Search results for: prediction modelling
3317 Multi-Linear Regression Based Prediction of Mass Transfer by Multiple Plunging Jets
Abstract:
The paper aims to compare the performance of vertical and inclined multiple plunging jets and to model and predict their mass transfer capacity by multi-linear regression based approach. The multiple vertical plunging jets have jet impact angle of θ = 90O; whereas, multiple inclined plunging jets have jet impact angle of θ = 600. The results of the study suggests that mass transfer is higher for multiple jets, and inclined multiple plunging jets have up to 1.6 times higher mass transfer than vertical multiple plunging jets under similar conditions. The derived relationship, based on multi-linear regression approach, has successfully predicted the volumetric mass transfer coefficient (KLa) from operational parameters of multiple plunging jets with a correlation coefficient of 0.973, root mean square error of 0.002 and coefficient of determination of 0.946. The results suggests that predicted overall mass transfer coefficient is in good agreement with actual experimental values; thereby suggesting the utility of derived relationship based on multi-linear regression based approach and can be successfully employed in modelling mass transfer by multiple plunging jets.Keywords: mass transfer, multiple plunging jets, multi-linear regression, earth sciences
Procedia PDF Downloads 4423316 Analytical Authentication of Butter Using Fourier Transform Infrared Spectroscopy Coupled with Chemometrics
Authors: M. Bodner, M. Scampicchio
Abstract:
Fourier Transform Infrared (FT-IR) spectroscopy coupled with chemometrics was used to distinguish between butter samples and non-butter samples. Further, quantification of the content of margarine in adulterated butter samples was investigated. Fingerprinting region (1400-800 cm–1) was used to develop unsupervised pattern recognition (Principal Component Analysis, PCA), supervised modeling (Soft Independent Modelling by Class Analogy, SIMCA), classification (Partial Least Squares Discriminant Analysis, PLS-DA) and regression (Partial Least Squares Regression, PLS-R) models. PCA of the fingerprinting region shows a clustering of the two sample types. All samples were classified in their rightful class by SIMCA approach; however, nine adulterated samples (between 1% and 30% w/w of margarine) were classified as belonging both at the butter class and at the non-butter one. In the two-class PLS-DA model’s (R2 = 0.73, RMSEP, Root Mean Square Error of Prediction = 0.26% w/w) sensitivity was 71.4% and Positive Predictive Value (PPV) 100%. Its threshold was calculated at 7% w/w of margarine in adulterated butter samples. Finally, PLS-R model (R2 = 0.84, RMSEP = 16.54%) was developed. PLS-DA was a suitable classification tool and PLS-R a proper quantification approach. Results demonstrate that FT-IR spectroscopy combined with PLS-R can be used as a rapid, simple and safe method to identify pure butter samples from adulterated ones and to determine the grade of adulteration of margarine in butter samples.Keywords: adulterated butter, margarine, PCA, PLS-DA, PLS-R, SIMCA
Procedia PDF Downloads 1293315 All-or-None Principle and Weakness of Hodgkin-Huxley Mathematical Model
Authors: S. A. Sadegh Zadeh, C. Kambhampati
Abstract:
Mathematical and computational modellings are the necessary tools for reviewing, analysing, and predicting processes and events in the wide spectrum range of scientific fields. Therefore, in a field as rapidly developing as neuroscience, the combination of these two modellings can have a significant role in helping to guide the direction the field takes. The paper combined mathematical and computational modelling to prove a weakness in a very precious model in neuroscience. This paper is intended to analyse all-or-none principle in Hodgkin-Huxley mathematical model. By implementation the computational model of Hodgkin-Huxley model and applying the concept of all-or-none principle, an investigation on this mathematical model has been performed. The results clearly showed that the mathematical model of Hodgkin-Huxley does not observe this fundamental law in neurophysiology to generating action potentials. This study shows that further mathematical studies on the Hodgkin-Huxley model are needed in order to create a model without this weakness.Keywords: all-or-none, computational modelling, mathematical model, transmembrane voltage, action potential
Procedia PDF Downloads 5973314 Genetic Algorithm Optimization of Microcantilever Based Resonator
Authors: Manjula Sutagundar, B. G. Sheeparamatti, D. S. Jangamshetti
Abstract:
Micro Electro Mechanical Systems (MEMS) resonators have shown the potential of replacing quartz crystal technology for sensing and high frequency signal processing applications because of inherent advantages like small size, high quality factor, low cost, compatibility with integrated circuit chips. This paper presents the optimization and modelling and simulation of the optimized micro cantilever resonator. The objective of the work is to optimize the dimensions of a micro cantilever resonator for a specified range of resonant frequency and specific quality factor. Optimization is carried out using genetic algorithm. The genetic algorithm is implemented using MATLAB. The micro cantilever resonator is modelled in CoventorWare using the optimized dimensions obtained from genetic algorithm. The modeled cantilever is analysed for resonance frequency.Keywords: MEMS resonator, genetic algorithm, modelling and simulation, optimization
Procedia PDF Downloads 5313313 Risk Assessment of Building Information Modelling Adoption in Construction Projects
Authors: Amirhossein Karamoozian, Desheng Wu, Behzad Abbasnejad
Abstract:
Building information modelling (BIM) is a new technology to enhance the efficiency of project management in the construction industry. In addition to the potential benefits of this useful technology, there are various risks and obstacles to applying it in construction projects. In this study, a decision making approach is presented for risk assessment in BIM adoption in construction projects. Various risk factors of exerting BIM during different phases of the project lifecycle are identified with the help of Delphi method, experts’ opinions and related literature. Afterward, Shannon’s entropy and Fuzzy TOPSIS (Technique for Order Preference by Similarity to Ideal Situation) are applied to derive priorities of the identified risk factors. Results indicated that lack of knowledge between professional engineers about workflows in BIM and conflict of opinions between different stakeholders are the risk factors with the highest priority.Keywords: risk, BIM, fuzzy TOPSIS, construction projects
Procedia PDF Downloads 2123312 Fuzzy Neuro Approach for Integrated Water Management System
Authors: Stuti Modi, Aditi Kambli
Abstract:
This paper addresses the need for intelligent water management and distribution system in smart cities to ensure optimal consumption and distribution of water for drinking and sanitation purposes. Water being a limited resource in cities require an effective system for collection, storage and distribution. In this paper, applications of two mostly widely used particular types of data-driven models, namely artificial neural networks (ANN) and fuzzy logic-based models, to modelling in the water resources management field are considered. The objective of this paper is to review the principles of various types and architectures of neural network and fuzzy adaptive systems and their applications to integrated water resources management. Final goal of the review is to expose and formulate progressive direction of their applicability and further research of the AI-related and data-driven techniques application and to demonstrate applicability of the neural networks, fuzzy systems and other machine learning techniques in the practical issues of the regional water management. Apart from this the paper will deal with water storage, using ANN to find optimum reservoir level and predicting peak daily demands.Keywords: artificial neural networks, fuzzy systems, peak daily demand prediction, water management and distribution
Procedia PDF Downloads 1663311 Integration of Big Data to Predict Transportation for Smart Cities
Authors: Sun-Young Jang, Sung-Ah Kim, Dongyoun Shin
Abstract:
The Intelligent transportation system is essential to build smarter cities. Machine learning based transportation prediction could be highly promising approach by delivering invisible aspect visible. In this context, this research aims to make a prototype model that predicts transportation network by using big data and machine learning technology. In detail, among urban transportation systems this research chooses bus system. The research problem that existing headway model cannot response dynamic transportation conditions. Thus, bus delay problem is often occurred. To overcome this problem, a prediction model is presented to fine patterns of bus delay by using a machine learning implementing the following data sets; traffics, weathers, and bus statues. This research presents a flexible headway model to predict bus delay and analyze the result. The prototyping model is composed by real-time data of buses. The data are gathered through public data portals and real time Application Program Interface (API) by the government. These data are fundamental resources to organize interval pattern models of bus operations as traffic environment factors (road speeds, station conditions, weathers, and bus information of operating in real-time). The prototyping model is designed by the machine learning tool (RapidMiner Studio) and conducted tests for bus delays prediction. This research presents experiments to increase prediction accuracy for bus headway by analyzing the urban big data. The big data analysis is important to predict the future and to find correlations by processing huge amount of data. Therefore, based on the analysis method, this research represents an effective use of the machine learning and urban big data to understand urban dynamics.Keywords: big data, machine learning, smart city, social cost, transportation network
