Search results for: predictive analytics
946 Evaluating the Suitability and Performance of Dynamic Modulus Predictive Models for North Dakota’s Asphalt Mixtures
Authors: Duncan Oteki, Andebut Yeneneh, Daba Gedafa, Nabil Suleiman
Abstract:
Most agencies lack the equipment required to measure the dynamic modulus (|E*|) of asphalt mixtures, necessitating the need to use predictive models. This study compared measured |E*| values for nine North Dakota asphalt mixes using the original Witczak, modified Witczak, and Hirsch models. The influence of temperature on the |E*| models was investigated, and Pavement ME simulations were conducted using measured |E*| and predictions from the most accurate |E*| model. The results revealed that the original Witczak model yielded the lowest Se/Sy and highest R² values, indicating the lowest bias and highest accuracy, while the poorest overall performance was exhibited by the Hirsch model. Using predicted |E*| as inputs in the Pavement ME generated conservative distress predictions compared to using measured |E*|. The original Witczak model was recommended for predicting |E*| for low-reliability pavements in North Dakota.Keywords: asphalt mixture, binder, dynamic modulus, MEPDG, pavement ME, performance, prediction
Procedia PDF Downloads 16945 Effectiveness of the Lacey Assessment of Preterm Infants to Predict Neuromotor Outcomes of Premature Babies at 12 Months Corrected Age
Authors: Thanooja Naushad, Meena Natarajan, Tushar Vasant Kulkarni
Abstract:
Background: The Lacey Assessment of Preterm Infants (LAPI) is used in clinical practice to identify premature babies at risk of neuromotor impairments, especially cerebral palsy. This study attempted to find the validity of the Lacey assessment of preterm infants to predict neuromotor outcomes of premature babies at 12 months corrected age and to compare its predictive ability with the brain ultrasound. Methods: This prospective cohort study included 89 preterm infants (45 females and 44 males) born below 35 weeks gestation who were admitted to the neonatal intensive care unit of a government hospital in Dubai. Initial assessment was done using the Lacey assessment after the babies reached 33 weeks postmenstrual age. Follow up assessment on neuromotor outcomes was done at 12 months (± 1 week) corrected age using two standardized outcome measures, i.e., infant neurological international battery and Alberta infant motor scale. Brain ultrasound data were collected retrospectively. Data were statistically analyzed, and the diagnostic accuracy of the Lacey assessment of preterm infants (LAPI) was calculated -when used alone and in combination with the brain ultrasound. Results: On comparison with brain ultrasound, the Lacey assessment showed superior specificity (96% vs. 77%), higher positive predictive value (57% vs. 22%), and higher positive likelihood ratio (18 vs. 3) to predict neuromotor outcomes at one year of age. The sensitivity of Lacey assessment was lower than brain ultrasound (66% vs. 83%), whereas specificity was similar (97% vs. 98%). A combination of Lacey assessment and brain ultrasound results showed higher sensitivity (80%), positive (66%), and negative (98%) predictive values, positive likelihood ratio (24), and test accuracy (95%) than Lacey assessment alone in predicting neurological outcomes. The negative predictive value of the Lacey assessment was similar to that of its combination with brain ultrasound (96%). Conclusion: Results of this study suggest that the Lacey assessment of preterm infants can be used as a supplementary assessment tool for premature babies in the neonatal intensive care unit. Due to its high specificity, Lacey assessment can be used to identify those babies at low risk of abnormal neuromotor outcomes at a later age. When used along with the findings of the brain ultrasound, Lacey assessment has better sensitivity to identify preterm babies at particular risk. These findings have applications in identifying premature babies who may benefit from early intervention services.Keywords: brain ultrasound, lacey assessment of preterm infants, neuromotor outcomes, preterm
Procedia PDF Downloads 104944 Psychological Testing in Industrial/Organizational Psychology: Validity and Reliability of Psychological Assessments in the Workplace
Authors: Melissa C. Monney
Abstract:
Psychological testing has been of interest to researchers for many years as useful tools in assessing and diagnosing various disorders as well as to assist in understanding human behavior. However, for over 20 years now, researchers and laypersons alike have been interested in using them for other purposes, such as determining factors in employee selection, promotion, and even termination. In recent years, psychological assessments have been useful in facilitating workplace decision processing, regarding employee circulation within organizations. This literature review explores four of the most commonly used psychological tests in workplace environments, namely cognitive ability, emotional intelligence, integrity, and personality tests, as organizations have used these tests to assess different factors of human behavior as predictive measures of future employee behaviors. The findings suggest that while there is much controversy and debate regarding the validity and reliability of these tests in workplace settings as they were not originally designed for these purposes, the use of such assessments in the workplace has been useful in decreasing costs and employee turnover as well as increase job satisfaction by ensuring the right employees are selected for their roles.Keywords: cognitive ability, personality testing, predictive validity, workplace behavior
Procedia PDF Downloads 212943 The Potential Impact of Big Data Analytics on Pharmaceutical Supply Chain Management
Authors: Maryam Ziaee, Himanshu Shee, Amrik Sohal
Abstract:
Big Data Analytics (BDA) in supply chain management has recently drawn the attention of academics and practitioners. Big data refers to a massive amount of data from different sources, in different formats, generated at high speed through transactions in business environments and supply chain networks. Traditional statistical tools and techniques find it difficult to analyse this massive data. BDA can assist organisations to capture, store, and analyse data specifically in the field of supply chain. Currently, there is a paucity of research on BDA in the pharmaceutical supply chain context. In this research, the Australian pharmaceutical supply chain was selected as the case study. This industry is highly significant since the right medicine must reach the right patients, at the right time, in right quantity, in good condition, and at the right price to save lives. However, drug shortages remain a substantial problem for hospitals across Australia with implications on patient care, staff resourcing, and expenditure. Furthermore, a massive volume and variety of data is generated at fast speed from multiple sources in pharmaceutical supply chain, which needs to be captured and analysed to