Search results for: time series prediction

Authors: Xuyang Song, Jun Yin

Abstract:

The smart card has been an extremely universal tool in public transit. It collects a large amount of data on buses, urban railway transit, and ferries and provides possibilities for passenger profiling. This paper combines offline analysis of passenger profiling and real-time prediction to propose a method that can accurately predict the destination station in real-time when passengers tag on. Firstly, this article constructs a static database of user travel characteristics after identifying passenger travel patterns based on the Density-Based Spatial Clustering of Applications with Noise (DBSCAN). The dual travel passenger habits are identified: OD travel habits and D station travel habits. Then a rapid real-time prediction algorithm based on Transit Passenger Profiling is proposed, which can predict the destination of in-board passengers. This article combines offline learning with online prediction, providing a technical foundation for real-time passenger flow prediction, monitoring and simulation, and short-term passenger behavior and demand prediction. This technology facilitates the efficient and real-time acquisition of passengers' travel destinations and demand. The last, an actual case was simulated and demonstrated feasibility and efficiency.

Keywords: travel behavior, destination prediction, public transit, passenger profiling

Procedia PDF Downloads 21

Authors: Salman Mohamadi, Seyed Mohammad Ali Tayaranian Hosseini, Hamidreza Amindavar

Abstract:

In this paper, we provide a procedure to analyze and model EEG (electroencephalogram) signal as a time series using ARIMA-GARCH to predict an epileptic attack. The heteroskedasticity of EEG signal is examined through the ARCH or GARCH, (Autore- gressive conditional heteroskedasticity, Generalized autoregressive conditional heteroskedasticity) test. The best ARIMA-GARCH model in AIC sense is utilized to measure the volatility of the EEG from epileptic canine subjects, to forecast the future values of EEG. ARIMA-only model can perform prediction, but the ARCH or GARCH model acting on the residuals of ARIMA attains a con- siderable improved forecast horizon. First, we estimate the best ARIMA model, then different orders of ARCH and GARCH modelings are surveyed to determine the best heteroskedastic model of the residuals of the mentioned ARIMA. Using the simulated conditional variance of selected ARCH or GARCH model, we suggest the procedure to predict the oncoming seizures. The results indicate that GARCH modeling determines the dynamic changes of variance well before the onset of seizure. It can be inferred that the prediction capability comes from the ability of the combined ARIMA-GARCH modeling to cover the heteroskedastic nature of EEG signal changes.

Keywords: epileptic seizure prediction , ARIMA, ARCH and GARCH modeling, heteroskedasticity, EEG

Procedia PDF Downloads 406

Authors: Olivier Delage, Hassan Bencherif, Alain Bourdier

Abstract:

Signal decomposition approaches represent an important step in time series analysis, providing useful knowledge and insight into the data and underlying dynamics characteristics while also facilitating tasks such as noise removal and feature extraction. As most of observational time series are nonlinear and nonstationary, resulting of several physical processes interaction at different time scales, experimental time series have fluctuations at all time scales and requires the development of specific signal decomposition techniques. Most commonly used techniques are data driven, enabling to obtain well-behaved signal components without making any prior-assumptions on input data. Among the most popular time series decomposition techniques, most cited in the literature, are the empirical mode decomposition and its variants, the empirical wavelet transform and singular spectrum analysis. With increasing popularity and utility of these methods in wide ranging applications, it is imperative to gain a good understanding and insight into the operation of these algorithms. In this work, we describe all of the techniques mentioned above as well as their ability to denoise signals, to capture trends, to identify components corresponding to the physical processes involved in the evolution of the observed system and deduce the dimensionality of the underlying dynamics. Results obtained with all of these methods on experimental total ozone columns and rainfall time series will be discussed and compared

Keywords: denoising, empirical mode decomposition, singular spectrum analysis, time series, underlying dynamics, wavelet analysis

Procedia PDF Downloads 118

Authors: Kamalpreet Kaur, Renu Dhir

Abstract:

Climate prediction is an essential component of climate research, which helps evaluate possible effects on economies, communities, and ecosystems. Climate prediction involves short-term weather prediction, seasonal prediction, and long-term climate change prediction. Climate prediction can use the information gathered from satellites, ground-based stations, and ocean buoys, among other sources. The paper's four architectures, such as ResNet50, VGG19, Inception-v3, and Xception, have been combined using an ensemble approach for overall performance and robustness. An ensemble of different models makes a prediction, and the majority vote determines the final prediction. The various architectures such as ResNet50, VGG19, Inception-v3, and Xception efficiently classify the dataset RSI-CB256, which contains satellite images into cloudy and non-cloudy. The generated ensembled S-E model (Sar-ensembled model) provides an accuracy of 99.25%.

