Search results for: time series regression

Authors: Mekamı Hayet, Bounoua Nacer, Benabderrahmane Sidahmed, Taleb Ahmed

Abstract:

Pose estimation is an important task in computer vision. Though the majority of the existing solutions provide good accuracy results, they are often overly complex and computationally expensive. In this perspective, we propose the use of dimensionality reduction techniques to address the problem of facial pose estimation. Firstly, a face image is converted into one-dimensional time series using Hilbert space filling curve, then the approach converts these time series data to a symbolic representation. Furthermore, a distance matrix is calculated between symbolic series of an input learning dataset of images, to generate classifiers of frontal vs. profile face pose. The proposed method is evaluated with three public datasets. Experimental results have shown that our approach is able to achieve a correct classification rate exceeding 97% with K-NN algorithm.

Keywords: machine learning, pattern recognition, facial pose classification, time series

Procedia PDF Downloads 351

Authors: Pui Shan Wong, Kosuke Tashiro, Satoru Kuhara, Sachiyo Aburatani

Abstract:

Functional genomics and gene regulation inference has readily expanded our knowledge and understanding of gene interactions with regards to expression regulation. With the advancement of transcriptome sequencing in time-series comes the ability to study the sequential changes of the transcriptome. This method presented here works to augment existing regulation networks accumulated in literature with transcriptome data gathered from time-series experiments to construct a sequential representation of transcription factor activity. This method is applied on a time-series RNA-Seq data set from Escherichia coli as it transitions from growth to stationary phase over five hours. Investigations are conducted on the various metabolic activities in gene regulation processes by taking advantage of the correlation between regulatory gene pairs to examine their activity on a dynamic network. Especially, the changes in metabolic activity during phase transition are analyzed with focus on the pagP gene as well as other associated transcription factors. The visualization of the sequential transcriptional activity is used to describe the change in metabolic pathway activity originating from the pagP transcription factor, phoP. The results show a shift from amino acid and nucleic acid metabolism, to energy metabolism during the transition to stationary phase in E. coli.

Keywords: Escherichia coli, gene regulation, network, time-series

Procedia PDF Downloads 373

Authors: Thrivikraman Aswathi, S. Advaith

Abstract:

As there can be cases where the use of real data is somehow limited, such as when it is hard to get access to a large volume of real data, we need to go for synthetic data generation. This produces high-quality synthetic data while maintaining the statistical properties of a specific dataset. In the present work, a generative adversarial network (GAN) is trained to produce multivariate time series (MTS) data since the MTS is now being gathered more often in various real-world systems. Furthermore, the GAN-generated MTS data is fed into a transformer-based deep learning architecture that carries out the data categorization into predefined classes. Further, the model is evaluated across various distinct domains by generating corresponding MTS data.

Keywords: GAN, transformer, classification, multivariate time series

Procedia PDF Downloads 131

Authors: Himayatullah Khan

Abstract:

This study estimates supply response function of sugarcane in Pakistan from 1980-81 to 2012-13. The study uses co-integration approach and error correction mechanism. Sugarcane production, area and price series were tested for unit root using Augmented Dickey Fuller (ADF). The study found that these series were stationary at their first differenced level. Using the Augmented Engle-Granger test and Cointegrating Regression Durbin-Watson (CRDW) test, the study found that “production and price” and “area and price” were co-integrated suggesting that the two sets of time series had long-run or equilibrium relationship. The results of the error correction models for the two sets of series showed that there was disequilibrium in the short run there may be disequilibrium. The Engle-Granger residual may be thought of as the equilibrium error which can be used to tie the short-run behavior of the dependent variable to its long-run value. The Granger-Causality test results showed that log of price granger caused both the long of production and log of area whereas, the log of production and log of area Granger caused each other.

Keywords: co-integration, error correction mechanism, Granger-causality, sugarcane, supply response

Procedia PDF Downloads 436

Authors: Hang Yang, Jichao Li, Kewei Yang, Tianyang Lei

Abstract:

Multivariate time series anomaly detection is a significant research topic in the field of data mining, encompassing a wide range of applications across various industrial sectors such as traffic roads, financial logistics, and corporate production. The inherent spatial dependencies and temporal characteristics present in multivariate time series introduce challenges to the anomaly detection task. Previous studies have typically been based on the assumption that all variables belong to the same spatial hierarchy, neglecting the multi-level spatial relationships. To address this challenge, this paper proposes a multi-scale spatial and unified temporal feature fusion network, denoted as MSUT-Net, for multivariate time series anomaly detection. The proposed model employs a multi-level modeling approach, incorporating both temporal and spatial modules. The spatial module is designed to capture the spatial characteristics of multivariate time series data, utilizing an adaptive graph structure learning model to identify the multi-level spatial relationships between data variables and their attributes. The temporal module consists of a unified temporal processing module, which is tasked with capturing the temporal features of multivariate time series. This module is capable of simultaneously identifying temporal dependencies among different variables. Extensive testing on multiple publicly available datasets confirms that MSUT-Net achieves superior performance on the majority of datasets. Our method is able to model and accurately detect systems data with multi-level spatial relationships from a spatial-temporal perspective, providing a novel perspective for anomaly detection analysis.