benefit operational decisions at every stage of supply chain processes. As the pharmaceutical industry lags behind other industries in using BDA, it raises the question of whether the use of BDA can improve transparency among pharmaceutical supply chain by enabling the partners to make informed-decisions across their operational activities. This presentation explores the impacts of BDA on supply chain management. An exploratory qualitative approach was adopted to analyse data collected through interviews. This study also explores the BDA potential in the whole pharmaceutical supply chain rather than focusing on a single entity. Twenty semi-structured interviews were undertaken with top managers in fifteen organisations (five pharmaceutical manufacturers, five wholesalers/distributors, and five public hospital pharmacies) to investigate their views on the use of BDA. The findings revealed that BDA can enable pharmaceutical entities to have improved visibility over the whole supply chain and also the market; it enables entities, especially manufacturers, to monitor consumption and the demand rate in real-time and make accurate demand forecasts which reduce drug shortages. Timely and precise decision-making can allow the entities to source and manage their stocks more effectively. This can likely address the drug demand at hospitals and respond to unanticipated issues such as drug shortages. Earlier studies explore BDA in the context of clinical healthcare; however, this presentation investigates the benefits of BDA in the Australian pharmaceutical supply chain. Furthermore, this research enhances managers’ insight into the potentials of BDA at every stage of supply chain processes and helps to improve decision-making in their supply chain operations. The findings will turn the rhetoric of data-driven decision into a reality where the managers may opt for analytics for improved decision-making in the supply chain processes.Keywords: big data analytics, data-driven decision, pharmaceutical industry, supply chain management
Procedia PDF Downloads 76942 Identification and Force Control of a Two Chambers Pneumatic Soft Actuator
Authors: Najib K. Dankadai, Ahmad 'Athif Mohd Faudzi, Khairuddin Osman, Muhammad Rusydi Muhammad Razif, IIi Najaa Aimi Mohd Nordin
Abstract:
Researches in soft actuators are now growing rapidly because of their adequacy to be applied in sectors like medical, agriculture, biological and welfare. This paper presents system identification (SI) and control of the force generated by a two chambers pneumatic soft actuator (PSA). A force mathematical model for the actuator was identified experimentally using data acquisition card and MATLAB SI toolbox. Two control techniques; a predictive functional control (PFC) and conventional proportional integral and derivative (PID) schemes are proposed and compared based on the identified model for the soft actuator flexible mechanism. Results of this study showed that both of the proposed controllers ensure accurate tracking when the closed loop system was tested with the step, sinusoidal and multi step reference input through MATLAB simulation although the PFC provides a better response than the PID.Keywords: predictive functional control (PFC), proportional integral and derivative (PID), soft actuator, system identification
Procedia PDF Downloads 285941 Improved Predictive Models for the IRMA Network Using Nonlinear Optimisation
Authors: Vishwesh Kulkarni, Nikhil Bellarykar
Abstract:
Cellular complexity stems from the interactions among thousands of different molecular species. Thanks to the emerging fields of systems and synthetic biology, scientists are beginning to unravel these regulatory, signaling, and metabolic interactions and to understand their coordinated action. Reverse engineering of biological networks has has several benefits but a poor quality of data combined with the difficulty in reproducing it limits the applicability of these methods. A few years back, many of the commonly used predictive algorithms were tested on a network constructed in the yeast Saccharomyces cerevisiae (S. cerevisiae) to resolve this issue. The network was a synthetic network of five genes regulating each other for the so-called in vivo reverse-engineering and modeling assessment (IRMA). The network was constructed in S. cereviase since it is a simple and well characterized organism. The synthetic network included a variety of regulatory interactions, thus capturing the behaviour of larger eukaryotic gene networks on a smaller scale. We derive a new set of algorithms by solving a nonlinear optimization problem and show how these algorithms outperform other algorithms on these datasets.Keywords: synthetic gene network, network identification, optimization, nonlinear modeling
Procedia PDF Downloads 121940 Comparison of Two Neural Networks To Model Margarine Age And Predict Shelf-Life Using Matlab
Authors: Phakamani Xaba, Robert Huberts, Bilainu Oboirien
Abstract:
The present study was aimed at developing & comparing two neural-network-based predictive models to predict shelf-life/product age of South African margarine using free fatty acid (FFA), water droplet size (D3.3), water droplet distribution (e-sigma), moisture content, peroxide value (PV), anisidine valve (AnV) and total oxidation (totox) value as input variables to the model. Brick margarine products which had varying ages ranging from fresh i.e. week 0 to week 47 were sourced. The brick margarine products which had been stored at 10 & 25 °C and were characterized. JMP and MATLAB models to predict shelf-life/ margarine age were developed and their performances were compared. The key performance indicators to evaluate the model performances were correlation coefficient (CC), root mean square error (RMSE), and mean absolute percentage error (MAPE) relative to the actual data. The MATLAB-developed model showed a better performance in all three performance indicators. The correlation coefficient of the MATLAB model was 99.86% versus 99.74% for the JMP model, the RMSE was 0.720 compared to 1.005 and the MAPE was 7.4% compared to 8.571%. The MATLAB model was selected to be the most accurate, and then, the number of hidden neurons/ nodes was optimized to develop a single predictive model. The optimized MATLAB with 10 neurons showed a better performance compared to the models with 1 & 5 hidden neurons. The developed models can be used by margarine manufacturers, food research institutions, researchers etc, to predict shelf-life/ margarine product age, optimize addition of antioxidants, extend shelf-life of products and proactively troubleshoot for problems related to changes which have an impact on shelf-life of margarine without conducting expensive trials.Keywords: margarine shelf-life, predictive modelling, neural networks, oil oxidation
Procedia PDF Downloads 164939 Dual-Channel Reliable Breast Ultrasound Image Classification Based on Explainable Attribution and Uncertainty Quantification
Authors: Haonan Hu, Shuge Lei, Dasheng Sun, Huabin Zhang, Kehong Yuan, Jian Dai, Jijun Tang