Keywords: climate, satellite images, prediction, classification

Procedia PDF Downloads 75

Authors: Qiang Zhang, Chun Yuan

Abstract:

Since the PU (Prediction Unit) decision process is the most time consuming part of the emerging HEVC (High Efficient Video Coding) standardin intra and inter frame coding, this paper proposes the fast PU decision algorithm and speed up the algorithm using CUDA (Compute Unified Device Architecture). In intra frame coding, the fast PU decision algorithm uses the texture features to skip intra-frame prediction or terminal the intra-frame prediction for smaller PU size. In inter frame coding of HEVC, the fast PU decision algorithm takes use of the similarity of its own two Nx2N size PU's motion vectors and the hierarchical structure of CU (Coding Unit) partition to skip some modes of PU partition, so as to reduce the motion estimation times. The accelerate algorithm using CUDA is based on the fast PU decision algorithm which uses the GPU to make the motion search and the gradient computation could be parallel computed. The proposed algorithm achieves up to 57% time saving compared to the HM 10.0 with little rate-distortion losses (0.043dB drop and 1.82% bitrate increase on average).

Keywords: HEVC, PU decision, inter prediction, intra prediction, CUDA, parallel

Procedia PDF Downloads 399

Authors: Jia Xin Low, Keng Wah Choo

Abstract:

This paper presents an automatic normal and abnormal heart sound classification model developed based on deep learning algorithm. MITHSDB heart sounds datasets obtained from the 2016 PhysioNet/Computing in Cardiology Challenge database were used in this research with the assumption that the electrocardiograms (ECG) were recorded simultaneously with the heart sounds (phonocardiogram, PCG). The PCG time series are segmented per heart beat, and each sub-segment is converted to form a square intensity matrix, and classified using convolutional neural network (CNN) models. This approach removes the need to provide classification features for the supervised machine learning algorithm. Instead, the features are determined automatically through training, from the time series provided. The result proves that the prediction model is able to provide reasonable and comparable classification accuracy despite simple implementation. This approach can be used for real-time classification of heart sounds in Internet of Medical Things (IoMT), e.g. remote monitoring applications of PCG signal.

Keywords: convolutional neural network, discrete wavelet transform, deep learning, heart sound classification

Procedia PDF Downloads 349

Authors: Md Al Masum Bhuiyan, Maria C. Mariani, Osei K. Tweneboah

Abstract:

In the present work, we develop a technique for estimating the volatility of financial time series by using stochastic differential equation. Taking the daily closing prices from developed and emergent stock markets as the basis, we argue that the incorporation of stochastic volatility into the time-varying parameter estimation significantly improves the forecasting performance via Maximum Likelihood Estimation. While using the technique, we see the long-memory behavior of data sets and one-step-ahead-predicted log-volatility with ±2 standard errors despite the variation of the observed noise from a Normal mixture distribution, because the financial data studied is not fully Gaussian. Also, the Ornstein-Uhlenbeck process followed in this work simulates well the financial time series, which aligns our estimation algorithm with large data sets due to the fact that this algorithm has good convergence properties.