Keywords: data mining, industrial system, multivariate time series, anomaly detection

Procedia PDF Downloads 17

Authors: Theophilus Adjirackor, Frederic Sam, Irene Opoku-Ntim, David Okoh Kpeglo, Prince K. Gyekye, Frank K. Quashie, Kofi Ofori

Abstract:

Indoor radon concentrations were collected monthly over a period of one year in 10 different levels in an underground goldmine, and the data was analyzed using a four-moving average time series to determine the relationship between the depths of the underground mine and the indoor radon concentration. The detectors were installed in batches within four quarters. The measurements were carried out using LR115 solid-state nuclear track detectors. Statistical models are applied in the prediction and analysis of the radon concentration at various depths. The time series model predicted a positive relationship between the depth of the underground mine and the indoor radon concentration. Thus, elevated radon concentrations are expected at deeper levels of the underground mine, but the relationship was insignificant at the 5% level of significance with a negative adjusted R2 (R2 = – 0.021) due to an appropriate engineering and adequate ventilation rate in the underground mine.

Keywords: LR115, radon concentration, rime series, underground goldmine

Procedia PDF Downloads 47

Authors: Cecile Edel, Sandrine Giscard D'Estaing, Elsa Labrune, Jacqueline Lornage, Mehdi Benchaib

Abstract:

Introduction: The use of incubator equipped with time-lapse technology in Artificial Reproductive Technology (ART) allows a continuous surveillance. With morphocinetic parameters, algorithms are available to predict the potential outcome of an embryo. However, the different proposed time-lapse algorithms do not take account the missing data, and then some embryos could not be classified. The aim of this work is to construct a predictive model even in the case of missing data. Materials and methods: Patients: A retrospective study was performed, in biology laboratory of reproduction at the hospital ‘Femme Mère Enfant’ (Lyon, France) between 1 May 2013 and 30 April 2015. Embryos (n= 557) obtained from couples (n=108) were cultured in a time-lapse incubator (Embryoscope®, Vitrolife, Goteborg, Sweden). Time-lapse incubator: The morphocinetic parameters obtained during the three first days of embryo life were used to build the predictive model. Predictive model: A quadratic regression was performed between the number of cells and time. N = a. T² + b. T + c. N: number of cells at T time (T in hours). The regression coefficients were calculated with Excel software (Microsoft, Redmond, WA, USA), a program with Visual Basic for Application (VBA) (Microsoft) was written for this purpose. The quadratic equation was used to find a value that allows to predict the blastocyst formation: the synthetize value. The area under the curve (AUC) obtained from the ROC curve was used to appreciate the performance of the regression coefficients and the synthetize value. A cut-off value has been calculated for each regression coefficient and for the synthetize value to obtain two groups where the difference of blastocyst formation rate according to the cut-off values was maximal. The data were analyzed with SPSS (IBM, Il, Chicago, USA). Results: Among the 557 embryos, 79.7% had reached the blastocyst stage. The synthetize value corresponds to the value calculated with time value equal to 99, the highest AUC was then obtained. The AUC for regression coefficient ‘a’ was 0.648 (p < 0.001), 0.363 (p < 0.001) for the regression coefficient ‘b’, 0.633 (p < 0.001) for the regression coefficient ‘c’, and 0.659 (p < 0.001) for the synthetize value. The results are presented as follow: blastocyst formation rate under cut-off value versus blastocyst rate formation above cut-off value. For the regression coefficient ‘a’ the optimum cut-off value was -1.14.10-3 (61.3% versus 84.3%, p < 0.001), 0.26 for the regression coefficient ‘b’ (83.9% versus 63.1%, p < 0.001), -4.4 for the regression coefficient ‘c’ (62.2% versus 83.1%, p < 0.001) and 8.89 for the synthetize value (58.6% versus 85.0%, p < 0.001). Conclusion: This quadratic regression allows to predict the outcome of an embryo even in case of missing data. Three regression coefficients and a synthetize value could represent the identity card of an embryo. ‘a’ regression coefficient represents the acceleration of cells division, ‘b’ regression coefficient represents the speed of cell division. We could hypothesize that ‘c’ regression coefficient could represent the intrinsic potential of an embryo. This intrinsic potential could be dependent from oocyte originating the embryo. These hypotheses should be confirmed by studies analyzing relationship between regression coefficients and ART parameters.