Abstract:
This paper focuses on the classification task of breast ultrasound images and conducts research on the reliability measurement of classification results. A dual-channel evaluation framework was developed based on the proposed inference reliability and predictive reliability scores. For the inference reliability evaluation, human-aligned and doctor-agreed inference rationals based on the improved feature attribution algorithm SP-RISA are gracefully applied. Uncertainty quantification is used to evaluate the predictive reliability via the test time enhancement. The effectiveness of this reliability evaluation framework has been verified on the breast ultrasound clinical dataset YBUS, and its robustness is verified on the public dataset BUSI. The expected calibration errors on both datasets are significantly lower than traditional evaluation methods, which proves the effectiveness of the proposed reliability measurement.Keywords: medical imaging, ultrasound imaging, XAI, uncertainty measurement, trustworthy AI
Procedia PDF Downloads 52938 The Predictive Value of Serum Bilirubin in the Post-Transplant De Novo Malignancy: A Data Mining Approach
Authors: Nasim Nosoudi, Amir Zadeh, Hunter White, Joshua Conrad, Joon W. Shim
Abstract:
De novo Malignancy has become one of the major causes of death after transplantation, so early cancer diagnosis and detection can drastically improve survival rates post-transplantation. Most previous work focuses on using artificial intelligence (AI) to predict transplant success or failure outcomes. In this work, we focused on predicting de novo malignancy after liver transplantation using AI. We chose the patients that had malignancy after liver transplantation with no history of malignancy pre-transplant. Their donors were cancer-free as well. We analyzed 254,200 patient profiles with post-transplant malignancy from the US Organ Procurement and Transplantation Network (OPTN). Several popular data mining methods were applied to the resultant dataset to build predictive models to characterize de novo malignancy after liver transplantation. Recipient's bilirubin, creatinine, weight, gender, number of days recipient was on the transplant waiting list, Epstein Barr Virus (EBV), International normalized ratio (INR), and ascites are among the most important factors affecting de novo malignancy after liver transplantationKeywords: De novo malignancy, bilirubin, data mining, transplantation
Procedia PDF Downloads 73937 The Investigation of Predictor Affect of Childhood Trauma, Dissociation, Alexithymia, and Gender on Dissociation in University Students
Authors: Gizem Akcan, Erdinc Ozturk
Abstract:
The purpose of the study was to determine some psychosocial variables that predict dissociation in university students. These psychosocial variables were perceived childhood trauma, alexithymia, and gender. 150 (75 males, 75 females) university students (bachelor, master and postgraduate) were enrolled in this study. They were chosen from universities in Istanbul at the education year of 2016-2017. Dissociative Experiences Scale (DES), Childhood Trauma Questionnaire (CTQ) and Toronto Alexithymia Scale were used to assess related variables. Demographic Information Form was given to students in order to have their demographic information. Frequency Distribution, Linear Regression Analysis, and t-test analysis were used for statistical analysis. Childhood trauma and alexithymia were found to have predictive value on dissociation among university students. However, physical abuse, physical neglect and emotional neglect sub dimensions of childhood trauma and externally-oriented thinking sub dimension of alexithymia did not have predictive value on dissociation. Moreover, there was no significant difference between males and females in terms of dissociation scores of participants.Keywords: childhood trauma, dissociation, alexithymia, gender
Procedia PDF Downloads 363936 Performance and Limitations of Likelihood Based Information Criteria and Leave-One-Out Cross-Validation Approximation Methods
Authors: M. A. C. S. Sampath Fernando, James M. Curran, Renate Meyer
Abstract:
Model assessment, in the Bayesian context, involves evaluation of the goodness-of-fit and the comparison of several alternative candidate models for predictive accuracy and improvements. In posterior predictive checks, the data simulated under the fitted model is compared with the actual data. Predictive model accuracy is estimated using information criteria such as the Akaike information criterion (AIC), the Bayesian information criterion (BIC), the Deviance information criterion (DIC), and the Watanabe-Akaike information criterion (WAIC). The goal of an information criterion is to obtain an unbiased measure of out-of-sample prediction error. Since posterior checks use the data twice; once for model estimation and once for testing, a bias correction which penalises the model complexity is incorporated in these criteria. Cross-validation (CV) is another method used for examining out-of-sample prediction accuracy. Leave-one-out cross-validation (LOO-CV) is the most computationally expensive variant among the other CV methods, as it fits as many models as the number of observations. Importance sampling (IS), truncated importance sampling (TIS) and Pareto-smoothed importance sampling (PSIS) are generally used as approximations to the exact LOO-CV and utilise the existing MCMC results avoiding expensive computational issues. The reciprocals of the predictive densities calculated over posterior draws for each observation are treated as the raw importance weights. These are in turn used to calculate the approximate LOO-CV of the observation as a weighted average of posterior densities. In IS-LOO, the raw weights are directly used. In contrast, the larger weights are replaced by their modified truncated weights in calculating TIS-LOO and PSIS-LOO. Although, information criteria and LOO-CV are unable to reflect the goodness-of-fit in absolute sense, the differences can be used to measure the relative performance of the models of interest. However, the use of these measures is only valid under specific circumstances. This study has developed 11 models using normal, log-normal, gamma, and student’s t distributions to improve the PCR stutter prediction with forensic data. These models are comprised of four with profile-wide variances, four with locus specific variances, and three which are two-component mixture models. The mean stutter ratio in each model is modeled as a locus specific simple linear regression against a feature of the alleles under study known as the longest uninterrupted sequence (LUS). The use of AIC, BIC, DIC, and WAIC in model comparison has some practical limitations. Even though, IS-LOO, TIS-LOO, and PSIS-LOO are considered to be approximations of the exact LOO-CV, the study observed some drastic deviations in the results. However, there are some interesting relationships among the logarithms of pointwise predictive densities (lppd) calculated under WAIC and the LOO approximation methods. The estimated overall lppd is a relative measure that reflects the overall goodness-of-fit of the model. Parallel log-likelihood profiles for the models conditional on equal posterior variances in lppds were observed. This study illustrates the limitations of the information criteria in practical model comparison problems. In addition, the relationships among LOO-CV approximation methods and WAIC with their limitations are discussed. Finally, useful recommendations that may help in practical model comparisons with these methods are provided.Keywords: cross-validation, importance sampling, information criteria, predictive accuracy
