Search results for: prediction error

Authors: Piotr A. Kowalski, Kasper Sapala, Wiktor Warchalowski

Abstract:

PM10 is a suspended dust that primarily has a negative effect on the respiratory system. PM10 is responsible for attacks of coughing and wheezing, asthma or acute, violent bronchitis. Indirectly, PM10 also negatively affects the rest of the body, including increasing the risk of heart attack and stroke. Unfortunately, Poland is a country that cannot boast of good air quality, in particular, due to large PM concentration levels. Therefore, based on the dense network of Airly sensors, it was decided to deal with the problem of prediction of suspended particulate matter concentration. Due to the very complicated nature of this issue, the Machine Learning approach was used. For this purpose, Convolution Neural Network (CNN) neural networks have been adopted, these currently being the leading information processing methods in the field of computational intelligence. The aim of this research is to show the influence of particular CNN network parameters on the quality of the obtained forecast. The forecast itself is made on the basis of parameters measured by Airly sensors and is carried out for the subsequent day, hour after hour. The evaluation of learning process for the investigated models was mostly based upon the mean square error criterion; however, during the model validation, a number of other methods of quantitative evaluation were taken into account. The presented model of pollution prediction has been verified by way of real weather and air pollution data taken from the Airly sensor network. The dense and distributed network of Airly measurement devices enables access to current and archival data on air pollution, temperature, suspended particulate matter PM1.0, PM2.5, and PM10, CAQI levels, as well as atmospheric pressure and air humidity. In this investigation, PM2.5, and PM10, temperature and wind information, as well as external forecasts of temperature and wind for next 24h served as inputted data. Due to the specificity of the CNN type network, this data is transformed into tensors and then processed. This network consists of an input layer, an output layer, and many hidden layers. In the hidden layers, convolutional and pooling operations are performed. The output of this system is a vector containing 24 elements that contain prediction of PM10 concentration for the upcoming 24 hour period. Over 1000 models based on CNN methodology were tested during the study. During the research, several were selected out that give the best results, and then a comparison was made with the other models based on linear regression. The numerical tests carried out fully confirmed the positive properties of the presented method. These were carried out using real ‘big’ data. Models based on the CNN technique allow prediction of PM10 dust concentration with a much smaller mean square error than currently used methods based on linear regression. What's more, the use of neural networks increased Pearson's correlation coefficient (R²) by about 5 percent compared to the linear model. During the simulation, the R² coefficient was 0.92, 0.76, 0.75, 0.73, and 0.73 for 1st, 6th, 12th, 18th, and 24th hour of prediction respectively.

Keywords: air pollution prediction (forecasting), machine learning, regression task, convolution neural networks

Procedia PDF Downloads 149

Authors: Seyoung Kim, Jeongmin Kim, Kwang Ryel Ryu

Abstract:

A database that records average traffic speeds measured at five-minute intervals for all the links in the traffic network of a metropolitan city. While learning from this data the models that can predict future traffic speed would be beneficial for the applications such as the car navigation system, building predictive models for every link becomes a nontrivial job if the number of links in a given network is huge. An advantage of adopting k-nearest neighbor (k-NN) as predictive models is that it does not require any explicit model building. Instead, k-NN takes a long time to make a prediction because it needs to search for the k-nearest neighbors in the database at prediction time. In this paper, we investigate how much we can speed up k-NN in making traffic speed predictions by reducing the amount of data to be searched for without a significant sacrifice of prediction accuracy. The rationale behind this is that we had a better look at only the recent data because the traffic patterns not only repeat daily or weekly but also change over time. In our experiments, we build several different k-NN models employing different sets of features which are the current and past traffic speeds of the target link and the neighbor links in its up/down-stream. The performances of these models are compared by measuring the average prediction accuracy and the average time taken to make a prediction using various amounts of data.