Keywords: financial time series, maximum likelihood estimation, Ornstein-Uhlenbeck type models, stochastic volatility model

Procedia PDF Downloads 242

Authors: Marina Benito-Parejo, Raul Perez-Lopez, Miguel Herraiz, Carolina Guardiola-Albert, Cesar Martinez

Abstract:

Persistency, long-term memory and randomness are intrinsic properties of time-series of earthquakes. The Rescaled Range Analysis (RS-Analysis) was introduced by Hurst in 1956 and modified by Mandelbrot and Wallis in 1964. This method represents a simple and elegant analysis which determines the range of variation of one natural property (the seismic energy released in this case) in a time interval. Despite the simplicity, there is complexity inherent in the property measured. The cumulative curve of the energy released in time is the well-known fractal geometry of a devil’s staircase. This geometry is used for determining the maximum and minimum value of the range, which is normalized by the standard deviation. The rescaled range obtained obeys a power-law with the time, and the exponent is the Hurst value. Depending on this value, time-series can be classified in long-term or short-term memory. Hence, an algorithm has been developed for compiling the RS-Analysis for time series of earthquakes by days. Completeness time distribution and locally stationarity of the time series are required. The interest of this analysis is their application for a complex seismic crisis where different earthquakes take place in clusters in a short period. Therefore, the Hurst exponent has been obtained for the seismic crisis of Alhoceima (Mediterranean Sea) of January-March, 2016, where at least five medium-sized earthquakes were triggered. According to the values obtained from the Hurst exponent for each cluster, a different mechanical origin can be detected, corroborated by the focal mechanisms calculated by the official institutions. Therefore, this type of analysis not only allows an approach to a greater understanding of a seismic series but also makes possible to discern different types of seismic origins.

Keywords: Alhoceima crisis, earthquake time series, Hurst exponent, rescaled range analysis

Procedia PDF Downloads 323

Authors: Hesham A. Elkaranshawy, Amr M. Abdelrazek, Hosam M. Ezzat

Abstract:

The system of ordinary nonlinear differential equations describing sliding velocity during impact with friction for a three-dimensional rigid-multibody system is developed. No analytical solutions have been obtained before for this highly nonlinear system. Hence, a power series solution is proposed. Since the validity of this solution is limited to its convergence zone, a suitable time step is chosen and at the end of it a new series solution is constructed. For a case study, the trajectory of the sliding velocity using the proposed method is built using 6 time steps, which coincides with a Runge-Kutta solution using 38 time steps.

Keywords: impact with friction, nonlinear ordinary differential equations, power series solutions, rough collision

Procedia PDF Downloads 488

Authors: Terigele

Abstract:

Cultural products are always shelved into different classes of a hierarchy that separates so-called highbrow and lowbrow cultures. This study illustrated that audiences might even construct a hierarchy according to the origin countries when consuming certain products. Chinese audiences now have access to TV series from all around the world thanks to the internet. TV series from different origin countries show some particular features in terms of length, theme, plots, accessibility, seriousness etc. Their audiences were therefore stereotyped because of what they watch. Based on in-depth interviews with 20 participants, this research has following findings: 1) Most popular origin countries of foreign TV series in China are Korea, the United States, the United Kingdom, Japan and European countries in a descending order. Korean TV series are most popular because they are less serious and more accessible compared to others. 2) In the hierarchy of the TV series, European TV series stand on the top followed by British and American TV series. Japanese TV series are also categorized into highbrow class. Korean TV series are at the bottom and always seen as lowbrow cultural products. 3) Most audiences consume TV series from more than one origin countries and have different needs when watching them. Participants reported that they watch European TV series because those TV series are more artistic than their counterparts and of great quality. They watch British and American TV series mainly to improve their English and to learn about the culture. They find Japanese TV series very enjoyable with a large variety of themes and impressive lines. Audiences watch Korean TV series mostly to entertain and kill time. 4) Audiences do care about cultural taste. Especially those who watch European, British and American TV series usually tend to consider audiences who watch nothing but Korean TV series to be shallow. On the other hand, Korean TV series’ audiences seem to care less about the hierarchy of the TV series. Even when they discuss the hierarchy, they tend to accept the judgments with ironies and jokes. Future studies can dig deeply into the genre and content of TV series with different origin countries and also investigate more about the psychology of audiences regarding the gender, age, education, socioeconomic status etc.