Keywords: ART procedure, blastocyst formation, time-lapse incubator, quadratic model

Procedia PDF Downloads 308

Authors: Sooyoung Cho, In-Gun Kim, Hyun-Seok Hong, Dong-Woo Kang, Ju Lee

Abstract:

Series hybrid UAV(Unmanned aerial vehicle) that is proposed in this paper performs VTOL(Vertical take-off and landing) using the battery and generator, and it applies the series hybrid system with combination of the small engine and generator when cruising flight. This system can be described as the next-generation system that can dramatically increase the UAV flight times. Also, UAV systems require a large energy at the time of VTOL to be conducted for a short time. Therefore, this paper designs PMSG(Permanent Magnet Synchronous Generator) having a high specific power considering VTOL through the FEA.

Keywords: PMSG, VTOL, UAV, high specific power density

Procedia PDF Downloads 521

Authors: Andres F. Ramirez, Carlos F. Valencia

Abstract:

The increasing interest in renewable energies strategies application and the path for diminishing the use of carbon related energy sources have encouraged the development of novel strategies for integration of solar energy into the electricity network. A correct inclusion of the fluctuating energy output of a photovoltaic (PV) energy system into an electric grid requires improvements in the forecasting and simulation methodologies for solar energy potential, and the understanding not only of the mean value of the series but the associated underlying stochastic process. We present a methodology for synthetic generation of solar irradiance (shortwave flux) and air temperature bivariate time series based on copula functions to represent the cross-dependence and temporal structure of the data. We explore the advantages of using this nonlinear time series method over traditional approaches that use a transformation of the data to normal distributions as an intermediate step. The use of copulas gives flexibility to represent the serial variability of the real data on the simulation and allows having more control on the desired properties of the data. We use discrete zero mass density distributions to assess the nature of solar irradiance, alongside vector generalized linear models for the bivariate time series time dependent distributions. We found that the copula autoregressive methodology used, including the zero mass characteristics of the solar irradiance time series, generates a significant improvement over state of the art strategies. These results will help to better understand the fluctuating nature of solar energy forecasting, the underlying stochastic process, and quantify the potential of a photovoltaic (PV) energy generating system integration into a country electricity network. Experimental analysis and real data application substantiate the usage and convenience of the proposed methodology to forecast solar irradiance time series and solar energy across northern hemisphere, southern hemisphere, and equatorial zones.

Keywords: copula autoregressive, solar irradiance forecasting, solar energy forecasting, time series generation

Procedia PDF Downloads 323

Authors: Jimeng Shi, Mahek Jain, Giri Narasimhan

Abstract:

Time Series Forecasting (TSF) is used to predict the target variables at a future time point based on the learning from previous time points. To keep the problem tractable, learning methods use data from a fixed-length window in the past as an explicit input. In this paper, we study how the performance of predictive models changes as a function of different look-back window sizes and different amounts of time to predict the future. We also consider the performance of the recent attention-based Transformer models, which have had good success in the image processing and natural language processing domains. In all, we compare four different deep learning methods (RNN, LSTM, GRU, and Transformer) along with a baseline method. The dataset (hourly) we used is the Beijing Air Quality Dataset from the UCI website, which includes a multivariate time series of many factors measured on an hourly basis for a period of 5 years (2010-14). For each model, we also report on the relationship between the performance and the look-back window sizes and the number of predicted time points into the future. Our experiments suggest that Transformer models have the best performance with the lowest Mean Average Errors (MAE = 14.599, 23.273) and Root Mean Square Errors (RSME = 23.573, 38.131) for most of our single-step and multi-steps predictions. The best size for the look-back window to predict 1 hour into the future appears to be one day, while 2 or 4 days perform the best to predict 3 hours into the future.

Keywords: air quality prediction, deep learning algorithms, time series forecasting, look-back window

Procedia PDF Downloads 156

Authors: Autcha Araveeporn

Abstract:

This paper presents a study about a nonparametric regression model consisting of a smoothing spline method and a penalized spline regression method. We also compare the techniques used for estimation and prediction of nonparametric regression model. We tried both methods with crude oil prices in dollars per barrel and the Stock Exchange of Thailand (SET) index. According to the results, it is concluded that smoothing spline method performs better than that of penalized spline regression method.