Keywords: big data, k-NN, machine learning, traffic speed prediction

Procedia PDF Downloads 363

Authors: Yan Cheng, Yijun Shao, James Rudolph, Charlene R. Weir, Beth Sahlmann, Qing Zeng-Treitler

Abstract:

Supervised predictive models require labeled data for training purposes. Complete and accurate labeled data, i.e., a ‘gold standard’, is not always available, and imperfectly labeled data may need to serve as an alternative. An important question is if the accuracy of the labeled data creates a performance ceiling for the trained model. In this study, we trained several models to recognize the presence of delirium in clinical documents using data with annotations that are not completely accurate (i.e., weakly-supervised learning). In the external evaluation, the support vector machine model with a linear kernel performed best, achieving an area under the curve of 89.3% and accuracy of 88%, surpassing the 80% accuracy of the training sample. We then generated a set of simulated data and carried out a series of experiments which demonstrated that models trained on imperfect data can (but do not always) outperform the accuracy of the training data, e.g., the area under the curve for some models is higher than 80% when trained on the data with an error rate of 40%. Our experiments also showed that the error resistance of linear modeling is associated with larger sample size, error type, and linearity of the data (all p-values < 0.001). In conclusion, this study sheds light on the usefulness of imperfect data in clinical research via weakly-supervised learning.

Keywords: weakly-supervised learning, support vector machine, prediction, delirium, simulation

Procedia PDF Downloads 199

Authors: Matevž Breška, Iztok Peruš, Vlado Stankovski

Abstract:

Systematic overview of existing Ground Motion Prediction Equations (GMPEs) has been published by Douglas. The number of earthquake recordings that have been used for fitting these equations has increased in the past decades. The current PF-L database contains 3550 recordings. Since the GMPEs frequently model the peak ground acceleration (PGA) the goal of the present study was to refit a selection of 44 of the existing equation models for PGA in light of the latest data. The algorithm Levenberg-Marquardt was used for fitting the coefficients of the equations and the results are evaluated both quantitatively by presenting the root mean squared error (RMSE) and qualitatively by drawing graphs of the five best fitted equations. The RMSE was found to be as low as 0.08 for the best equation models. The newly estimated coefficients vary from the values published in the original works.

Keywords: Ground Motion Prediction Equations, Levenberg-Marquardt algorithm, refitting PF-L database, peak ground acceleration

Procedia PDF Downloads 462

Authors: Jiun Jeon, Byung-Ju Yi

Abstract:

This paper investigates modeling and shape prediction of elastic kinematic chains such as colonoscopy. 2D and 3D models of elastic kinematic chains are suggested and their behaviors are demonstrated through simulation. To corroborate the effectiveness of those models, experimental work is performed using a magnetic sensor system.

Keywords: elastic kinematic chain, shape prediction, colonoscopy, modeling

Procedia PDF Downloads 605

Authors: Neetu Manocha

Abstract:

Image segmentation is a technique where a picture is parted into distinct parts having similar features which have a place with similar items. Various segmentation strategies have been proposed as of late by prominent analysts. But, after ultimate thorough research, the novelists have analyzed that generally, the old methods do not decrease the segmentation error rate. Then author finds the technique HIST-SI to decrease the segmentation error rates. In this technique, cluster-based and threshold-based segmentation techniques are merged together. After then, to improve the result of HIST-SI, the authors added the method of filtering and linking in this technique named Imp_HIST-SI to decrease the segmentation error rates. The goal of this research is to find a new technique to decrease the segmentation error rates and produce much better results than the HIST-SI technique. For testing the proposed technique, a dataset of Bhuvan – a National Geoportal developed and hosted by ISRO (Indian Space Research Organisation) is used. Experiments are conducted using Scikit-image & OpenCV tools of Python, and performance is evaluated and compared over various existing image segmentation techniques for several matrices, i.e., Mean Square Error (MSE) and Peak Signal Noise Ratio (PSNR).

Keywords: satellite image, image segmentation, edge detection, error rate, MSE, PSNR, HIST-SI, linking, filtering, imp_HIST-SI

Procedia PDF Downloads 141

Authors: Ugwu O. C., Mamah R. O., Awudu W. S.

Abstract:

This work is aimed at enhancing signal reception on a mobile radio network and minimizing outage probability in a mobile radio network using adaptive beamforming antenna arrays. In this research work, an empirical real-time drive measurement was done in a cellular network of Globalcom Nigeria Limited located at Ikeja, the headquarters of Lagos State, Nigeria, with reference base station number KJA 004. The empirical measurement includes Received Signal Strength and Bit Error Rate which were recorded for exact prediction of the signal strength of the network as at the time of carrying out this research work. The Received Signal Strength and Bit Error Rate were measured with a spectrum monitoring Van with the help of a Ray Tracer at an interval of 100 meters up to 700 meters from the transmitting base station. The distance and angular location measurements from the reference network were done with the help Global Positioning System (GPS). The other equipment used were transmitting equipment measurements software (Temsoftware), Laptops and log files, which showed received signal strength with distance from the base station. Results obtained were about 11% from the real-time experiment, which showed that mobile radio networks are prone to signal failure and can be minimized using an Adaptive Beamforming Antenna Array in terms of a significant reduction in Bit Error Rate, which implies improved performance of the mobile radio network. In addition, this work did not only include experiments done through empirical measurement but also enhanced mathematical models that were developed and implemented as a reference model for accurate prediction. The proposed signal models were based on the analysis of continuous time and discrete space, and some other assumptions. These developed (proposed) enhanced models were validated using MATLAB (version 7.6.3.35) program and compared with the conventional antenna for accuracy. These outage models were used to manage the blocked call experience in the mobile radio network. 20% improvement was obtained when the adaptive beamforming antenna arrays were implemented on the wireless mobile radio network.

Keywords: beamforming algorithm, adaptive beamforming, simulink, reception

Procedia PDF Downloads 41

Authors: Cinna Soltanpur, Mohammad Ghamari, Behzad Momahed Heravi, Fatemeh Zare

Abstract:

Low-density parity-check (LDPC) codes have been shown to deliver capacity approaching performance; however, problematic graphical structures (e.g. trapping sets) in the Tanner graph of some LDPC codes can cause high error floors in bit-error-ratio (BER) performance under conventional sum-product algorithm (SPA). This paper presents a serial concatenation scheme to avoid the trapping sets and to lower the error floors of LDPC code. The outer code in the proposed concatenation is the LDPC, and the inner code is a high rate array code. This approach applies an interactive hybrid process between the BCJR decoding for the array code and the SPA for the LDPC code together with bit-pinning and bit-flipping techniques. Margulis code of size (2640, 1320) has been used for the simulation and it has been shown that the proposed concatenation and decoding scheme can considerably improve the error floor performance with minimal rate loss.

Keywords: concatenated coding, low–density parity–check codes, array code, error floors

Procedia PDF Downloads 356

Authors: Li Yu, Chenlu Jiao, Hongrun Xin, Yan Wang, Kaiyang Wang

Abstract:

In order to study the impact of various factors on the housing price, we propose to build different prediction models based on deep learning to determine the existing data of the real estate in order to more accurately predict the housing price or its changing trend in the future. Considering that the factors which affect the housing price vary widely, the proposed prediction models include two categories. The first one is based on multiple characteristic factors of the real estate. We built Convolution Neural Network (CNN) prediction model and Long Short-Term Memory (LSTM) neural network prediction model based on deep learning, and logical regression model was implemented to make a comparison between these three models. Another prediction model is time series model. Based on deep learning, we proposed an LSTM-1 model purely regard to time series, then implementing and comparing the LSTM model and the Auto-Regressive and Moving Average (ARMA) model. In this paper, comprehensive study of the second-hand housing price in Beijing has been conducted from three aspects: crawling and analyzing, housing price predicting, and the result comparing. Ultimately the best model program was produced, which is of great significance to evaluation and prediction of the housing price in the real estate industry.

Keywords: deep learning, convolutional neural network, LSTM, housing prediction

Procedia PDF Downloads 306

Authors: Bharatendra Rai

Abstract:

Predictive data analysis and modeling involving machine learning techniques become challenging in presence of too many explanatory variables or features. Presence of too many features in machine learning is known to not only cause algorithms to slow down, but they can also lead to decrease in model prediction accuracy. This study involves housing dataset with 79 quantitative and qualitative features that describe various aspects people consider while buying a new house. Boruta algorithm that supports feature selection using a wrapper approach build around random forest is used in this study. This feature selection process leads to 49 confirmed features which are then used for developing predictive random forest models. The study also explores five different data partitioning ratios and their impact on model accuracy are captured using coefficient of determination (r-square) and root mean square error (rsme).