Keywords: foreign TV series, hierarchy, popularity, uses and gratification

Procedia PDF Downloads 243

Authors: Maurice Ntahobari, Levin Kuhlmann, Mario Boley, Zhinoos Razavi Hesabi

Abstract:

For machine learning based epileptic seizure prediction, it is important for the model to be implemented in small implantable or wearable devices that can be used to monitor epilepsy patients; however, current state-of-the-art methods are complex and computationally intensive. We use Shapley Additive Explanation (SHAP) to find relevant intracranial electroencephalogram (iEEG) features and improve the computational efficiency of a state-of-the-art seizure prediction method based on the extra trees classifier while maintaining prediction performance. Results for a small contest dataset and a much larger dataset with continuous recordings of up to 3 years per patient from 15 patients yield better than chance prediction performance (p < 0.004). Moreover, while the performance of the SHAP-based model is comparable to that of the benchmark, the overall training and prediction time of the model has been reduced by a factor of 1.83. It can also be noted that the feature called zero crossing value is the best EEG feature for seizure prediction. These results suggest state-of-the-art seizure prediction performance can be achieved using efficient methods based on optimal feature selection.

Keywords: machine learning, seizure prediction, extra tree classifier, SHAP, epilepsy

Procedia PDF Downloads 113

Authors: Karim Hamidi Machekposhti, Hossein Sedghi, Abdolrasoul Telvari, Hossein Babazadeh

Abstract:

Floods have huge environmental and economic impact. Therefore, flood prediction is given a lot of attention due to its importance. This study analysed the annual maximum streamflow (discharge) (AMS or AMD) of Karkheh River in Karkheh River Basin for flood predicting using ARIMA model. For this purpose, we use the Box-Jenkins approach, which contains four-stage method model identification, parameter estimation, diagnostic checking and forecasting (predicting). The main tool used in ARIMA modelling was the SAS and SPSS software. Model identification was done by visual inspection on the ACF and PACF. SAS software computed the model parameters using the ML, CLS and ULS methods. The diagnostic checking tests, AIC criterion, RACF graph and RPACF graphs, were used for selected model verification. In this study, the best ARIMA models for Annual Maximum Discharge (AMD) time series was (4,1,1) with their AIC value of 88.87. The RACF and RPACF showed residuals’ independence. To forecast AMD for 10 future years, this model showed the ability of the model to predict floods of the river under study in the Karkheh River Basin. Model accuracy was checked by comparing the predicted and observation series by using coefficient of determination (R²).

Keywords: time series modelling, stochastic processes, ARIMA model, Karkheh river

Procedia PDF Downloads 288

Authors: Heramb Somthankar, Anujoy Chakraborty

Abstract:

This paper presents a classification study on yield prediction in semiconductor production using machine learning approaches. A complicated semiconductor production process is generally monitored continuously by signals acquired from sensors and measurement sites. A monitoring system contains a variety of signals, all of which contain useful information, irrelevant information, and noise. In the case of each signal being considered a feature, "Feature Selection" is used to find the most relevant signals. The open-source UCI SECOM Dataset provides 1567 such samples, out of which 104 fail in quality assurance. Feature extraction and selection are performed on the dataset, and useful signals were considered for further study. Afterward, common machine learning algorithms were employed to predict whether the signal yields pass or fail. The most relevant algorithm is selected for prediction based on the accuracy and loss of the ML model.

Keywords: deep learning, feature extraction, feature selection, machine learning classification algorithms, semiconductor production monitoring, signal processing, time-series analysis

Procedia PDF Downloads 110

Authors: Abiodun Ayodele Sanyaolu

Abstract:

The research was carried out to investigate the time series analysis on quarterly production of fruit juice at the National Horticultural Research Institute Ibadan from 2010 to 2018. Documentary method of data collection was used, and the method of least square and moving average were used in the analysis. From the calculation and the graph, it was glaring that there was increase, decrease, and uniform movements in both the graph of the original data and the tabulated quarter values of the original data. Time series analysis was used to detect the trend in the highest number of fruit juice and it appears to be good over a period of time and the methods used to forecast are additive and multiplicative models. Since it was observed that the production of fruit juice is usually high in January of every year, it is strongly advised that National Horticultural Research Institute should make more provision for fruit juice storage outside this period of the year.