Procedia PDF Downloads 2413310 Development of Deep Neural Network-Based Strain Values Prediction Models for Full-Scale Reinforced Concrete Frames Using Highly Flexible Sensing Sheets
Authors: Hui Zhang, Sherif Beskhyroun
Abstract:
Structural Health monitoring systems (SHM) are commonly used to identify and assess structural damage. In terms of damage detection, SHM needs to periodically collect data from sensors placed in the structure as damage-sensitive features. This includes abnormal changes caused by the strain field and abnormal symptoms of the structure, such as damage and deterioration. Currently, deploying sensors on a large scale in a building structure is a challenge. In this study, a highly stretchable strain sensors are used in this study to collect data sets of strain generated on the surface of full-size reinforced concrete (RC) frames under extreme cyclic load application. This sensing sheet can be switched freely between the test bending strain and the axial strain to achieve two different configurations. On this basis, the deep neural network prediction model of the frame beam and frame column is established. The training results show that the method can accurately predict the strain value and has good generalization ability. The two deep neural network prediction models will also be deployed in the SHM system in the future as part of the intelligent strain sensor system.Keywords: strain sensing sheets, deep neural networks, strain measurement, SHM system, RC frames
Procedia PDF Downloads 773309 Drying and Transport Processes in Distributed Hydrological Modelling Based on Finite Volume Schemes (Iber Model)
Authors: Carlos Caro, Ernest Bladé, Pedro Acosta, Camilo Lesmes
Abstract:
The drying-wet process is one of the topics to be more careful in distributed hydrological modeling using finite volume schemes as a means of solving the equations of Saint Venant. In a hydrologic and hydraulic computer model, surface flow phenomena depend mainly on the different flow accumulation and subsequent runoff generation. These accumulations are generated by routing, cell by cell, from the heights of water, which begin to appear due to the rain at each instant of time. Determine when it is considered a dry cell and when considered wet to include in the full calculation is an issue that directly affects the quantification of direct runoff or generation of flow at the end of a zone of contribution by accumulations flow generated from cells or finite volume.Keywords: hydrology, transport processes, hydrological modelling, finite volume schemes
Procedia PDF Downloads 3753308 Algorithm and Software Based on Multilayer Perceptron Neural Networks for Estimating Channel Use in the Spectral Decision Stage in Cognitive Radio Networks
Authors: Danilo López, Johana Hernández, Edwin Rivas
Abstract:
The use of the Multilayer Perceptron Neural Networks (MLPNN) technique is presented to estimate the future state of use of a licensed channel by primary users (PUs); this will be useful at the spectral decision stage in cognitive radio networks (CRN) to determine approximately in which time instants of future may secondary users (SUs) opportunistically use the spectral bandwidth to send data through the primary wireless network. To validate the results, sequences of occupancy data of channel were generated by simulation. The results show that the prediction percentage is greater than 60% in some of the tests carried out.Keywords: cognitive radio, neural network, prediction, primary user
Procedia PDF Downloads 3533307 Metabolic Predictive Model for PMV Control Based on Deep Learning
Authors: Eunji Choi, Borang Park, Youngjae Choi, Jinwoo Moon
Abstract:
In this study, a predictive model for estimating the metabolism (MET) of human body was developed for the optimal control of indoor thermal environment. Human body images for indoor activities and human body joint coordinated values were collected as data sets, which are used in predictive model. A deep learning algorithm was used in an initial model, and its number of hidden layers and hidden neurons were optimized. Lastly, the model prediction performance was analyzed after the model being trained through collected data. In conclusion, the possibility of MET prediction was confirmed, and the direction of the future study was proposed as developing various data and the predictive model.Keywords: deep learning, indoor quality, metabolism, predictive model
Procedia PDF Downloads 2413306 Modelling the Effect of Distancing and Wearing of Face Masks on Transmission of COVID-19 Infection Dynamics
Authors: Nurudeen Oluwasola Lasisi
Abstract:
The COVID-19 is an infection caused by coronavirus, which has been designated as a pandemic in the world. In this paper, we proposed a model to study the effect of distancing and wearing masks on the transmission of COVID-19 infection dynamics. The invariant region of the model is established. The COVID-19 free equilibrium and the reproduction number of the model were obtained. The local and global stability of the model is determined using the linearization technique method and Lyapunov method. It was found that COVID-19 free equilibrium state is locally asymptotically stable in feasible region Ω if R₀ < 1 and globally asymptomatically stable if R₀ < 1, otherwise unstable if R₀ > 1. More so, numerical analysis and simulations of the dynamics of the COVID-19 infection are presented.Keywords: distancing, reproduction number, wearing of mask, local and global stability, modelling, transmission
Procedia PDF Downloads 1263305 Business Intelligence for Profiling of Telecommunication Customer
Authors: Rokhmatul Insani, Hira Laksmiwati Soemitro
Abstract:
Business Intelligence is a methodology that exploits the data to produce information and knowledge systematically, business intelligence can support the decision-making process. Some methods in business intelligence are data warehouse and data mining. A data warehouse can store historical data from transactional data. For data modelling in data warehouse, we apply dimensional modelling by Kimball. While data mining is used to extracting patterns from the data and get insight from the data. Data mining has many techniques, one of which is segmentation. For profiling of telecommunication customer, we use customer segmentation according to customer’s usage of services, customer invoice and customer payment. Customers can be grouped according to their characteristics and can be identified the profitable customers. We apply K-Means Clustering Algorithm for segmentation. The input variable for that algorithm we use RFM (Recency, Frequency and Monetary) model. All process in data mining, we use tools IBM SPSS modeller.Keywords: business intelligence, customer segmentation, data warehouse, data mining
Procedia PDF Downloads 4613304 Linking Business Process Models and System Models Based on Business Process Modelling
Authors: Faisal A. Aburub
Abstract:
Organizations today need to invest in software in order to run their businesses, and to the organizations’ objectives, the software should be in line with the business process. This research presents an approach for linking process models and system models. Particularly, the new approach aims to synthesize sequence diagram based on role activity diagram (RAD) model. The approach includes four steps namely: Create business process model using RAD, identify computerized activities, identify entities in sequence diagram and identify messages in sequence diagram. The new approach has been validated using the process of student registration in University of Petra as a case study. Further research is required to validate the new approach using different domains.Keywords: business process modelling, system models, role activity diagrams, sequence diagrams
Procedia PDF Downloads 3683303 Probabilistic-Based Design of Bridges under Multiple Hazards: Floods and Earthquakes
Authors: Kuo-Wei Liao, Jessica Gitomarsono
Abstract:
Bridge reliability against natural hazards such as floods or earthquakes is an interdisciplinary problem that involves a wide range of knowledge. Moreover, due to the global climate change, engineers have to design a structure against the multi-hazard threats. Currently, few of the practical design guideline has included such concept. The bridge foundation in Taiwan often does not have a uniform width. However, few of the researches have focused on safety evaluation of a bridge with a complex pier. Investigation of the scouring depth under such situation is very important. Thus, this study first focuses on investigating and improving the scour prediction formula for a bridge with complicated foundation via experiments and artificial intelligence. Secondly, a probabilistic design procedure is proposed using the established prediction formula for practical engineers under the multi-hazard attacks.Keywords: bridge, reliability, multi-hazards, scour
Procedia PDF Downloads 3573302 Machine Learning Development Audit Framework: Assessment and Inspection of Risk and Quality of Data, Model and Development Process
Authors: Jan Stodt, Christoph Reich
Abstract:
The usage of machine learning models for prediction is growing rapidly and proof that the intended requirements are met is essential. Audits are a proven method to determine whether requirements or guidelines are met. However, machine learning models have intrinsic characteristics, such as the quality of training data, that make it difficult to demonstrate the required behavior and make audits more challenging. This paper describes an ML audit framework that evaluates and reviews the risks of machine learning applications, the quality of the training data, and the machine learning model. We evaluate and demonstrate the functionality of the proposed framework by auditing an steel plate fault prediction model.Keywords: audit, machine learning, assessment, metrics
Procedia PDF Downloads 2513301 Measuring the Unmeasurable: A Project of High Risk Families Prediction and Management