Keywords: nonparametric regression model, penalized spline regression method, smoothing spline method, Stock Exchange of Thailand (SET)

Procedia PDF Downloads 440

Authors: Ramesh K. Sunkaria, Puneeta Marwaha

Abstract:

Nonlinear methods of heart rate variability (HRV) analysis are becoming more popular. It has been observed that complexity measures quantify the regularity and uncertainty of cardiovascular RR-interval time series. In the present work, SampEn has been evaluated in healthy Normal Sinus Rhythm (NSR) male and female subjects for different data lengths and tolerance level r. It is demonstrated that SampEn is small for higher values of tolerance r. Also SampEn value of healthy female group is higher than that of healthy male group for short data length and with increase in data length both groups overlap each other and it is difficult to distinguish them. The SampEn gives inaccurate results by assigning higher value to female group, because male subject have more complex HRV pattern than that of female subjects. Therefore, this traditional algorithm exhibits higher complexity for healthy female subjects than for healthy male subjects, which is misleading observation. This may be due to the fact that SampEn do not account for multiple time scales inherent in the physiologic time series and the hidden spatial and temporal fluctuations remains unexplored.

Keywords: heart rate variability, normal sinus rhythm group, RR interval time series, sample entropy

Procedia PDF Downloads 282

Authors: Victoria Siriaki Jorry, I. S. Mbalawata, Hayong Shin

Abstract:

The main objective in a change detection problem is to develop algorithms for efficient detection of gradual and/or abrupt changes in the parameter distribution of a process or time series data. In this paper, we present a modified cumulative (MCUSUM) algorithm to detect the start and end of a time-varying linear drift in mean value of a time series data based on likelihood ratio test procedure. The design, implementation and performance of the proposed algorithm for a linear drift detection is evaluated and compared to the existing CUSUM algorithm using different performance measures. An approach to accurately approximate the threshold of the MCUSUM is also provided. Performance of the MCUSUM for gradual change-point detection is compared to that of standard cumulative sum (CUSUM) control chart designed for abrupt shift detection using Monte Carlo Simulations. In terms of the expected time for detection, the MCUSUM procedure is found to have a better performance than a standard CUSUM chart for detection of the gradual change in mean. The algorithm is then applied and tested to a randomly generated time series data with a gradual linear trend in mean to demonstrate its usefulness.

Keywords: average run length, CUSUM control chart, gradual change detection, likelihood ratio test

Procedia PDF Downloads 299

Authors: Siriporn Meenanan

Abstract:

The objective of this study is to investigate the attitude of Thai students towards the Thai series "Hormones the Series Season 2". This study was conducted in the quantitative research, and the questionnaires were used to collect data from 400 people of the sample group. Descriptive statistics were used in data analysis. The findings reveal that most participants have positive comments regarding the series. They strongly agreed that the series reflects on the way of life and problems of teenagers in Thailand. Hence, the participants believe that if adults have a chance to watch the series, they will have the better understanding of the teenagers. In addition, the participants also agreed that the contents of the play are appropriate and satisfiable as the contents of “Hormones the Series Season 2” will raise awareness among the teens and use it as a guide to prevent problems that might happen during their teenage life.

Keywords: content analysis, attitude, Thai series, hormones the Series

Procedia PDF Downloads 230

Authors: Ashneel Ajay Singh, Naoki Suzuki, Kazumi Sakuramoto,

Abstract:

The North Pacific albacore (Thunnus alalunga) is a temperate tuna species distributed in the North Pacific which is of significant economic importance to the Pacific Island Nations and Territories. Despite its importance, the stock dynamics and ecological characteristics of albacore still, have gaps in knowledge. The stock-recruitment relationship of the North Pacific stock of albacore tuna was investigated for different density-dependent effects and a regime shift in the stock characteristics in response to changes in environmental and climatic conditions. Linear regression analysis for recruit per spawning biomass (RPS) and recruitment (R) against the female spawning stock biomass (SSB) were significant for the presence of different density-dependent effects and positive for a regime shift in the stock time series. Application of Deming regression to RPS against SSB with the assumption for the presence of observation and process errors in both the dependent and independent variables confirmed the results of simple regression. However, R against SSB results disagreed given variance level of < 3 and agreed with linear regression results given the assumption of variance ≥ 3. Assuming the presence of different density-dependent effects in the albacore tuna time series, environmental and climatic condition variables were compared with R, RPS, and SSB. The significant relationship of R, RPS and SSB were determined with the sea surface temperature (SST), Pacific Decadal Oscillation (PDO) and multivariate El Niño Southern Oscillation (ENSO) with SST being the principal variable exhibiting significantly similar trend with R and RPS. Recruitment is significantly influenced by the dynamics of the SSB as well as environmental conditions which demonstrates that the stock-recruitment relationship is multidimensional. Further investigation of the North Pacific albacore tuna age-class and structure is necessary for further support the results presented here. It is important for fishery managers and decision makers to be vigilant of regime shifts in environmental conditions relating to albacore tuna as it may possibly cause regime shifts in the albacore R and RPS which should be taken into account to effectively and sustainability formulate harvesting plans and management of the species in the North Pacific oceanic region.