Keywords: housing data, feature selection, random forest, Boruta algorithm, root mean square error

Procedia PDF Downloads 323

Authors: Madhawa Basnayaka, Jouni Paltakari

Abstract:

Fully passive backscattering chipless RFID tags are an emerging wireless technology with low cost, higher reading distance, and fast automatic identification without human interference, unlike already available technologies like optical barcodes. The design optimization of chipless RFID tags is crucial as it requires replacing integrated chips found in conventional RFID tags with printed geometric designs. These designs enable data encoding and decoding through backscattered electromagnetic (EM) signatures. The applications of chipless RFID tags have been limited due to the constraints of data encoding capacity and the ability to design accurate yet efficient configurations. The traditional approach to accomplishing design parameters for a desired EM response involves iterative adjustment of design parameters and simulating until the desired EM spectrum is achieved. However, traditional numerical simulation methods encounter limitations in optimizing design parameters efficiently due to the speed and resource consumption. In this work, a deep learning neural network (DNN) is utilized to establish a correlation between the EM spectrum and the dimensional parameters of nested centric rings, specifically square and octagonal. The proposed bi-directional DNN has two simultaneously running neural networks, namely spectrum prediction and design parameters prediction. First, spectrum prediction DNN was trained to minimize mean square error (MSE). After the training process was completed, the spectrum prediction DNN was able to accurately predict the EM spectrum according to the input design parameters within a few seconds. Then, the trained spectrum prediction DNN was connected to the design parameters prediction DNN and trained two networks simultaneously. For the first time in chipless tag design, design parameters were predicted accurately after training bi-directional DNN for a desired EM spectrum. The model was evaluated using a randomly generated spectrum and the tag was manufactured using the predicted geometrical parameters. The manufactured tags were successfully tested in the laboratory. The amount of iterative computer simulations has been significantly decreased by this approach. Therefore, highly efficient but ultrafast bi-directional DNN models allow rapid and complicated chipless RFID tag designs.

Keywords: artificial intelligence, chipless RFID, deep learning, machine learning

Procedia PDF Downloads 50

Authors: Sunyoung Bu, Wonkyu Chung, Philsu Kim

Abstract:

In this paper, we introduce a method for improving the embedded Runge-Kutta-Fehlberg 4(5) method. At each integration step, the proposed method is comprised of two equations for the solution and the error, respectively. This solution and error are obtained by solving an initial value problem whose solution has the information of the error at each integration step. The constructed algorithm controls both the error and the time step size simultaneously and possesses a good performance in the computational cost compared to the original method. For the assessment of the effectiveness, EULR problem is numerically solved.

Keywords: embedded Runge-Kutta-Fehlberg method, initial value problem, EULR problem, integration step

Procedia PDF Downloads 463

Authors: James Purnama, Riri Fitri Sari

Abstract:

We investigate the machine learning algorithm selection problem in the term of a depth image based eye gaze estimation, with respect to its essential difficulty in reducing the number of required training samples and duration time of training. Statistics based prediction accuracy are increasingly used to assess and evaluate prediction or estimation in gaze estimation. This article evaluates Root Mean Squared Error (RMSE) and R-Squared statistical analysis to assess machine learning methods on depth camera data for gaze estimation. There are 4 machines learning methods have been evaluated: Random Forest Regression, Regression Tree, Support Vector Machine (SVM), and Linear Regression. The experiment results show that the Random Forest Regression has the lowest RMSE and the highest R-Squared, which means that it is the best among other methods.

Keywords: gaze estimation, gaze tracking, eye tracking, kinect, regression model, orange python

Procedia PDF Downloads 538

Authors: A. Ejah Umraeni Salam, M. Tola, M. Selintung, F. Maricar

Abstract:

Clean water is an essential and fundamental human need. Therefore, its supply must be assured by maintaining the quality, quantity and water pressure. However the fact is, on its distribution system, leakage happens and becomes a common world issue. One of the technical causes of the leakage is a leaking pipe. The purpose of the research is how to use the Radial Basis Function Neural (RBFNN) model to detect the location and the magnitude of the pipeline leakage rapidly and efficiently. In this study the RBFNN are trained and tested on data from EPANET hydraulic modeling system. Method of Radial Basis Function Neural Network is proved capable to detect location and magnitude of pipeline leakage with of the accuracy of the prediction results based on the value of RMSE (Root Meant Square Error), comparison prediction and actual measurement approaches 0.000049 for the whole pipeline system.