Authors: Peifang Hsieh
Abstract:
The prevention of child abuse has aroused serious concerns in Taiwan because of the disparity between the increasing amount of reported child abuse cases that doubled over the past decade and the scarcity of social workers. New Taipei city, with the most population in Taiwan and over 70% of its 4 million citizens are migrant families in which the needs of children can be easily neglected due to insufficient support from relatives and communities, sees urgency for a social support system, by preemptively identifying and outreaching high-risk families of child abuse, so as to offer timely assistance and preventive measure to safeguard the welfare of the children. Big data analysis is the inspiration. As it was clear that high-risk families of child abuse have certain characteristics in common, New Taipei city decides to consolidate detailed background information data from departments of social affairs, education, labor, and health (for example considering status of parents’ employment, health, and if they are imprisoned, fugitives or under substance abuse), to cross-reference for accurate and prompt identification of the high-risk families in need. 'The Service Center for High-Risk Families' (SCHF) was established to integrate data cross-departmentally. By utilizing the machine learning 'random forest method' to build a risk prediction model which can early detect families that may very likely to have child abuse occurrence, the SCHF marks high-risk families red, yellow, or green to indicate the urgency for intervention, so as to those families concerned can be provided timely services. The accuracy and recall rates of the above model were 80% and 65%. This prediction model can not only improve the child abuse prevention process by helping social workers differentiate the risk level of newly reported cases, which may further reduce their major workload significantly but also can be referenced for future policy-making.Keywords: child abuse, high-risk families, big data analysis, risk prediction model
Procedia PDF Downloads 1183300 Non-Destructive Prediction System Using near Infrared Spectroscopy for Crude Palm Oil
Authors: Siti Nurhidayah Naqiah Abdull Rani, Herlina Abdul Rahim
Abstract:
Near infrared (NIR) spectroscopy has always been of great interest in the food and agriculture industries. The development of predictive models has facilitated the estimation process in recent years. In this research, 176 crude palm oil (CPO) samples acquired from Felda Johor Bulker Sdn Bhd were studied. A FOSS NIRSystem was used to tak e absorbance measurements from the sample. The wavelength range for the spectral measurement is taken at 1600nm to 1900nm. Partial Least Square Regression (PLSR) prediction model with 50 optimal number of principal components was implemented to study the relationship between the measured Free Fatty Acid (FFA) values and the measured spectral absorption. PLSR showed predictive ability of FFA values with correlative coefficient (R) of 0.9808 for the training set and 0.9684 for the testing set.Keywords: palm oil, fatty acid, NIRS, PLSR
Procedia PDF Downloads 1973299 Lexicon-Based Sentiment Analysis for Stock Movement Prediction
Authors: Zane Turner, Kevin Labille, Susan Gauch
Abstract:
Sentiment analysis is a broad and expanding field that aims to extract and classify opinions from textual data. Lexicon-based approaches are based on the use of a sentiment lexicon, i.e., a list of words each mapped to a sentiment score, to rate the sentiment of a text chunk. Our work focuses on predicting stock price change using a sentiment lexicon built from financial conference call logs. We present a method to generate a sentiment lexicon based upon an existing probabilistic approach. By using a domain-specific lexicon, we outperform traditional techniques and demonstrate that domain-specific sentiment lexicons provide higher accuracy than generic sentiment lexicons when predicting stock price change.Keywords: computational finance, sentiment analysis, sentiment lexicon, stock movement prediction
Procedia PDF Downloads 1133298 Lexicon-Based Sentiment Analysis for Stock Movement Prediction
Authors: Zane Turner, Kevin Labille, Susan Gauch
Abstract:
Sentiment analysis is a broad and expanding field that aims to extract and classify opinions from textual data. Lexicon-based approaches are based on the use of a sentiment lexicon, i.e., a list of words each mapped to a sentiment score, to rate the sentiment of a text chunk. Our work focuses on predicting stock price change using a sentiment lexicon built from financial conference call logs. We introduce a method to generate a sentiment lexicon based upon an existing probabilistic approach. By using a domain-specific lexicon, we outperform traditional techniques and demonstrate that domain-specific sentiment lexicons provide higher accuracy than generic sentiment lexicons when predicting stock price change.Keywords: computational finance, sentiment analysis, sentiment lexicon, stock movement prediction