Keywords: fruit juice, least square, multiplicative models, time series

Procedia PDF Downloads 142

Authors: B. Sir, M. Podhoranyi, S. Kuchar, T. Kocyan

Abstract:

Rainfall-runoff models play important role in hydrological predictions. However, the model is only one part of the process for creation of flood prediction. The aim of this paper is to show the process of successful prediction for flood event (May 15–May 18 2014). The prediction was performed by rainfall runoff model HEC–HMS, one of the models computed within Floreon+ system. The paper briefly evaluates the results of automatic hydrologic prediction on the river Olše catchment and its gages Český Těšín and Věřňovice.

Keywords: flood, HEC-HMS, prediction, rainfall, runoff

Procedia PDF Downloads 395

Authors: Bum-Soo Kim, Jin-Uk Kim

Abstract:

In this paper, we design and implement a partial denoising boundary image matching system using indexing techniques. Converting boundary images to time-series makes it feasible to perform fast search using indexes even on a very large image database. Thus, using this converting method we develop a client-server system based on the previous partial denoising research in the GUI (graphical user interface) environment. The client first converts a query image given by a user to a time-series and sends denoising parameters and the tolerance with this time-series to the server. The server identifies similar images from the index by evaluating a range query, which is constructed using inputs given from the client, and sends the resulting images to the client. Experimental results show that our system provides much intuitive and accurate matching result.

Keywords: boundary image matching, indexing, partial denoising, time-series matching

Procedia PDF Downloads 141

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 73

Authors: Ibrahim Tiryakioglu, Mehmet Ali Ugur, Caglar Ozkaymak

Abstract:

The development of GNSS technology has led to increasingly widespread and successful applications of GNSS surveys for monitoring crustal movements. However, multi-period GPS survey solutions have not been applied in monitoring vertical surface deformation. This study uses long-term GNSS time series that are required to determine vertical deformations. In recent years, the surface deformations that are parallel and semi-parallel to Bolvadin fault have occurred in Western Anatolia. These surface deformations have continued to occur in Bolvadin settlement area that is located mostly on alluvium ground. Due to these surface deformations, a number of cracks in the buildings located in the residential areas and breaks in underground water and sewage systems have been observed. In order to determine the amount of vertical surface deformations, two continuous GNSS stations have been established in the region. The stations have been operating since 2015 and 2017, respectively. In this study, GNSS observations from the mentioned two GNSS stations were processed with GAMIT/GLOBK (GNSS Analysis Massachusetts Institute of Technology/GLOBal Kalman) program package to create a coordinate time series. With the time series analyses, the GNSS stations’ behavior models (linear, periodical, etc.), the causes of these behaviors, and mathematical models were determined. The study results from the time series analysis of these two 2 GNSS stations shows approximately 50-80 mm/yr vertical movement.

Keywords: Bolvadin fault, GAMIT, GNSS time series, surface deformations

Procedia PDF Downloads 165

Authors: Lule Basha, Eralda Gjika

Abstract:

The exchange rate is one of the most important economic variables, especially for a small, open economy such as Albania. Its effect is noticeable in one country's competitiveness, trade and current account, inflation, wages, domestic economic activity, and bank stability. This study investigates the fluctuation of Albania’s exchange rates using monthly average foreign currency, Euro (Eur) to Albanian Lek (ALL) exchange rate with a time span from January 2008 to June 2021, and the macroeconomic factors that have a significant effect on the exchange rate. Initially, the Random Forest Regression algorithm is constructed to understand the impact of economic variables on the behavior of monthly average foreign currencies exchange rates. Then the forecast of macro-economic indicators for 12 months was performed using time series models. The predicted values received are placed in the random forest model in order to obtain the average monthly forecast of the Euro to Albanian Lek (ALL) exchange rate for the period July 2021 to June 2022.

