of such systems. However, the uncertainty estimates provided by neural networks (NNs) tend to be overconfident and unreasonable. Ensemble of NNs typically produce good predictions but uncertainty estimates tend to be inconsistent. Inspired by these, this paper presents a framework that can quantitatively estimate the uncertainties by leveraging the advances in transfer learning through slight modification to the existing training pipelines. This promising algorithm is developed with an intention of deployment in real world problems which already boast a good predictive performance by reusing those pretrained models. The idea is to capture the behavior of the trained NNs for the base task by augmenting it with the uncertainty estimates from a supplementary network. A series of experiments with known and unknown distributions show that the proposed approach produces well calibrated uncertainty estimates with high quality predictions.

Keywords: uncertainty estimation, neural networks, transfer learning, regression

Procedia PDF Downloads 135

Authors: Sonia Anand Knowlton

Abstract:

There are some cases where we as a society feel deeply uncomfortable with the use of Artificial Intelligence (AI) tools in the judicial decision-making process, and justifiably so. A perfect example is COMPAS, an algorithmic model that predicts recidivism rates of offenders to assist in the determination of their bail conditions. COMPAS turned out to be extremely racist: it massively overpredicted recidivism rates of Black offenders and underpredicted recidivism rates of white offenders. At the same time, there are certain uses of AI in the judicial decision-making process that many would feel more comfortable with and even support. Take, for example, a “super-breathalyzer,” an (albeit imaginary) tool that uses AI to deliver highly detailed information about the subject of the breathalyzer test to the legal decision-makers analyzing their drunk-driving case. This article evaluates the point at which a judge’s use of AI tools begins to undermine the public’s trust in the administration of justice. It argues that the answer to this question depends on whether the AI tool is in a role in which it must perform a moral evaluation of a human being.

Keywords: artificial intelligence, judicial reasoning, morality, technology, algorithm

Procedia PDF Downloads 81