Procedia PDF Downloads 1563297 Estimating Cyclone Intensity Using INSAT-3D IR Images Based on Convolution Neural Network Model
Authors: Divvela Vishnu Sai Kumar, Deepak Arora, Sheenu Rizvi
Abstract:
Forecasting a cyclone through satellite images consists of the estimation of the intensity of the cyclone and predicting it before a cyclone comes. This research work can help people to take safety measures before the cyclone comes. The prediction of the intensity of a cyclone is very important to save lives and minimize the damage caused by cyclones. These cyclones are very costliest natural disasters that cause a lot of damage globally due to a lot of hazards. Authors have proposed five different CNN (Convolutional Neural Network) models that estimate the intensity of cyclones through INSAT-3D IR images. There are a lot of techniques that are used to estimate the intensity; the best model proposed by authors estimates intensity with a root mean squared error (RMSE) of 10.02 kts.Keywords: estimating cyclone intensity, deep learning, convolution neural network, prediction models
Procedia PDF Downloads 1013296 Application of EEG Wavelet Power to Prediction of Antidepressant Treatment Response
Authors: Dorota Witkowska, Paweł Gosek, Lukasz Swiecicki, Wojciech Jernajczyk, Bruce J. West, Miroslaw Latka
Abstract:
In clinical practice, the selection of an antidepressant often degrades to lengthy trial-and-error. In this work we employ a normalized wavelet power of alpha waves as a biomarker of antidepressant treatment response. This novel EEG metric takes into account both non-stationarity and intersubject variability of alpha waves. We recorded resting, 19-channel EEG (closed eyes) in 22 inpatients suffering from unipolar (UD, n=10) or bipolar (BD, n=12) depression. The EEG measurement was done at the end of the short washout period which followed previously unsuccessful pharmacotherapy. The normalized alpha wavelet power of 11 responders was markedly different than that of 11 nonresponders at several, mostly temporoparietal sites. Using the prediction of treatment response based on the normalized alpha wavelet power, we achieved 81.8% sensitivity and 81.8% specificity for channel T4.Keywords: alpha waves, antidepressant, treatment outcome, wavelet
Procedia PDF Downloads 2973295 Social Media Data Analysis for Personality Modelling and Learning Styles Prediction Using Educational Data Mining
Authors: Srushti Patil, Preethi Baligar, Gopalkrishna Joshi, Gururaj N. Bhadri
Abstract:
In designing learning environments, the instructional strategies can be tailored to suit the learning style of an individual to ensure effective learning. In this study, the information shared on social media like Facebook is being used to predict learning style of a learner. Previous research studies have shown that Facebook data can be used to predict user personality. Users with a particular personality exhibit an inherent pattern in their digital footprint on Facebook. The proposed work aims to correlate the user's’ personality, predicted from Facebook data to the learning styles, predicted through questionnaires. For Millennial learners, Facebook has become a primary means for information sharing and interaction with peers. Thus, it can serve as a rich bed for research and direct the design of learning environments. The authors have conducted this study in an undergraduate freshman engineering course. Data from 320 freshmen Facebook users was collected. The same users also participated in the learning style and personality prediction survey. The Kolb’s Learning style questionnaires and Big 5 personality Inventory were adopted for the survey. The users have agreed to participate in this research and have signed individual consent forms. A specific page was created on Facebook to collect user data like personal details, status updates, comments, demographic characteristics and egocentric network parameters. This data was captured by an application created using Python program. The data captured from Facebook was subjected to text analysis process using the Linguistic Inquiry and Word Count dictionary. An analysis of the data collected from the questionnaires performed reveals individual student personality and learning style. The results obtained from analysis of Facebook, learning style and personality data were then fed into an automatic classifier that was trained by using the data mining techniques like Rule-based classifiers and Decision trees. This helps to predict the user personality and learning styles by analysing the common patterns. Rule-based classifiers applied for text analysis helps to categorize Facebook data into positive, negative and neutral. There were totally two models trained, one to predict the personality from Facebook data; another one to predict the learning styles from the personalities. The results show that the classifier model has high accuracy which makes the proposed method to be a reliable one for predicting the user personality and learning styles.Keywords: educational data mining, Facebook, learning styles, personality traits