Keywords: exchange rate, random forest, time series, machine learning, prediction

Procedia PDF Downloads 104

Authors: M. Karimpour, L. Hitihamillage, N. Elkhoury, S. Moridpour, R. Hesami

Abstract:

Rail transport authorities around the world have been facing a significant challenge when predicting rail infrastructure maintenance work for a long period of time. Generally, maintenance monitoring and prediction is conducted manually. With the restrictions in economy, the rail transport authorities are in pursuit of improved modern methods, which can provide precise prediction of rail maintenance time and location. The expectation from such a method is to develop models to minimize the human error that is strongly related to manual prediction. Such models will help them in understanding how the track degradation occurs overtime under the change in different conditions (e.g. rail load, rail type, rail profile). They need a well-structured technique to identify the precise time that rail tracks fail in order to minimize the maintenance cost/time and secure the vehicles. The rail track characteristics that have been collected over the years will be used in developing rail track degradation prediction models. Since these data have been collected in large volumes and the data collection is done both electronically and manually, it is possible to have some errors. Sometimes these errors make it impossible to use them in prediction model development. This is one of the major drawbacks in rail track degradation prediction. An accurate model can play a key role in the estimation of the long-term behavior of rail tracks. Accurate models increase the track safety and decrease the cost of maintenance in long term. In this research, a short review of rail track degradation prediction models has been discussed before estimating rail track degradation for the curve sections of Melbourne tram track system using Adaptive Network-based Fuzzy Inference System (ANFIS) model.

Keywords: ANFIS, MGT, prediction modeling, rail track degradation

Procedia PDF Downloads 337

Authors: Reza Bazargan lari, Mohammad H. Fattahi

Abstract:

Natural resources management including water resources requires reliable estimations of time variant environmental parameters. Small improvements in the estimation of environmental parameters would result in grate effects on managing decisions. Noise reduction using wavelet techniques is an effective approach for pre-processing of practical data sets. Predictability enhancement of the river flow time series are assessed using fractal approaches before and after applying wavelet based pre-processing. Time series correlation and persistency, the minimum sufficient length for training the predicting model and the maximum valid length of predictions were also investigated through a fractal assessment.

Keywords: wavelet, de-noising, predictability, time series fractal analysis, valid length, ANN

Procedia PDF Downloads 372

Authors: R. Rajeswari

Abstract:

An investigation was carried out to know the SOC stock and its change over time in benchmark soils of different agroclimatic zones of Tamil Nadu. Roth.C model was used to assess SOC stock under existing and alternate cropping pattern. Soil map prepared on 1:50,000 scale from Natural Resources Information System (NRIS) employed under satellite data (IRS-1C/1D-PAN sharpened LISS-III image) was used to estimate SOC stock in different agroclimatic zones of Tamil Nadu. Fifteen benchmark soils were selected in different agroclimatic zones of Tamil Nadu based on their land use and the areal extent to assess SOC level and its change overtime. This revealed that, between eleven years of period (1997 - 2007). SOC buildup was higher in soils under horticulture system, followed by soils under rice cultivation. Among different agroclimatic zones of Tamil Nadu hilly zone have the highest SOC stock, followed by north eastern, southern, western, cauvery delta, north western, and high rainfall zone. Although organic carbon content in the soils of North eastern, southern, western, North western, Cauvery delta were less than high rainfall zone, the SOC stock was high. SOC density was higher in high rainfall and hilly zone than other agroclimatic zones of Tamil Nadu. Among low rainfall regions of Tamil Nadu cauvery delta zone recorded higher SOC density. Roth.C model was used to assess SOC stock under existing and alternate cropping pattern in viz., Periyanaickenpalayam series (western zone), Peelamedu series (southern zone), Vallam series (north eastern zone), Vannappatti series (north western zone) and Padugai series (cauvery delta zone). Padugai series recorded higher TOC, BIO, and HUM, followed by Periyanaickenpalayam series, Peelamedu series, Vallam series, and Vannappatti series. Vannappatti and Padugai series develop high TOC, BIO, and HUM under existing cropping pattern. Periyanaickenpalayam, Peelamedu, and Vallam series develop high TOC, BIO, and HUM under alternate cropping pattern. Among five selected soil series, Periyanaickenpalayam, Peelamedu, and Padugai series recorded 0.75 per cent TOC during 2025 and 2018, 2100 and 2035, 2013 and 2014 under existing and alternate cropping pattern, respectively.