Keywords: radial basis function neural network, leakage pipeline, EPANET, RMSE

Procedia PDF Downloads 358

Authors: Houcine Naim, Abdelatif Hassini, Noureddine Benabadji, Alex Van Den Bossche

Abstract:

In this paper, two models to estimate the overall monthly average daily radiation on a horizontal surface were applied to the site of Oran (35.38 ° N, 0.37 °W). We present a comparison between the first one is a regression equation of the Angstrom type and the second model is developed by the present authors some modifications were suggested using as input parameters: the astronomical parameters as (latitude, longitude, and altitude) and meteorological parameters as (relative humidity). The comparisons are made using the mean bias error (MBE), root mean square error (RMSE), mean percentage error (MPE), and mean absolute bias error (MABE). This comparison shows that the second model is closer to the experimental values that the model of Angstrom.

Keywords: meteorology, global radiation, Angstrom model, Oran

Procedia PDF Downloads 233

Authors: Gaurav Sancheti

Abstract:

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

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

Procedia PDF Downloads 221

Authors: Rami El-Hajj Mohamad, Mahmoud Skafi, Ali Massoud Haidar

Abstract:

Several meteorological parameters were used for the prediction of monthly average daily global solar radiation on horizontal using recurrent neural networks (RNNs). Climatological data and measures, mainly air temperature, humidity, sunshine duration, and wind speed between 1995 and 2007 were used to design and validate a feed forward and recurrent neural network based prediction systems. In this paper we present our reference system based on a feed-forward multilayer perceptron (MLP) as well as the proposed approach based on an RNN model. The obtained results were promising and comparable to those obtained by other existing empirical and neural models. The experimental results showed the advantage of RNNs over simple MLPs when we deal with time series solar radiation predictions based on daily climatological data.

Keywords: recurrent neural networks, global solar radiation, multi-layer perceptron, gradient, root mean square error

Procedia PDF Downloads 444

Authors: Dimitrios Triantakonstantis, Demetris Stathakis

Abstract:

Urban areas have been expanded throughout the globe. Monitoring and modeling urban growth have become a necessity for a sustainable urban planning and decision making. Urban prediction models are important tools for analyzing the causes and consequences of urban land use dynamics. The objective of this research paper is to analyze and model the urban change, which has been occurred from 1990 to 2000 using CORINE land cover maps. The model was developed using drivers of urban changes (such as road distance, slope, etc.) under an Artificial Neural Network modeling approach. Validation was achieved using a prediction map for 2006 which was compared with a real map of Urban Atlas of 2006. The accuracy produced a Kappa index of agreement of 0,639 and a value of Cramer's V of 0,648. These encouraging results indicate the importance of the developed urban growth prediction model which using a set of available common biophysical drivers could serve as a management tool for the assessment of urban change.

Keywords: artificial neural networks, CORINE, urban atlas, urban growth prediction

Procedia PDF Downloads 529

Authors: Takuro Kida, Yuichi Kida

Abstract:

We show a base of the new trend of algorithm mathematically that treats a historical reason of continuous discrimination in the world as well as its solution by introducing new concepts of parallel world that includes an invisible set of errors as its companion. With respect to a matrix operator-filter bank that the matrix operator-analysis-filter bank H and the matrix operator-sampling-filter bank S are given, firstly, we introduce the detailed algorithm to derive the optimum matrix operator-synthesis-filter bank Z that minimizes all the worst-case measures of the matrix operator-error-signals E(ω) = F(ω) − Y(ω) between the matrix operator-input-signals F(ω) and the matrix operator-output signals Y(ω) of the matrix operator-filter bank at the same time. Further, feedback is introduced to the above approximation theory and it is indicated that introducing conversations with feedback does not superior automatically to the accumulation of existing knowledge of signal prediction. Secondly, the concept of category in the field of mathematics is applied to the above optimum signal approximation and is indicated that the category-based approximation theory is applied to the set-theoretic consideration of the recognition of humans. Based on this discussion, it is shown naturally why the narrow perception that tends to create isolation shows an apparent advantage in the short term and, often, why such narrow thinking becomes intimate with discriminatory action in a human group. Throughout these considerations, it is presented that, in order to abolish easy and intimate discriminatory behavior, it is important to create a parallel world of conception where we share the set of invisible error signals, including the words and the consciousness of both worlds.