Procedia PDF Downloads 2123294 GraphNPP: A Graphormer-Based Architecture for Network Performance Prediction in Software-Defined Networking
Authors: Hanlin Liu, Hua Li, Yintan AI
Abstract:
Network performance prediction (NPP) is essential for the management and optimization of software-defined networking (SDN) and contributes to improving the quality of service (QoS) in SDN to meet the requirements of users. Although current deep learning-based methods can achieve high effectiveness, they still suffer from some problems, such as difficulty in capturing global information of the network, inefficiency in modeling end-to-end network performance, and inadequate graph feature extraction. To cope with these issues, our proposed Graphormer-based architecture for NPP leverages the powerful graph representation ability of Graphormer to effectively model the graph structure data, and a node-edge transformation algorithm is designed to transfer the feature extraction object from nodes to edges, thereby effectively extracting the end-to-end performance characteristics of the network. Moreover, routing oriented centrality measure coefficient for nodes and edges is proposed respectively to assess their importance and influence within the graph. Based on this coefficient, an enhanced feature extraction method and an advanced centrality encoding strategy are derived to fully extract the structural information of the graph. Experimental results on three public datasets demonstrate that the proposed GraphNPP architecture can achieve state-of-the-art results compared to current NPP methods.Keywords: software-defined networking, network performance prediction, Graphormer, graph neural network
Procedia PDF Downloads 253293 Modelling Conceptual Quantities Using Support Vector Machines
Authors: Ka C. Lam, Oluwafunmibi S. Idowu
Abstract:
Uncertainty in cost is a major factor affecting performance of construction projects. To our knowledge, several conceptual cost models have been developed with varying degrees of accuracy. Incorporating conceptual quantities into conceptual cost models could improve the accuracy of early predesign cost estimates. Hence, the development of quantity models for estimating conceptual quantities of framed reinforced concrete structures using supervised machine learning is the aim of the current research. Using measured quantities of structural elements and design variables such as live loads and soil bearing pressures, response and predictor variables were defined and used for constructing conceptual quantities models. Twenty-four models were developed for comparison using a combination of non-parametric support vector regression, linear regression, and bootstrap resampling techniques. R programming language was used for data analysis and model implementation. Gross soil bearing pressure and gross floor loading were discovered to have a major influence on the quantities of concrete and reinforcement used for foundations. Building footprint and gross floor loading had a similar influence on beams and slabs. Future research could explore the modelling of other conceptual quantities for walls, finishes, and services using machine learning techniques. Estimation of conceptual quantities would assist construction planners in early resource planning and enable detailed performance evaluation of early cost predictions.Keywords: bootstrapping, conceptual quantities, modelling, reinforced concrete, support vector regression
Procedia PDF Downloads 1993292 Reliability Assessment of Various Empirical Formulas for Prediction of Scour Hole Depth (Plunge Pool) Using a Comprehensive Physical Model
Authors: Majid Galoie, Khodadad Safavi, Abdolreza Karami Nejad, Reza Roshan
Abstract:
In this study, a comprehensive scouring model has been developed in order to evaluate the accuracy of various empirical relationships which were suggested for prediction of scour hole depth in plunge pools by Martins, Mason, Chian and Veronese. For this reason, scour hole depths caused by free falling jets from a flip bucket to a plunge pool were investigated. In this study various discharges, angles, scouring times, etc. have been considered. The final results demonstrated that the all mentioned empirical formulas, except Mason formula, were reasonably agreement with the experimental data.Keywords: scour hole depth, plunge pool, physical model, reliability assessment
Procedia PDF Downloads 5163291 Neural Network Based Path Loss Prediction for Global System for Mobile Communication in an Urban Environment
Authors: Danladi Ali
Abstract:
In this paper, we measured GSM signal strength in the Dnepropetrovsk city in order to predict path loss in study area using nonlinear autoregressive neural network prediction and we also, used neural network clustering to determine average GSM signal strength receive at the study area. The nonlinear auto-regressive neural network predicted that the GSM signal is attenuated with the mean square error (MSE) of 2.6748dB, this attenuation value is used to modify the COST 231 Hata and the Okumura-Hata models. The neural network clustering revealed that -75dB to -95dB is received more frequently. This means that the signal strength received at the study is mostly weak signalKeywords: one-dimensional multilevel wavelets, path loss, GSM signal strength, propagation, urban environment and model