Procedia PDF Downloads 360935 Design and Development of Real-Time Optimal Energy Management System for Hybrid Electric Vehicles
Authors: Masood Roohi, Amir Taghavipour
Abstract:
This paper describes a strategy to develop an energy management system (EMS) for a charge-sustaining power-split hybrid electric vehicle. This kind of hybrid electric vehicles (HEVs) benefit from the advantages of both parallel and series architecture. However, it gets relatively more complicated to manage power flow between the battery and the engine optimally. The applied strategy in this paper is based on nonlinear model predictive control approach. First of all, an appropriate control-oriented model which was accurate enough and simple was derived. Towards utilization of this controller in real-time, the problem was solved off-line for a vast area of reference signals and initial conditions and stored the computed manipulated variables inside look-up tables. Look-up tables take a little amount of memory. Also, the computational load dramatically decreased, because to find required manipulated variables the controller just needed a simple interpolation between tables.Keywords: hybrid electric vehicles, energy management system, nonlinear model predictive control, real-time
Procedia PDF Downloads 315934 Twitter Ego Networks and the Capital Markets: A Social Network Analysis Perspective of Market Reactions to Earnings Announcement Events
Authors: Gregory D. Saxton
Abstract:
Networks are everywhere: lunch ties among co-workers, golfing partnerships among employees, interlocking board-of-director connections, Facebook friendship ties, etc. Each network varies in terms of its structure -its size, how inter-connected network members are, and the prevalence of sub-groups and cliques. At the same time, within any given network, some network members will have a more important, more central position on account of their greater number of connections or their capacity as “bridges” connecting members of different network cliques. The logic of network structure and position is at the heart of what is known as social network analysis, and this paper applies this logic to the study of the stock market. Using an array of data analytics and machine learning tools, this study will examine 17 million Twitter messages discussing the stocks of the firms in the S&P 1,500 index in 2018. Each of these 1,500 stocks has a distinct Twitter discussion network that varies in terms of core network characteristics such as size, density, influence, norms and values, level of activity, and embedded resources. The study’s core proposition is that the ultimate effect of any market-relevant information is contingent on the characteristics of the network through which it flows. To test this proposition, this study operationalizes each of the core network characteristics and examines their influence on market reactions to 2018 quarterly earnings announcement events.Keywords: data analytics, investor-to-investor communication, social network analysis, Twitter
Procedia PDF Downloads 84933 Development of a Predictive Model to Prevent Financial Crisis
Authors: Tengqin Han
Abstract:
Delinquency has been a crucial factor in economics throughout the years. Commonly seen in credit card and mortgage, it played one of the crucial roles in causing the most recent financial crisis in 2008. In each case, a delinquency is a sign of the loaner being unable to pay off the debt, and thus may cause a lost of property in the end. Individually, one case of delinquency seems unimportant compared to the entire credit system. China, as an emerging economic entity, the national strength and economic strength has grown rapidly, and the gross domestic product (GDP) growth rate has remained as high as 8% in the past decades. However, potential risks exist behind the appearance of prosperity. Among the risks, the credit system is the most significant one. Due to long term and a large amount of balance of the mortgage, it is critical to monitor the risk during the performance period. In this project, about 300,000 mortgage account data are analyzed in order to develop a predictive model to predict the probability of delinquency. Through univariate analysis, the data is cleaned up, and through bivariate analysis, the variables with strong predictive power are detected. The project is divided into two parts. In the first part, the analysis data of 2005 are split into 2 parts, 60% for model development, and 40% for in-time model validation. The KS of model development is 31, and the KS for in-time validation is 31, indicating the model is stable. In addition, the model is further validation by out-of-time validation, which uses 40% of 2006 data, and KS is 33. This indicates the model is still stable and robust. In the second part, the model is improved by the addition of macroeconomic economic indexes, including GDP, consumer price index, unemployment rate, inflation rate, etc. The data of 2005 to 2010 is used for model development and validation. Compared with the base model (without microeconomic variables), KS is increased from 41 to 44, indicating that the macroeconomic variables can be used to improve the separation power of the model, and make the prediction more accurate.Keywords: delinquency, mortgage, model development, model validation
Procedia PDF Downloads 196932 Online Learning for Modern Business Models: Theoretical Considerations and Algorithms
Authors: Marian Sorin Ionescu, Olivia Negoita, Cosmin Dobrin
Abstract:
This scientific communication reports and discusses learning models adaptable to modern business problems and models specific to digital concepts and paradigms. In the PAC (probably approximately correct) learning model approach, in which the learning process begins by receiving a batch of learning examples, the set of learning processes is used to acquire a hypothesis, and when the learning process is fully used, this hypothesis is used in the prediction of new operational examples. For complex business models, a lot of models should be introduced and evaluated to estimate the induced results so that the totality of the results are used to develop a predictive rule, which anticipates the choice of new models. In opposition, for online learning-type processes, there is no separation between the learning (training) and predictive phase. Every time a business model is approached, a test example is considered from the beginning until the prediction of the appearance of a model considered correct from the point of view of the business decision. After choosing choice a part of the business model, the label with the logical value "true" is known. Some of the business models are used as examples of learning (training), which helps to improve the prediction mechanisms for future business models.Keywords: machine learning, business models, convex analysis, online learning