Keywords: signal prediction, pseudo inverse matrix, artificial intelligence, conditional optimization

Procedia PDF Downloads 156

Authors: P. S. Jagadeesh Kumar, S. Meenakshi Sundaram, Wenli Hu, Yang Yung

Abstract:

In 3D video games, the dominance of production is unceasingly growing with a protruding level of affordability in terms of budget. Afterward, the automation of speech-lip synchronization technique is customarily onerous and has advanced a critical research subject in virtual reality based 3D video games. This paper presents one of these automatic tools, precisely riveted on the synchronization of the speech and the lip movement of the game characters. A robust and precise speech recognition segment that systematized with Algebraic Code Excited Linear Prediction method is developed which unconventionally delivers lip sync results. The Algebraic Code Excited Linear Prediction algorithm is constructed on that used in code-excited linear prediction, but Algebraic Code Excited Linear Prediction codebooks have an explicit algebraic structure levied upon them. This affords a quicker substitute to the software enactments of lip sync algorithms and thus advances the superiority of service factors abridged production cost.

Keywords: algebraic code excited linear prediction, speech-lip synchronization, video games, virtual reality

Procedia PDF Downloads 474

Authors: Pradeep Singh, Shrish Verma

Abstract:

Software fault prediction models are created by using the source code, processed metrics from the same or previous version of code and related fault data. Some company do not store and keep track of all artifacts which are required for software fault prediction. To construct fault prediction model for such company, the training data from the other projects can be one potential solution. The earlier we predict the fault the less cost it requires to correct. The training data consists of metrics data and related fault data at function/module level. This paper investigates fault predictions at early stage using the cross-project data focusing on the design metrics. In this study, empirical analysis is carried out to validate design metrics for cross project fault prediction. The machine learning techniques used for evaluation is Naïve Bayes. The design phase metrics of other projects can be used as initial guideline for the projects where no previous fault data is available. We analyze seven data sets from NASA Metrics Data Program which offer design as well as code metrics. Overall, the results of cross project is comparable to the within company data learning.

Keywords: software metrics, fault prediction, cross project, within project.

Procedia PDF Downloads 344

Authors: Watchareeporn Chaimongkol

Abstract:

In this paper, we compare the statistical measures accuracy of composite forecasting model to estimate automobile customer demand in Thailand. A modified simple exponential smoothing and autoregressive integrate moving average (ARIMA) forecasting model is built to estimate customer demand of passenger cars, instead of using information of historical sales data. Our model takes into account special characteristic of the Thai automobile market such as sales promotion, advertising and publicity, petrol price, and interest rate for loan. We evaluate our forecasting model by comparing forecasts with actual data using six accuracy measurements, mean absolute percentage error (MAPE), geometric mean absolute error (GMAE), symmetric mean absolute percentage error (sMAPE), mean absolute scaled error (MASE), median relative absolute error (MdRAE), and geometric mean relative absolute error (GMRAE).

Keywords: composite forecasting, simple exponential smoothing model, autoregressive integrate moving average model selection, accuracy measurements

Procedia PDF Downloads 362

Authors: Man Fung Ho, Lap So, Jiaqi Zhang, Yuheng Zhao, Huiyang Lu, Tat Shing Choi, K. Y. Michael Wong

Abstract:

Nowadays, a tremendous amount of data is available in the transportation system, enabling the development of various machine learning approaches to make short-term latency predictions. A natural question is then the choice of relevant information to enable accurate predictions. Using traffic data collected from the Taiwan Freeway System, we consider the prediction of short-term latency of a freeway segment with a length of 17 km covering 5 measurement points, each collecting vehicle-by-vehicle data through the electronic toll collection system. The processed data include the past latencies of the freeway segment with different time lags, the traffic conditions of the individual segments (the accumulations, the traffic fluxes, the entrance and exit rates), the total accumulations, and the weekday latency profiles obtained by Gaussian process regression of past data. We arrive at several important conclusions about how data should be refined to obtain accurate predictions, which have implications for future system-wide latency predictions. (1) We find that the prediction of median latency is much more accurate and meaningful than the prediction of average latency, as the latter is plagued by outliers. This is verified by machine-learning prediction using XGBoost that yields a 35% improvement in the mean square error of the 5-minute averaged latencies. (2) We find that the median latency of the segment 15 minutes ago is a very good baseline for performance comparison, and we have evidence that further improvement is achieved by machine learning approaches such as XGBoost and Long Short-Term Memory (LSTM). (3) By analyzing the feature importance score in XGBoost and calculating the mutual information between the inputs and the latencies to be predicted, we identify a sequence of inputs ranked in importance. It confirms that the past latencies are most informative of the predicted latencies, followed by the total accumulation, whereas inputs such as the entrance and exit rates are uninformative. It also confirms that the inputs are much less informative of the average latencies than the median latencies. (4) For predicting the latencies of segments composed of two or three sub-segments, summing up the predicted latencies of each sub-segment is more accurate than the one-step prediction of the whole segment, especially with the latency prediction of the downstream sub-segments trained to anticipate latencies several minutes ahead. The duration of the anticipation time is an increasing function of the traveling time of the upstream segment. The above findings have important implications to predicting the full set of latencies among the various locations in the freeway system.