Keywords: Albacore tuna, Thunnus alalunga, recruitment, spawning stock biomass, recruits per spawning biomass, sea surface temperature, pacific decadal oscillation, El Niño southern oscillation, density-dependent effects, regime shift

Procedia PDF Downloads 307

Authors: Kehinde Peter Oyeduntan, Kayode Oshinubi

Abstract:

The west African ports have been experiencing large inflow and outflow of containerized cargo in the last decades, and this has created a quest amongst the countries to attain the status of hub port for the sub-region. This study analyzed the relationship between the container throughput and Gross Domestic Products (GDP) of nine west African countries, using Simple Linear Regression (SLR), Polynomial Regression Model (PRM) and Support Vector Machines (SVM) with a time series of 20 years. The results showed that there exists a high correlation between the GDP and container throughput. The model also predicted the container throughput in west Africa for the next 20 years. The findings and recommendations presented in this research will guide policy makers and help improve the management of container ports and terminals in west Africa, thereby boosting the economy.

Keywords: container, ports, terminals, throughput

Procedia PDF Downloads 215

Authors: Vishal Das, Tianyi Mao, Zhicheng Geng, Carmen Flores, Diego Pelloso, Fang Wang

Abstract:

Accurate sell-in and sell-out forecasting is a ubiquitous problem in the retail industry. It is an important element of any demand planning activity. As a global food and beverage company, Nestlé has hundreds of products in each geographical location that they operate in. Each product has its sell-in and sell-out time series data, which are forecasted on a weekly and monthly scale for demand and financial planning. To address this challenge, Nestlé Chilein collaboration with Amazon Machine Learning Solutions Labhas developed their in-house solution of using machine learning models for forecasting. Similar products are combined together such that there is one model for each product category. In this way, the models learn from a larger set of data, and there are fewer models to maintain. The solution is scalable to all product categories and is developed to be flexible enough to include any new product or eliminate any existing product in a product category based on requirements. We show how we can use the machine learning development environment on Amazon Web Services (AWS) to explore a set of forecasting models and create business intelligence dashboards that can be used with the existing demand planning tools in Nestlé. We explored recent deep learning networks (DNN), which show promising results for a variety of time series forecasting problems. Specifically, we used a DeepAR autoregressive model that can group similar time series together and provide robust predictions. To further enhance the accuracy of the predictions and include domain-specific knowledge, we designed an ensemble approach using DeepAR and XGBoost regression model. As part of the ensemble approach, we interlinked the sell-out and sell-in information to ensure that a future sell-out influences the current sell-in predictions. Our approach outperforms the benchmark statistical models by more than 50%. The machine learning (ML) pipeline implemented in the cloud is currently being extended for other product categories and is getting adopted by other geomarkets.

Keywords: sell-in and sell-out forecasting, demand planning, DeepAR, retail, ensemble machine learning, time-series

Procedia PDF Downloads 276

Authors: Tamas Madl

Abstract:

Sudden cardiac death (SCD) constitutes a large proportion of cardiovascular mortalities, provides little advance warning, and the risk is difficult to recognize based on ubiquitous, low cost medical equipment such as the standard, 12-lead, ten second ECG. Autonomic abnormalities have been shown to be strongly predictive of SCD risk; yet current methods are not trivially applicable to the brevity and low temporal and electrical resolution of standard ECGs. Here, we build horizontal visibility graph representations of very short inter-beat interval time series, and perform unsuper- vised representation learning in order to convert these variable size objects into fixed-length vectors preserving similarity rela- tions. We show that such representations facilitate classification into healthy vs. at-risk patients on two different datasets, the Mul- tiparameter Intelligent Monitoring in Intensive Care II and the PhysioNet Sudden Cardiac Death Holter Database. Our results suggest that graph representation learning of heartbeat interval time series facilitates robust classification even in sequences as short as ten seconds.

Keywords: sudden cardiac death, heart rate variability, ECG analysis, time series classification

Procedia PDF Downloads 236

Authors: Ahmad Moussavi

Abstract:

In this paper, we are concerned with the Jacobson semisimple skew inverse Laurent series rings R((x−1; α, δ)) and the skew Laurent power series rings R[[x, x−1; α]], where R is an associative ring equipped with an automorphism α and an α-derivation δ. Examples to illustrate and delimit the theory are provided.

Keywords: skew polynomial rings, Laurent series, skew inverse Laurent series rings

Procedia PDF Downloads 166

Authors: Messis A., Adjebli A., Ayeche R., Talbi M., Tighilet K., Louardiane M.