Procedia PDF Downloads 114931 Green Wave Control Strategy for Optimal Energy Consumption by Model Predictive Control in Electric Vehicles
Authors: Furkan Ozkan, M. Selcuk Arslan, Hatice Mercan
Abstract:
Electric vehicles are becoming increasingly popular asa sustainable alternative to traditional combustion engine vehicles. However, to fully realize the potential of EVs in reducing environmental impact and energy consumption, efficient control strategies are essential. This study explores the application of green wave control using model predictive control for electric vehicles, coupled with energy consumption modeling using neural networks. The use of MPC allows for real-time optimization of the vehicles’ energy consumption while considering dynamic traffic conditions. By leveraging neural networks for energy consumption modeling, the EV's performance can be further enhanced through accurate predictions and adaptive control. The integration of these advanced control and modeling techniques aims to maximize energy efficiency and range while navigating urban traffic scenarios. The findings of this research offer valuable insights into the potential of green wave control for electric vehicles and demonstrate the significance of integrating MPC and neural network modeling for optimizing energy consumption. This work contributes to the advancement of sustainable transportation systems and the widespread adoption of electric vehicles. To evaluate the effectiveness of the green wave control strategy in real-world urban environments, extensive simulations were conducted using a high-fidelity vehicle model and realistic traffic scenarios. The results indicate that the integration of model predictive control and energy consumption modeling with neural networks had a significant impact on the energy efficiency and range of electric vehicles. Through the use of MPC, the electric vehicle was able to adapt its speed and acceleration profile in realtime to optimize energy consumption while maintaining travel time objectives. The neural network-based energy consumption modeling provided accurate predictions, enabling the vehicle to anticipate and respond to variations in traffic flow, further enhancing energy efficiency and range. Furthermore, the study revealed that the green wave control strategy not only reduced energy consumption but also improved the overall driving experience by minimizing abrupt acceleration and deceleration, leading to a smoother and more comfortable ride for passengers. These results demonstrate the potential for green wave control to revolutionize urban transportation by enhancing the performance of electric vehicles and contributing to a more sustainable and efficient mobility ecosystem.Keywords: electric vehicles, energy efficiency, green wave control, model predictive control, neural networks
Procedia PDF Downloads 13930 Comparative Analysis of Predictive Models for Customer Churn Prediction in the Telecommunication Industry
Authors: Deepika Christopher, Garima Anand
Abstract:
To determine the best model for churn prediction in the telecom industry, this paper compares 11 machine learning algorithms, namely Logistic Regression, Support Vector Machine, Random Forest, Decision Tree, XGBoost, LightGBM, Cat Boost, AdaBoost, Extra Trees, Deep Neural Network, and Hybrid Model (MLPClassifier). It also aims to pinpoint the top three factors that lead to customer churn and conducts customer segmentation to identify vulnerable groups. According to the data, the Logistic Regression model performs the best, with an F1 score of 0.6215, 81.76% accuracy, 68.95% precision, and 56.57% recall. The top three attributes that cause churn are found to be tenure, Internet Service Fiber optic, and Internet Service DSL; conversely, the top three models in this article that perform the best are Logistic Regression, Deep Neural Network, and AdaBoost. The K means algorithm is applied to establish and analyze four different customer clusters. This study has effectively identified customers that are at risk of churn and may be utilized to develop and execute strategies that lower customer attrition.Keywords: attrition, retention, predictive modeling, customer segmentation, telecommunications
Procedia PDF Downloads 19929 A Real-Time Bayesian Decision-Support System for Predicting Suspect Vehicle’s Intended Target Using a Sparse Camera Network
Authors: Payam Mousavi, Andrew L. Stewart, Huiwen You, Aryeh F. G. Fayerman
Abstract:
We present a decision-support tool to assist an operator in the detection and tracking of a suspect vehicle traveling to an unknown target destination. Multiple data sources, such as traffic cameras, traffic information, weather, etc., are integrated and processed in real-time to infer a suspect’s intended destination chosen from a list of pre-determined high-value targets. Previously, we presented our work in the detection and tracking of vehicles using traffic and airborne cameras. Here, we focus on the fusion and processing of that information to predict a suspect’s behavior. The network of cameras is represented by a directional graph, where the edges correspond to direct road connections between the nodes and the edge weights are proportional to the average time it takes to travel from one node to another. For our experiments, we construct our graph based on the greater Los Angeles subset of the Caltrans’s “Performance Measurement System” (PeMS) dataset. We propose a Bayesian approach where a posterior probability for each target is continuously updated based on detections of the suspect in the live video feeds. Additionally, we introduce the concept of ‘soft interventions’, inspired by the field of Causal Inference. Soft interventions are herein defined as interventions that do not immediately interfere with the suspect’s movements; rather, a soft intervention may induce the suspect into making a new decision, ultimately making their intent more transparent. For example, a soft intervention could be temporarily closing a road a few blocks from the suspect’s current location, which may require the suspect to change their current course. The objective of these interventions is to gain the maximum amount of information about the suspect’s intent in the shortest possible time. Our system currently operates in a human-on-the-loop mode where at each step, a set of recommendations are presented to the operator to aid in decision-making. In principle, the system could operate autonomously, only prompting the operator for critical decisions, allowing the system to significantly scale up to larger areas and multiple suspects. Once the intended target is identified with sufficient confidence, the vehicle is reported to the authorities to take further action. Other recommendations include a selection of road closures, i.e., soft interventions, or to continue monitoring. We evaluate the performance of the proposed system using simulated scenarios where the suspect, starting at random locations, takes a noisy shortest path to their intended target. In all scenarios, the suspect’s intended target is unknown to our system. The decision thresholds are selected to maximize the chances of determining the suspect’s intended target in the minimum amount of time and with the smallest number of interventions. We conclude by discussing the limitations of our current approach to motivate a machine learning approach, based on reinforcement learning in order to relax some of the current limiting assumptions.Keywords: autonomous surveillance, Bayesian reasoning, decision support, interventions, patterns of life, predictive analytics, predictive insights