Keywords: data refinement, machine learning, mutual information, short-term latency prediction

Procedia PDF Downloads 169

Authors: Abdullahi Jibrin, Aishetu Abdulkadir

Abstract:

The development of allometric models is crucial to accurate forest biomass/carbon stock assessment. The aim of this study was to develop a set of biomass prediction models that will enable the determination of total tree aboveground biomass for savannah woodland area in Niger State, Nigeria. Based on the data collected through biometric measurements of 1816 trees and destructive sampling of 36 trees, five species specific and one site specific models were developed. The sample size was distributed equally between the five most dominant species in the study site (Vitellaria paradoxa, Irvingia gabonensis, Parkia biglobosa, Anogeissus leiocarpus, Pterocarpus erinaceous). Firstly, the equations were developed for five individual species. Secondly these five species were mixed and were used to develop an allometric equation of mixed species. Overall, there was a strong positive relationship between total tree biomass and the stem diameter. The coefficient of determination (R2 values) ranging from 0.93 to 0.99 P < 0.001 were realised for the models; with considerable low standard error of the estimates (SEE) which confirms that the total tree above ground biomass has a significant relationship with the dbh. The F-test value for the biomass prediction models were also significant at p < 0.001 which indicates that the biomass prediction models are valid. This study recommends that for improved biomass estimates in the study site, the site specific biomass models should preferably be used instead of using generic models.

Keywords: allometriy, biomass, carbon stock , model, regression equation, woodland, inventory

Procedia PDF Downloads 448

Authors: Haozhe Xiang

Abstract:

With the rise of the Internet of Things era, intelligent products are gradually integrating into people's lives. Pedestrian trajectory prediction has become a key issue, which is crucial for the motion path planning of intelligent agents such as autonomous vehicles, robots, and drones. In the current technological context, deep learning technology is becoming increasingly sophisticated and gradually replacing traditional models. The pedestrian trajectory prediction algorithm combining neural networks and attention mechanisms has significantly improved prediction accuracy. Based on in-depth research on deep learning and pedestrian trajectory prediction algorithms, this article focuses on physical environment modeling and learning of historical trajectory time dependence. At the same time, social interaction between pedestrians and scene interaction between pedestrians and the environment were handled. An improved pedestrian trajectory prediction algorithm is proposed by analyzing the existing model architecture. With the help of these improvements, acceptable predicted trajectories were successfully obtained. Experiments on public datasets have demonstrated the algorithm's effectiveness and achieved acceptable results.

Keywords: deep learning, graph convolutional network, attention mechanism, LSTM

Procedia PDF Downloads 71

Authors: Vijay Pal Singh, Yogesh Chandra Kotiyal

Abstract:

Structures are a combination of various load carrying members which transfer the loads to the foundation from the superstructure safely. At the design stage, the loading of the structure is defined and appropriate material choices are made based upon their properties, mainly related to strength. The strength of materials kept on reducing with time because of many factors like environmental exposure and deformation caused by unpredictable external loads. Hence, to predict the strength of materials used in structures, various techniques are used. Among these techniques, Non-Destructive Techniques (NDT) are the one that can be used to predict the strength without damaging the structure. In the present study, the compressive strength of concrete has been predicted using Artificial Neural Network (ANN). The predicted strength was compared with the experimentally obtained actual compressive strength of concrete and equations were developed for different models. A good co-relation has been obtained between the predicted strength by these models and experimental values. Further, the co-relation has been developed using two NDT techniques for prediction of strength by regression analysis. It was found that the percentage error has been reduced between the predicted strength by using combined techniques in place of single techniques.