Abstract:

Cancers are the second cause of death worldwide. Prevalence and incidence of cancers is getting increased by aging and population growth. This study aims to predict and modeling the evolution of breast, Colorectal, Lung, Bladder and Prostate cancers over the period of 2014-2019. In this study, data were analyzed using time series analysis with double exponential smoothing method to forecast the future pattern. To describe and fit the appropriate models, Minitab statistical software version 17 was used. Between 2014 and 2019, the overall trend in the raw number of new cancer cases registered has been increasing over time; the change in observations over time has been increasing. Our forecast model is validated since we have good prediction for the period 2020 and data not available for 2021 and 2022. Time series analysis showed that the double exponential smoothing is an efficient tool to model the future data on the raw number of new cancer cases.

Keywords: cancer, time series, prediction, double exponential smoothing

Procedia PDF Downloads 89

Authors: Kehinde Peter Oyeduntan, Kayode Oshinubi

Abstract:

Nigeria is referred as the ‘Giant of Africa’ due to high population, land mass and large economy. However, it still trails far behind many smaller economies in the continent in terms of maritime operations. As we have seen that the maritime industry is the spark plug for national growth, because it houses the most crucial infrastructure that generates wealth for a nation, it is worrisome that a nation with six seaports lag in maritime activities. In this research, we have studied how the Gross Domestic Product (GDP) of the maritime transport influences the Nigerian economy. To do this, we applied Simple Linear Regression (SLR), Support Vector Machine (SVM), Polynomial Regression Model (PRM), Generalized Additive Model (GAM) and Generalized Linear Mixed Model (GLMM) to model the relationship between the nation’s Total GDP (TGDP) and the Maritime Transport GDP (MGDP) using a time series data of 20 years. The result showed that the MGDP is statistically significant to the Nigerian economy. Amongst the statistical tool applied, the PRM of order 4 describes the relationship better when compared to other methods. The recommendations presented in this study will guide policy makers and help improve the economy of Nigeria in terms of its GDP.

Keywords: maritime transport, economy, GDP, regression, port

Procedia PDF Downloads 155

Authors: Asif Zaman, Moinul Islam Zaber, Amin Ahsan Ali

Abstract:

In recent years of industrialization, the South Asian countries are being affected by air pollution due to a severe increase in fine particulate matter 2.5 (PM₂.₅). Among them, Bangladesh is one of the most polluting countries. In this paper, statistical analyses were conducted on the time series of PM₂.₅ from various districts in Bangladesh, mostly around Dhaka city. Research has been conducted on the dynamic interactions and relationships between PM₂.₅ concentrations in different zones. The study is conducted toward understanding the characteristics of PM₂.₅, such as spatial-temporal characterization, correlation of other contributors behind air pollution such as human activities, driving factors and environmental casualties. Clustering on the data gave an insight on the districts groups based on their AQI frequency as representative districts. Seasonality analysis on hourly and monthly frequency found higher concentration of fine particles in nighttime and winter season, respectively. Cross correlation analysis discovered a phenomenon of correlations among cities based on time-lagged series of air particle readings and visualization framework is developed for observing interaction in PM₂.₅ concentrations between cities. Significant time-lagged correlations were discovered between the PM₂.₅ time series in different city groups throughout the country by cross correlation analysis. Additionally, seasonal heatmaps depict that the pooled series correlations are less significant in warmer months, and among cities of greater geographic distance as well as time lag magnitude and direction of the best shifted correlated particulate matter time series among districts change seasonally. The geographic map visualization demonstrates spatial behaviour of air pollution among districts around Dhaka city and the significant effect of wind direction as the vital actor on correlated shifted time series. The visualization framework has multipurpose usage from gathering insight of general and seasonal air quality of Bangladesh to determining the pathway of regional transportation of air pollution.

Keywords: air quality, particles, cross correlation, seasonality

Procedia PDF Downloads 105

Authors: Vikrant Gupta, Amrit Goswami

Abstract:

The fixed income market forms the basis of the modern financial market. All other assets in financial markets derive their value from the bond market. Owing to its over-the-counter nature, corporate bonds have relatively less data publicly available and thus is researched upon far less compared to Equities. Bond price prediction is a complex financial time series forecasting problem and is considered very crucial in the domain of finance. The bond prices are highly volatile and full of noise which makes it very difficult for traditional statistical time-series models to capture the complexity in series patterns which leads to inefficient forecasts. To overcome the inefficiencies of statistical models, various machine learning techniques were initially used in the literature for more accurate forecasting of time-series. However, simple machine learning methods such as linear regression, support vectors, random forests fail to provide efficient results when tested on highly complex sequences such as stock prices and bond prices. hence to capture these intricate sequence patterns, various deep learning-based methodologies have been discussed in the literature. In this study, a recurrent neural network-based deep learning model using long short term networks for prediction of corporate bond prices has been discussed. Long Short Term networks (LSTM) have been widely used in the literature for various sequence learning tasks in various domains such as machine translation, speech recognition, etc. In recent years, various studies have discussed the effectiveness of LSTMs in forecasting complex time-series sequences and have shown promising results when compared to other methodologies. LSTMs are a special kind of recurrent neural networks which are capable of learning long term dependencies due to its memory function which traditional neural networks fail to capture. In this study, a simple LSTM, Stacked LSTM and a Masked LSTM based model has been discussed with respect to varying input sequences (three days, seven days and 14 days). In order to facilitate faster learning and to gradually decompose the complexity of bond price sequence, an Empirical Mode Decomposition (EMD) has been used, which has resulted in accuracy improvement of the standalone LSTM model. With a variety of Technical Indicators and EMD decomposed time series, Masked LSTM outperformed the other two counterparts in terms of prediction accuracy. To benchmark the proposed model, the results have been compared with traditional time series models (ARIMA), shallow neural networks and above discussed three different LSTM models. In summary, our results show that the use of LSTM models provide more accurate results and should be explored more within the asset management industry.

Keywords: bond prices, long short-term memory, time series forecasting, empirical mode decomposition

Procedia PDF Downloads 137

Authors: Sandeep Kaur, Jenny Vuong, Marcel Julliard, Sean O'Donoghue

Abstract:

Time-series data are useful for modelling as they can enable model-evaluation. However, when reconstructing models from phosphoproteomic data, often non-exact methods are utilised, as the knowledge regarding the network structure, such as, which kinases and phosphatases lead to the observed phosphorylation state, is incomplete. Thus, such reactions are often hypothesised, which gives rise to uncertainty. Here, we propose a framework, implemented via a web-based tool (as an extension to Minardo), which given time-series phosphoproteomic datasets, can generate κ models. The incompleteness and uncertainty in the generated model and reactions are clearly presented to the user via the visual method. Furthermore, we demonstrate, via a toy EGF signalling model, the use of algorithmic verification to verify κ models. Manually formulated requirements were evaluated with regards to the model, leading to the highlighting of the nodes causing unsatisfiability (i.e. error causing nodes). We aim to integrate such methods into our web-based tool and demonstrate how the identified erroneous nodes can be presented to the user via the visual method. Thus, in this research we present a framework, to enable a user to explore phosphorylation proteomic time-series data in the context of models. The observer can visualise which reactions in the model are highly uncertain, and which nodes cause incorrect simulation outputs. A tool such as this enables an end-user to determine the empirical analysis to perform, to reduce uncertainty in the presented model - thus enabling a better understanding of the underlying system.

Keywords: κ-models, model verification, time-series phosphoproteomic datasets, uncertainty and error visualisation

Procedia PDF Downloads 259

Authors: Charline David, Alexandre Blondin Massé, Arnaud Zinflou

Abstract:

In 2016, Clements, Hurn, and Li proposed a multiple equation time series approach for the short-term load forecasting, reporting an average mean absolute percentage error (MAPE) of 1.36% on an 11-years dataset for the Queensland region in Australia. We present an adaptation of their model to the electrical power load consumption for the whole Quebec province in Canada. More precisely, we take into account two additional meteorological variables — cloudiness and wind speed — on top of temperature, as well as the use of multiple meteorological measurements taken at different locations on the territory. We also consider other minor improvements. Our final model shows an average MAPE score of 1:79% over an 8-years dataset.

Keywords: short-term load forecasting, special days, time series, multiple equations, parallelization, clustering

Procedia PDF Downloads 104

Authors: Hervice Roméo Fogno Fotsoa, Germaine Djuidje Kenmoe, Claude Vidal Aloyem Kazé

Abstract:

One of the main applications of machine learning is the prediction of time series. But a more accurate prediction requires a more optimal model of machine learning. Several optimization techniques have been developed, but without considering the input variables disposition of the system. Thus, this work aims to present a new machine learning architecture optimization technique based on their optimal input variables disposition. The validations are done on the prediction of wind time series, using data collected in Cameroon. The number of possible dispositions with four input variables is determined, i.e., twenty-four. Each of the dispositions is used to perform the prediction, with the main criteria being the training and prediction performances. The results obtained from a static architecture and a dynamic architecture of neural networks have shown that these performances are a function of the input variable's disposition, and this is in a different way from the architectures. This analysis revealed that it is necessary to take into account the input variable's disposition for the development of a more optimal neural network model. Thus, a new neural network training algorithm is proposed by introducing the search for the optimal input variables disposition in the traditional back-propagation algorithm. The results of the application of this new optimization approach on the two single neural network architectures are compared with the previously obtained results step by step. Moreover, this proposed approach is validated in a collaborative optimization method with a single objective optimization technique, i.e., genetic algorithm back-propagation neural networks. From these comparisons, it is concluded that each proposed model outperforms its traditional model in terms of training and prediction performance of time series. Thus the proposed optimization approach can be useful in improving the accuracy of time series forecasts. This proves that the proposed optimization approach can be useful in improving the accuracy of time series prediction based on machine learning.