Procedia PDF Downloads 90928 Detection of Autistic Children's Voice Based on Artificial Neural Network
Authors: Royan Dawud Aldian, Endah Purwanti, Soegianto Soelistiono
Abstract:
In this research we have been developed an automatic investigation to classify normal children voice or autistic by using modern computation technology that is computation based on artificial neural network. The superiority of this computation technology is its capability on processing and saving data. In this research, digital voice features are gotten from the coefficient of linear-predictive coding with auto-correlation method and have been transformed in frequency domain using fast fourier transform, which used as input of artificial neural network in back-propagation method so that will make the difference between normal children and autistic automatically. The result of back-propagation method shows that successful classification capability for normal children voice experiment data is 100% whereas, for autistic children voice experiment data is 100%. The success rate using back-propagation classification system for the entire test data is 100%.Keywords: autism, artificial neural network, backpropagation, linier predictive coding, fast fourier transform
Procedia PDF Downloads 420927 DeepLig: A de-novo Computational Drug Design Approach to Generate Multi-Targeted Drugs
Authors: Anika Chebrolu
Abstract:
Mono-targeted drugs can be of limited efficacy against complex diseases. Recently, multi-target drug design has been approached as a promising tool to fight against these challenging diseases. However, the scope of current computational approaches for multi-target drug design is limited. DeepLig presents a de-novo drug discovery platform that uses reinforcement learning to generate and optimize novel, potent, and multitargeted drug candidates against protein targets. DeepLig’s model consists of two networks in interplay: a generative network and a predictive network. The generative network, a Stack- Augmented Recurrent Neural Network, utilizes a stack memory unit to remember and recognize molecular patterns when generating novel ligands from scratch. The generative network passes each newly created ligand to the predictive network, which then uses multiple Graph Attention Networks simultaneously to forecast the average binding affinity of the generated ligand towards multiple target proteins. With each iteration, given feedback from the predictive network, the generative network learns to optimize itself to create molecules with a higher average binding affinity towards multiple proteins. DeepLig was evaluated based on its ability to generate multi-target ligands against two distinct proteins, multi-target ligands against three distinct proteins, and multi-target ligands against two distinct binding pockets on the same protein. With each test case, DeepLig was able to create a library of valid, synthetically accessible, and novel molecules with optimal and equipotent binding energies. We propose that DeepLig provides an effective approach to design multi-targeted drug therapies that can potentially show higher success rates during in-vitro trials.Keywords: drug design, multitargeticity, de-novo, reinforcement learning
Procedia PDF Downloads 49926 Comparison of Techniques for Detection and Diagnosis of Eccentricity in the Air-Gap Fault in Induction Motors
Authors: Abrahão S. Fontes, Carlos A. V. Cardoso, Levi P. B. Oliveira
Abstract:
The induction motors are used worldwide in various industries. Several maintenance techniques are applied to increase the operating time and the lifespan of these motors. Among these, the predictive maintenance techniques such as Motor Current Signature Analysis (MCSA), Motor Square Current Signature Analysis (MSCSA), Park's Vector Approach (PVA) and Park's Vector Square Modulus (PVSM) are used to detect and diagnose faults in electric motors, characterized by patterns in the stator current frequency spectrum. In this article, these techniques are applied and compared on a real motor, which has the fault of eccentricity in the air-gap. It was used as a theoretical model of an electric induction motor without fault in order to assist comparison between the stator current frequency spectrum patterns with and without faults. Metrics were purposed and applied to evaluate the sensitivity of each technique fault detection. The results presented here show that the above techniques are suitable for the fault of eccentricity in the air gap, whose comparison between these showed the suitability of each one.Keywords: eccentricity in the air-gap, fault diagnosis, induction motors, predictive maintenance
Procedia PDF Downloads 320925 Optimal Tamping for Railway Tracks, Reducing Railway Maintenance Expenditures by the Use of Integer Programming
Authors: Rui Li, Min Wen, Kim Bang Salling
Abstract:
For the modern railways, maintenance is critical for ensuring safety, train punctuality and overall capacity utilization. The cost of railway maintenance in Europe is high, on average between 30,000 – 100,000 Euros per kilometer per year. In order to reduce such maintenance expenditures, this paper presents a mixed 0-1 linear mathematical model designed to optimize the predictive railway tamping activities for ballast track in the planning horizon of three to four years. The objective function is to minimize the tamping machine actual costs. The approach of the research is using the simple dynamic model for modelling condition-based tamping process and the solution method for finding optimal condition-based tamping schedule. Seven technical and practical aspects are taken into account to schedule tamping: (1) track degradation of the standard deviation of the longitudinal level over time; (2) track geometrical alignment; (3) track quality thresholds based on the train speed limits; (4) the dependency of the track quality recovery on the track quality after tamping operation; (5) Tamping machine operation practices (6) tamping budgets and (7) differentiating the open track from the station sections. A Danish railway track between Odense and Fredericia with 42.6 km of length is applied for a time period of three and four years in the proposed maintenance model. The generated tamping schedule is reasonable and robust. Based on the result from the Danish railway corridor, the total costs can be reduced significantly (50%) than the previous model which is based on optimizing the number of tamping. The different maintenance strategies have been discussed in the paper. The analysis from the results obtained from the model also shows a longer period of predictive tamping planning has more optimal scheduling of maintenance actions than continuous short term preventive maintenance, namely yearly condition-based planning.Keywords: integer programming, railway tamping, predictive maintenance model, preventive condition-based maintenance