Keywords: rebound, ultra-sonic pulse, penetration, ANN, NDT, regression

Procedia PDF Downloads 428

Authors: Yong-Il Lee, Do-Yeon Hwang, Won-Seog Jeong, Duckshin Park

Abstract:

Recently, because of the high-speed trains forced ventilation, it is important to control the ventilation. The ventilation is for controlling various contaminants, temperature, and humidity. The high-speed train route is straight to a destination having a high speed. And there are many mountainous areas in Korea. So, tunnel rate is higher then other country. KTX HVAC block off the outdoor air, when entering tunnel. So the high tunnel rate is an effect of ventilation in the KTX cabin. It is important to reduction rate in CO2 concentration prediction. To meet the air quality of the public transport vehicles recommend standards, the KTX cabin of CO2 concentration should be managed. In this study, the concentration change was predicted by CO2 prediction simulation in route to be opened.

Keywords: CO2 prediction, KTX, ventilation, infrastructure and transportation engineering

Procedia PDF Downloads 544

Authors: Katawut Kaewbanjong

Abstract:

We analyzed a volume of data and found significant user satisfaction in software project factors. A statistical significance analysis (logistic regression) and collinearity analysis determined the significance factors from a group of 71 pre-defined factors from 191 software projects in ISBSG Release 12. The eight prediction models used for testing the prediction potential of these factors were Neural network, k-NN, Naïve Bayes, Random forest, Decision tree, Gradient boosted tree, linear regression and logistic regression prediction model. Fifteen pre-defined factors were truly significant in predicting user satisfaction, and they provided 82.71% prediction accuracy when used with a neural network prediction model. These factors were client-server, personnel changes, total defects delivered, project inactive time, industry sector, application type, development type, how methodology was acquired, development techniques, decision making process, intended market, size estimate approach, size estimate method, cost recording method, and effort estimate method. These findings may benefit software development managers considerably.

Keywords: prediction model, statistical analysis, software project, user satisfaction factor

Procedia PDF Downloads 124

Authors: I. Alhajri, G. Zahedi, R. Alazmi, A. Akbari

Abstract:

In this paper, an Artificial Neural Network (ANN) was developed to predict Asphaltene Precipitation (AP) during the injection of carbon dioxide into crude oil reservoirs. In this study, the experimental data from six different oil fields were collected. Seventy percent of the data was used to develop the ANN model, and different ANN architectures were examined. A network with the Trainlm training algorithm was found to be the best network to estimate the AP. To check the validity of the proposed model, the model was used to predict the AP for the thirty percent of the data that was unevaluated. The Mean Square Error (MSE) of the prediction was 0.0018, which confirms the excellent prediction capability of the proposed model. In the second part of this study, the ANN model predictions were compared with modified Hirschberg model predictions. The ANN was found to provide more accurate estimates compared to the modified Hirschberg model. Finally, the proposed model was employed to examine the effect of different operating parameters during gas injection on the AP. It was found that the AP is mostly sensitive to the reservoir temperature. Furthermore, the carbon dioxide concentration in liquid phase increases the AP.

Keywords: artificial neural network, asphaltene, CO2 injection, Hirschberg model, oil reservoirs

Procedia PDF Downloads 364

Authors: Aaditya U. Jhamb

Abstract:

Due to the current energy crisis that many countries are battling, energy-efficient buildings are the subject of extensive research in the modern technological era because of growing worries about energy consumption and its effects on the environment. The paper explores 8 factors that help determine energy efficiency for a building: (relative compactness, surface area, wall area, roof area, overall height, orientation, glazing area, and glazing area distribution), with Tsanas and Xifara providing a dataset. The data set employed 768 different residential building models to anticipate heating and cooling loads with a low mean squared error. By optimizing these characteristics, machine learning algorithms may assess and properly forecast a building's heating and cooling loads, lowering energy usage while increasing the quality of people's lives. As a result, the paper studied the magnitude of the correlation between these input factors and the two output variables using various statistical methods of analysis after determining which input variable was most closely associated with the output loads. The most conclusive model was the Decision Tree Regressor, which had a mean squared error of 0.258, whilst the least definitive model was the Isotonic Regressor, which had a mean squared error of 21.68. This paper also investigated the KNN Regressor and the Linear Regression, which had to mean squared errors of 3.349 and 18.141, respectively. In conclusion, the model, given the 8 input variables, was able to predict the heating and cooling loads of a residential building accurately and precisely.

Keywords: energy efficient buildings, heating load, cooling load, machine learning models

Procedia PDF Downloads 96