Keywords: agro climatic zones, benchmark soil, land use, soil organic carbon

Procedia PDF Downloads 96

Authors: Wullapa Wongsinlatam

Abstract:

Back propagation algorithm (BP) is a widely used technique in artificial neural network and has been used as a tool for solving the time series problems, such as decreasing training time, maximizing the ability to fall into local minima, and optimizing sensitivity of the initial weights and bias. This paper proposes an improvement of a BP technique which is called IM-COH algorithm (IM-COH). By combining IM-COH algorithm with cuckoo search algorithm (CS), the result is cuckoo search improved control output hidden layer algorithm (CS-IM-COH). This new algorithm has a better ability in optimizing sensitivity of the initial weights and bias than the original BP algorithm. In this research, the algorithm of CS-IM-COH is compared with the original BP, the IM-COH, and the original BP with CS (CS-BP). Furthermore, the selected benchmarks, four time series samples, are shown in this research for illustration. The research shows that the CS-IM-COH algorithm give the best forecasting results compared with the selected samples.

Keywords: artificial neural networks, back propagation algorithm, time series, local minima problem, metaheuristic optimization

Procedia PDF Downloads 154

Authors: Aswathi Thrivikraman, S. Advaith

Abstract:

The study is aimed at designing encoders for deciphering intricacies in time series data by redescribing the dynamics operating on a lower-dimensional manifold. A semi-supervised LSTM autoencoder is devised and investigated to see if the latent representation of the time series data can better forecast the data. End-to-end training of the LSTM autoencoder, together with another LSTM network that is connected to the latent space, forces the hidden states of the encoder to represent the most meaningful latent variables relevant for forecasting. Furthermore, the study compares the predictions with those of a traditional seq2seq model.

Keywords: LSTM, autoencoder, forecasting, seq2seq model

Procedia PDF Downloads 156

Authors: Omar Alzeley, Sergey Utev

Abstract:

Minimizing a constrained multivariate function is the fundamental of Machine learning, and these algorithms are at the core of data mining and data visualization techniques. The decision function that maps input points to output points is based on the result of optimization. This optimization is the central of learning theory. One approach to complex systems where the dynamics of the system is inferred by a statistical analysis of the fluctuations in time of some associated observable is time series analysis. The purpose of this paper is a mathematical transition from the autoregressive model of classical time series to the matrix formalization of quantum theory. Firstly, we have proposed a quantum time series model (QTS). Although Hamiltonian technique becomes an established tool to detect a deterministic chaos, other approaches emerge. The quantum probabilistic technique is used to motivate the construction of our QTS model. The QTS model resembles the quantum dynamic model which was applied to financial data. Secondly, various statistical methods, including machine learning algorithms such as the Kalman filter algorithm, are applied to estimate and analyses the unknown parameters of the model. Finally, simulation techniques such as Markov chain Monte Carlo have been used to support our investigations. The proposed model has been examined by using real and simulated data. We establish the relation between quantum statistical machine and quantum time series via random matrix theory. It is interesting to note that the primary focus of the application of QTS in the field of quantum chaos was to find a model that explain chaotic behaviour. Maybe this model will reveal another insight into quantum chaos.

Keywords: machine learning, simulation techniques, quantum probability, tensor product, time series

Procedia PDF Downloads 469

Authors: N. H. Adenan, M. S. M. Noorani

Abstract:

River flow prediction is an essential to ensure proper management of water resources can be optimally distribute water to consumers. This study presents an analysis and prediction by using nonlinear prediction method involving monthly river flow data in Tanjung Tualang from 1976 to 2006. Nonlinear prediction method involves the reconstruction of phase space and local linear approximation approach. The phase space reconstruction involves the reconstruction of one-dimensional (the observed 287 months of data) in a multidimensional phase space to reveal the dynamics of the system. Revenue of phase space reconstruction is used to predict the next 72 months. A comparison of prediction performance based on correlation coefficient (CC) and root mean square error (RMSE) have been employed to compare prediction performance for nonlinear prediction method, ARIMA and SVM. Prediction performance comparisons show the prediction results using nonlinear prediction method is better than ARIMA and SVM. Therefore, the result of this study could be used to developed an efficient water management system to optimize the allocation water resources.