Procedia PDF Downloads 408924 Establishing a Surrogate Approach to Assess the Exposure Concentrations during Coating Process
Authors: Shan-Hong Ying, Ying-Fang Wang
Abstract:
A surrogate approach was deployed for assessing exposures of multiple chemicals at the selected working area of coating processes and applied to assess the exposure concentration of similar exposed groups using the same chemicals but different formula ratios. For the selected area, 6 to 12 portable photoionization detector (PID) were placed uniformly in its workplace to measure its total VOCs concentrations (CT-VOCs) for 6 randomly selected workshifts. Simultaneously, one sampling strain was placed beside one of these portable PIDs, and the collected air sample was analyzed for individual concentration (CVOCi) of 5 VOCs (xylene, butanone, toluene, butyl acetate, and dimethylformamide). Predictive models were established by relating the CT-VOCs to CVOCi of each individual compound via simple regression analysis. The established predictive models were employed to predict each CVOCi based on the measured CT-VOC for each the similar working area using the same portable PID. Results show that predictive models obtained from simple linear regression analyses were found with an R2 = 0.83~0.99 indicating that CT-VOCs were adequate for predicting CVOCi. In order to verify the validity of the exposure prediction model, the sampling analysis of the above chemical substances was further carried out and the correlation between the measured value (Cm) and the predicted value (Cp) was analyzed. It was found that there is a good correction between the predicted value and measured value of each measured chemical substance (R2=0.83~0.98). Therefore, the surrogate approach could be assessed the exposure concentration of similar exposed groups using the same chemicals but different formula ratios. However, it is recommended to establish the prediction model between the chemical substances belonging to each coater and the direct-reading PID, which is more representative of reality exposure situation and more accurately to estimate the long-term exposure concentration of operators.Keywords: exposure assessment, exposure prediction model, surrogate approach, TVOC
Procedia PDF Downloads 115923 Assessment of Predictive Confounders for the Prevalence of Breast Cancer among Iraqi Population: A Retrospective Study from Baghdad, Iraq
Authors: Nadia H. Mohammed, Anmar Al-Taie, Fadia H. Al-Sultany
Abstract:
Although breast cancer prevalence continues to increase, mortality has been decreasing as a result of early detection and improvement in adjuvant systemic therapy. Nevertheless, this disease required further efforts to understand and identify the associated potential risk factors that could play a role in the prevalence of this malignancy among Iraqi women. The objective of this study was to assess the perception of certain predictive risk factors on the prevalence of breast cancer types among a sample of Iraqi women diagnosed with breast cancer. This was a retrospective observational study carried out at National Cancer Research Center in College of Medicine, Baghdad University from November 2017 to January 2018. Data of 100 patients with breast cancer whose biopsies examined in the National Cancer Research Center were included in this study. Data were collected to structure a detailed assessment regarding the patients’ demographic, medical and cancer records. The majority of study participants (94%) suffered from ductal breast cancer with mean age 49.57 years. Among those women, 48.9% were obese with body mass index (BMI) 35 kg/m2. 68.1% of them had positive family history of breast cancer and 66% had low parity. 40.4% had stage II ductal breast cancer followed by 25.5% with stage III. It was found that 59.6% and 68.1% had positive oestrogen receptor sensitivity and positive human epidermal growth factor (HER2/neu) receptor sensitivity respectively. In regard to the impact of prediction of certain variables on the incidence of ductal breast cancer, positive family history of breast cancer (P < 0.0001), low parity (P< 0.0001), stage I and II breast cancer (P = 0.02) and positive HER2/neu status (P < 0.0001) were significant predictive factors among the study participants. The results from this study provide relevant evidence for a significant positive and potential association between certain risk factors and the prevalence of breast cancer among Iraqi women.Keywords: Ductal Breast Cancer, Hormone Sensitivity, Iraq, Risk Factors
Procedia PDF Downloads 104922 Application of Global Predictive Real Time Control Strategy to Improve Flooding Prevention Performance of Urban Stormwater Basins
Authors: Shadab Shishegar, Sophie Duchesne, Genevieve Pelletier
Abstract:
Sustainability as one of the key elements of Smart cities, can be realized by employing Real Time Control Strategies for city’s infrastructures. Nowadays Stormwater management systems play an important role in mitigating the impacts of urbanization on natural hydrological cycle. These systems can be managed in such a way that they meet the smart cities standards. In fact, there is a huge potential for sustainable management of urban stormwater and also its adaptability to global challenges like climate change. Hence, a dynamically managed system that can adapt itself to instability of the environmental conditions is desirable. A Global Predictive Real Time Control approach is proposed in this paper to optimize the performance of stormwater management basins in terms of flooding prevention. To do so, a mathematical optimization model is developed then solved using Genetic Algorithm (GA). Results show an improved performance at system-level for the stormwater basins in comparison to static strategy.Keywords: environmental sustainability, optimization, real time control, storm water management
Procedia PDF Downloads 151921 Reinforcement Learning for Quality-Oriented Production Process Parameter Optimization Based on Predictive Models
Authors: Akshay Paranjape, Nils Plettenberg, Robert Schmitt
Abstract:
Producing faulty products can be costly for manufacturing companies and wastes resources. To reduce scrap rates in manufacturing, process parameters can be optimized using machine learning. Thus far, research mainly focused on optimizing specific processes using traditional algorithms. To develop a framework that enables real-time optimization based on a predictive model for an arbitrary production process, this study explores the application of reinforcement learning (RL) in this field. Based on a thorough review of literature about RL and process parameter optimization, a model based on maximum a posteriori policy optimization that can handle both numerical and categorical parameters is proposed. A case study compares the model to state–of–the–art traditional algorithms and shows that RL can find optima of similar quality while requiring significantly less time. These results are confirmed in a large-scale validation study on data sets from both production and other fields. Finally, multiple ways to improve the model are discussed.Keywords: reinforcement learning, production process optimization, evolutionary algorithms, policy optimization, actor critic approach
Procedia PDF Downloads 59920 Inclusion of Students with Disabilities (SWD) in Higher Education Institutions (HEIs): Self-Advocacy and Engagement as Central
Authors: Tadesse Abera
Abstract:
This study aimed to investigate the contribution of self-advocacy and engagement in the inclusion of SWDs in HEIs. A convergent parallel mixed methods design was employed. This article reports the quantitative strand. A total of 246 SWDs were selected through stratified proportionate random sampling technique from five public HEIs in Ethiopia. Data were collected through Self-advocacy questionnaire, student engagement scale, and college student experience questionnaire and analyzed through frequency, percentage, mean, standard deviation, correlation, one sample t-test and multiple regression. Both self-advocacy and engagement were found to have a predictive power on inclusion of respondents in the HEIs, where engagement was found to be more predictor. From the components of self-advocacy, knowledge of self and leadership and from engagement dimensions sense of belonging, cognitive, and valuing in their respective orders were found to have a stronger predictive power on the inclusion of respondents in the institutions. Based on the findings it was concluded that, if students with disabilities work hard to be self-determined, strive for realizing social justice, exert quality effort and seek active involvement, their inclusion in the institutions would be ensured.Keywords: self-advocacy, engagement, inclusion, students with disabilities, higher education institution
Procedia PDF Downloads 34919 Deepnic, A Method to Transform Each Variable into Image for Deep Learning
Authors: Nguyen J. M., Lucas G., Brunner M., Ruan S., Antonioli D.
Abstract:
Deep learning based on convolutional neural networks (CNN) is a very powerful technique for classifying information from an image. We propose a new method, DeepNic, to transform each variable of a tabular dataset into an image where each pixel represents a set of conditions that allow the variable to make an error-free prediction. The contrast of each pixel is proportional to its prediction performance and the color of each pixel corresponds to a sub-family of NICs. NICs are probabilities that depend on the number of inputs to each neuron and the range of coefficients of the inputs. Each variable can therefore be expressed as a function of a matrix of 2 vectors corresponding to an image whose pixels express predictive capabilities. Our objective is to transform each variable of tabular data into images into an image that can be analysed by CNNs, unlike other methods which use all the variables to construct an image. We analyse the NIC information of each variable and express it as a function of the number of neurons and the range of coefficients used. The predictive value and the category of the NIC are expressed by the contrast and the color of the pixel. We have developed a pipeline to implement this technology and have successfully applied it to genomic expressions on an Affymetrix chip.Keywords: tabular data, deep learning, perfect trees, NICS
Procedia PDF Downloads 52918 Implementing a Neural Network on a Low-Power and Mobile Cluster to Aide Drivers with Predictive AI for Traffic Behavior
Authors: Christopher Lama, Alix Rieser, Aleksandra Molchanova, Charles Thangaraj
Abstract:
New technologies like Tesla’s Dojo have made high-performance embedded computing more available. Although automobile computing has developed and benefited enormously from these more recent technologies, the costs are still high, prohibitively high in some cases for broader adaptation, particularly for the after-market and enthusiast markets. This project aims to implement a Raspberry Pi-based low-power (under one hundred Watts) highly mobile computing cluster for a neural network. The computing cluster built from off-the-shelf components is more affordable and, therefore, makes wider adoption possible. The paper describes the design of the neural network, Raspberry Pi-based cluster, and applications the cluster will run. The neural network will use input data from sensors and cameras to project a live view of the road state as the user drives. The neural network will be trained to predict traffic behavior and generate warnings when potentially dangerous situations are predicted. The significant outcomes of this study will be two folds, firstly, to implement and test the low-cost cluster, and secondly, to ascertain the effectiveness of the predictive AI implemented on the cluster.Keywords: CS pedagogy, student research, cluster computing, machine learning
Procedia PDF Downloads 53917 Carbohydrate Intake Estimation in Type I Diabetic Patients Described by UVA/Padova Model
Authors: David A. Padilla, Rodolfo Villamizar
Abstract:
In recent years, closed loop control strategies have been developed in order to establish a healthy glucose profile in type 1 diabetic mellitus (T1DM) patients. However, the controller itself is unable to define a suitable reference trajectory for glucose. In this paper, a control strategy Is proposed where the shape of the reference trajectory is generated bases in the amount of carbohydrates present during the digestive process, due to the effect of carbohydrate intake. Since there no exists a sensor to measure the amount of carbohydrates consumed, an estimator is proposed. Thus this paper presents the entire process of designing a carbohydrate estimator, which allows estimate disturbance for a predictive controller (MPC) in a T1MD patient, the estimation will be used to establish a profile of reference and improve the response of the controller by providing the estimated information of ingested carbohydrates. The dynamics of the diabetic model used are due to the equations described by the UVA/Padova model of the T1DMS simulator, the system was developed and simulated in Simulink, taking into account the noise and limitations of the glucose control system actuators.Keywords: estimation, glucose control, predictive controller, MPC, UVA/Padova
Procedia PDF Downloads 229