Keywords: input variable disposition, machine learning, optimization, performance, time series prediction

Procedia PDF Downloads 111

Authors: M. Ouassaf, S. Belaid

Abstract:

A series of quinoline derivatives with antimalarial activity were subjected to two-dimensional quantitative structure-activity relationship (2D-QSAR) studies. Three models were implemented using multiple regression linear MLR, a regression partial least squares (PLS), nonlinear regression (MNLR), to see which descriptors are closely related to the activity biologic. We relied on a principal component analysis (PCA). Based on our results, a comparison of the quality of, MLR, PLS, and MNLR models shows that the MNLR (R = 0.914 and R² = 0.835, RCV= 0.853) models have substantially better predictive capability because the MNLR approach gives better results than MLR (R = 0.835 and R² = 0,752, RCV=0.601)), PLS (R = 0.742 and R² = 0.552, RCV=0.550) The model of MNLR gave statistically significant results and showed good stability to data variation in leave-one-out cross-validation. The obtained results suggested that our proposed model MNLR may be useful to predict the biological activity of derivatives of quinoline.

Keywords: antimalarial, quinoline, QSAR, PCA, MLR , MNLR, MLR

Procedia PDF Downloads 157

Authors: Ramin Vafadary, Maryam Khanbaghi

Abstract:

Forecasting electricity load is important for various purposes like planning, operation, and control. Forecasts can save operating and maintenance costs, increase the reliability of power supply and delivery systems, and correct decisions for future development. This paper compares various time series methods to forecast 24 hours ahead of electricity load. The methods considered are the Holt-Winters smoothing, SARIMA Modeling, LSTM Network, Fbprophet, and Tensorflow probability. The performance of each method is evaluated by using the forecasting accuracy criteria, namely, the mean absolute error and root mean square error. The National Renewable Energy Laboratory (NREL) residential energy consumption data is used to train the models. The results of this study show that the SARIMA model is superior to the others for 24 hours ahead forecasts. Furthermore, a Bagging technique is used to make the predictions more robust. The obtained results show that by Bagging multiple time-series forecasts, we can improve the robustness of the models for 24 hours ahead of electricity load forecasting.

Keywords: bagging, Fbprophet, Holt-Winters, LSTM, load forecast, SARIMA, TensorFlow probability, time series

Procedia PDF Downloads 96

Authors: Hisyam Hidayatullah

Abstract:

This world events consciously or not realize everything has a pattern, until the events of the universe according to the Big Bang theory of the solar system which makes so regular in the rotation. The author would like to create a results curve area between the quadratic function y=kx2 and line y=ka2 using GeoGebra application version 4.2. This paper can provide a series that is no less interesting with Fourier series, so that will add new material about the series can be calculated with sigma notation. In addition, the ranks of the unique natural numbers of extensive changes in established areas. Finally, this paper provides analytical and geometric proof of the vast area in between the lines and curves that give the area is formed by y=ka2 dan kurva y=kx2, x-axis, line x=√a and x=-√a make a series of numbers for k=1 and a ∈ original numbers. ∑_(i=0)^n=(4n√n)/3=0+4/3+(8√2)/3+4√3+⋯+(4n√n)/3. The author calls the series “H154M”.

Keywords: sequence, series, sigma notation, application GeoGebra

Procedia PDF Downloads 377

Authors: Arindam Chaudhuri

Abstract:

The research presents epsilon- hierarchical fuzzy twin support vector regression (epsilon-HFTSVR) based on epsilon-fuzzy twin support vector regression (epsilon-FTSVR) and epsilon-twin support vector regression (epsilon-TSVR). Epsilon-FTSVR is achieved by incorporating trapezoidal fuzzy numbers to epsilon-TSVR which takes care of uncertainty existing in forecasting problems. Epsilon-FTSVR determines a pair of epsilon-insensitive proximal functions by solving two related quadratic programming problems. The structural risk minimization principle is implemented by introducing regularization term in primal problems of epsilon-FTSVR. This yields dual stable positive definite problems which improves regression performance. Epsilon-FTSVR is then reformulated as epsilon-HFTSVR consisting of a set of hierarchical layers each containing epsilon-FTSVR. Experimental results on both synthetic and real datasets reveal that epsilon-HFTSVR has remarkable generalization performance with minimum training time.

Keywords: regression, epsilon-TSVR, epsilon-FTSVR, epsilon-HFTSVR

Procedia PDF Downloads 376