Keywords: river flow, nonlinear prediction method, phase space, local linear approximation

Procedia PDF Downloads 413

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 44

Authors: Md. Mahadi Hasan Sany, Mumenunnessa Keya, Sharun Khushbu, Sheikh Abujar

Abstract:

Since the beginning of China's economic reform, trade between the U.S. and China has grown rapidly, and has increased since China's accession to the World Trade Organization in 2001. The US imports more than it exports from China, reducing the trade war between China and the U.S. for the 2019 trade deficit, but in 2020, the opposite happens. In international and U.S. trade, Washington launched a full-scale trade war against China in March 2016, which occurred a catastrophic epidemic. The main goal of our study is to measure and predict trade relations between China and the U.S., before and after the arrival of the COVID epidemic. The ML model uses different data as input but has no time dimension that is present in the time series models and is only able to predict the future from previously observed data. The LSTM (a well-known Recurrent Neural Network) model is applied as the best time series model for trading forecasting. We have been able to create a sustainable forecasting system in trade between China and the US by closely monitoring a dataset published by the State Website NZ Tatauranga Aotearoa from January 1, 2015, to April 30, 2021. Throughout the survey, we provided a 180-day forecast that outlined what would happen to trade between China and the US during COVID-19. In addition, we have illustrated that the LSTM model provides outstanding outcome in time series data analysis rather than RFR and SVR (e.g., both ML models). The study looks at how the current Covid outbreak affects China-US trade. As a comparative study, RMSE transmission rate is calculated for LSTM, RFR and SVR. From our time series analysis, it can be said that the LSTM model has given very favorable thoughts in terms of China-US trade on the future export situation.

Keywords: RFR, China-U.S. trade war, SVR, LSTM, deep learning, Covid-19, export value, forecasting, time series analysis

Procedia PDF Downloads 198

Authors: Ioannis P. Panapakidis, Marios N. Moschakis

Abstract:

The Greek Energy Market is structured as a mandatory pool where the producers make their bid offers in day-ahead basis. The System Operator solves an optimization routine aiming at the minimization of the cost of produced electricity. The solution of the optimization problem leads to the calculation of the System Marginal Price (SMP). Accurate forecasts of the SMP can lead to increased profits and more efficient portfolio management from the producer`s perspective. Aim of this study is to provide a comparative analysis of various machine learning models such as artificial neural networks and neuro-fuzzy models for the prediction of the SMP of the Greek market. Machine learning algorithms are favored in predictions problems since they can capture and simulate the volatilities of complex time series.

Keywords: deregulated energy market, forecasting, machine learning, system marginal price

Procedia PDF Downloads 216

Authors: Bikis Muhammed, Sehra Sedigh Sarvestani, Ali R. Hurson, Lasanthi Gamage

Abstract:

Vehicular traffic events have overly complex spatial correlations and temporal interdependencies and are also influenced by environmental events such as weather conditions. To capture these spatial and temporal interdependencies and make more realistic vehicular traffic predictions, graph neural networks (GNN) based traffic prediction models have been extensively utilized due to their capability of capturing non-Euclidean spatial correlation very effectively. However, most of the already existing GNN-based traffic prediction models have some limitations during learning complex and dynamic spatial and temporal patterns due to the following missing factors. First, most GNN-based traffic prediction models have used static distance or sometimes haversine distance mechanisms between spatially separated traffic observations to estimate spatial correlation. Secondly, most GNN-based traffic prediction models have not incorporated environmental events that have a major impact on the normal traffic states. Finally, most of the GNN-based models did not use an attention mechanism to focus on only important traffic observations. The objective of this paper is to study and make real-time vehicular traffic predictions while incorporating the effect of weather conditions. To fill the previously mentioned gaps, our prediction model uses a real-time driving distance between sensors to build a distance matrix or spatial adjacency matrix and capture spatial correlation. In addition, our prediction model considers the effect of six types of weather conditions and has an attention mechanism in both spatial and temporal data aggregation. Our prediction model efficiently captures the spatial and temporal correlation between traffic events, and it relies on the graph attention network (GAT) and Bidirectional bidirectional long short-term memory (Bi-LSTM) plus attention layers and is called GAT-BILSTMA.

Keywords: deep learning, real time prediction, GAT, Bi-LSTM, attention

Procedia PDF Downloads 73