Procedia PDF Downloads 3683290 Movie Genre Preference Prediction Using Machine Learning for Customer-Based Information
Authors: Haifeng Wang, Haili Zhang
Abstract:
Most movie recommendation systems have been developed for customers to find items of interest. This work introduces a predictive model usable by small and medium-sized enterprises (SMEs) who are in need of a data-based and analytical approach to stock proper movies for local audiences and retain more customers. We used classification models to extract features from thousands of customers’ demographic, behavioral and social information to predict their movie genre preference. In the implementation, a Gaussian kernel support vector machine (SVM) classification model and a logistic regression model were established to extract features from sample data and their test error-in-sample were compared. Comparison of error-out-sample was also made under different Vapnik–Chervonenkis (VC) dimensions in the machine learning algorithm to find and prevent overfitting. Gaussian kernel SVM prediction model can correctly predict movie genre preferences in 85% of positive cases. The accuracy of the algorithm increased to 93% with a smaller VC dimension and less overfitting. These findings advance our understanding of how to use machine learning approach to predict customers’ preferences with a small data set and design prediction tools for these enterprises.Keywords: computational social science, movie preference, machine learning, SVM
Procedia PDF Downloads 2473289 Hybrid Renewable Energy System Development Towards Autonomous Operation: The Deployment Potential in Greece
Authors: Afroditi Zamanidou, Dionysios Giannakopoulos, Konstantinos Manolitsis
Abstract:
A notable amount of electrical energy demand in many countries worldwide is used to cover public energy demand for road, square and other public spaces’ lighting. Renewable energy can contribute in a significant way to the electrical energy demand coverage for public lighting. This paper focuses on the sizing and design of a hybrid energy system (HES) exploiting the solar-wind energy potential to meet the electrical energy needs of lighting roads, squares and other public spaces. Moreover, the proposed HES provides coverage of the electrical energy demand for a Wi-Fi hotspot and a charging hotspot for the end-users. Alongside the sizing of the energy production system of the proposed HES, in order to ensure a reliable supply without interruptions, a storage system is added and sized. Multiple scenarios of energy consumption are assumed and applied in order to optimize the sizing of the energy production system and the energy storage system. A database with meteorological prediction data for 51 areas in Greece is developed in order to assess the possible deployment of the proposed HES. Since there are detailed meteorological prediction data for all 51 areas under investigation, the use of these data is evaluated, comparing them to real meteorological data. The meteorological prediction data are exploited to form three hourly production profiles for each area for every month of the year; minimum, average and maximum energy production. The energy production profiles are combined with the energy consumption scenarios and the sizing results of the energy production system and the energy storage system are extracted and presented for every area. Finally, the economic performance of the proposed HES in terms of Levelized cost of energy is estimated by calculating and assessing construction, operation and maintenance costs.Keywords: energy production system sizing, Greece’s deployment potential, meteorological prediction data, wind-solar hybrid energy system, levelized cost of energy
Procedia PDF Downloads 1343288 Mixtures of Length-Biased Weibull Distributions for Loss Severity Modelling
Authors: Taehan Bae
Abstract:
In this paper, a class of length-biased Weibull mixtures is presented to model loss severity data. The proposed model generalizes the Erlang mixtures with the common scale parameter, and it shares many important modelling features, such as flexibility to fit various data distribution shapes and weak-denseness in the class of positive continuous distributions, with the Erlang mixtures. We show that the asymptotic tail estimate of the length-biased Weibull mixture is Weibull-type, which makes the model effective to fit loss severity data with heavy-tailed observations. A method of statistical estimation is discussed with applications on real catastrophic loss data sets.Keywords: Erlang mixture, length-biased distribution, transformed gamma distribution, asymptotic tail estimate, EM algorithm, expectation-maximization algorithm
Procedia PDF Downloads 209