Forecasting the carsharing service demand using uni and multivariable models

AbstractCarsharing is ana lternative to urban mobility that has been widely adopted recently. This service presents three main business models: two of these models base their services on stations while the remainder, the free-floating service, is free of fixed stations. Despite the notable advantages of carsharing, this service is prone to several problems, such as fleet imbalance due to the variance of the daily demand in large urban centers. Forecasting the demand for the service is a key task to deal with this issue. In this sense, in this work, we analyze the use of well-known techniques to forecast a carsharing service demand. More in deep, we evaluate the use of the Long Short-Term Memory (LSTM) and Prophet techniques to predict the demand of three real carsharing services. Moreover, we also evaluate seven state-of-the-art forecasting models on a given free-floating carsharing service, highlighting the potentials of each technique. In addition to historical carsharing service data, we have also used climatic series to enhance the forecasting. Indeed, the results of our analysis have shown that the addition of meteorological data improved the models’ performance. In this case, the mean absolute error of LSTM may fall by half, when using the climate data. When considering the free-floating carsharing service, and prediction for the short-term (i.e., 12 hours), the boosting algorithms (e.g. XGBoost, Catboost, and LightGBM) present superior performance, with less than 20% of mean absolute error when compared to the next best-ranked model (Prophet). On the other hand, Prophet performed better for predictions conducted on long-term periods.

Download Full-text

Forecasting the Carsharing Service Demand Using Uni and Multivariable Models

10.21203/rs.3.rs-99113/v1 ◽

2020 ◽

Author(s):

Victor Aquiles Alencar ◽

Lucas Ribeiro Pessamilio ◽

Felipe Rooke da Silva ◽

Heder Soares Bernardino ◽

Alex Borges Vieira

Keyword(s):

Short Term Memory ◽

Meteorological Data ◽

Absolute Error ◽

Climatic Data ◽

Short Term ◽

Car Sharing ◽

Urban Centers ◽

Service Demand ◽

Boosting Algorithms

Abstract Car-sharing is an alternative to urban mobility that has been widely adopted. However, this approach is prone to several problems, such as fleet imbalance, due to the variance of the daily demand in large urban centers. In this work, we apply two time series techniques, namely, Long Short-Term Memory (LSTM) and Prophet, to infer the demand for three real car-sharing services. We also apply several state-of-the-art models on free-floating data in order to get a better understanding of what works best for this type of data. In addition to historical data, we also use climatic attributes in LSTM applications. As a result, the addition of meteorological data improved the model’s performance, especially on Evo: an average Mean Absolute Error (MAE) of approximately 61.13 travels was obtained with the demand data on Evo, while MAE equals 32.72 travels was observed when adding the climatic data, the other datasets also improved but none other improved this much. For the free-floating data test, we got the Boosting Algorithms (XGBoost, Catboost, and LightGBM) got the best performance short term, the worst one has an improvement of around 22% of MAE over the next best-ranked (Prophet). Meanwhile in the long term Prophet got the best MAE result, around 22.5% better than the second-best (LSTM).

Download Full-text

Recurrent Neural Network-Based Hourly Prediction of Photovoltaic Power Output Using Meteorological Information

Energies ◽

10.3390/en12020215 ◽

2019 ◽

Vol 12 (2) ◽

pp. 215 ◽

Cited By ~ 31

Author(s):

Donghun Lee ◽

Kwanho Kim

Keyword(s):

Neural Network ◽

Power Output ◽

Short Term Memory ◽

Mean Absolute Error ◽

Absolute Error ◽

Energy Market ◽

Seasonal Patterns ◽

Short Term ◽

Artificial Neural Network Ann ◽

Power Outputs

Recently, the prediction of photovoltaic (PV) power has become of paramount importance to improve the expected revenue of PV operators and the effective operations of PV facility systems. Additionally, the precise PV power output prediction in an hourly manner enables more sophisticated strategies for PV operators and markets as the electricity price in a renewable energy market is continuously changing. However, the hourly prediction of PV power outputs is considered as a challenging problem due to the dynamic natures of meteorological information not only in a day but also across days. Therefore, in this paper, we suggest three PV power output prediction methods such as artificial neural network (ANN)-, deep neural network (DNN)-, and long and short term memory (LSTM)-based models that are capable to understand the hidden relationships between meteorological information and actual PV power outputs. In particular, the proposed LSTM based model is designed to capture both hourly patterns in a day and seasonal patterns across days. We conducted the experiments by using a real-world dataset. The experimental results show that the proposed ANN based model fails to yield satisfactory results, and the proposed LSTM based model successfully better performs more than 50% compared to the conventional statistical models in terms of mean absolute error.

Download Full-text

Previsão de casos de dengue através de Machine Learning e Deep Learning: uma revisão sistemática

Research Society and Development ◽

10.33448/rsd-v10i11.19347 ◽

2021 ◽

Vol 10 (11) ◽

pp. e33101119347

Author(s):

Ewethon Dyego de Araujo Batista ◽

Wellington Candeia de Araújo ◽

Romeryto Vieira Lira ◽

Laryssa Izabel de Araujo Batista

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Short Term Memory ◽

Mean Absolute Error ◽

Absolute Error ◽

Support Vector ◽

Short Term ◽

Term Memory ◽

Long Short Term Memory ◽

De Medicina

Introdução: a dengue é uma arbovirose causada pelo vírus DENV e transmitida para o homem através do mosquito Aedes aegypti. Atualmente, não existe uma vacina eficaz para combater todas as sorologias do vírus. Diante disso, o combate à doença se volta para medidas preventivas contra a proliferação do mosquito. Os pesquisadores estão utilizando Machine Learning (ML) e Deep Learning (DL) como ferramentas para prever casos de dengue e ajudar os governantes nesse combate. Objetivo: identificar quais técnicas e abordagens de ML e de DL estão sendo utilizadas na previsão de dengue. Métodos: revisão sistemática realizada nas bases das áreas de Medicina e de Computação com intuito de responder as perguntas de pesquisa: é possível realizar previsões de casos de dengue através de técnicas de ML e de DL, quais técnicas são utilizadas, onde os estudos estão sendo realizados, como e quais dados estão sendo utilizados? Resultados: após realizar as buscas, aplicar os critérios de inclusão, exclusão e leitura aprofundada, 14 artigos foram aprovados. As técnicas Random Forest (RF), Support Vector Regression (SVR), e Long Short-Term Memory (LSTM) estão presentes em 85% dos trabalhos. Em relação aos dados, na maioria, foram utilizados 10 anos de dados históricos da doença e informações climáticas. Por fim, a técnica Root Mean Absolute Error (RMSE) foi a preferida para mensurar o erro. Conclusão: a revisão evidenciou a viabilidade da utilização de técnicas de ML e de DL para a previsão de casos de dengue, com baixa taxa de erro e validada através de técnicas estatísticas.

Download Full-text

Task Scheduling in Fog Node within the Tactical Cloud

Defence Science Journal ◽

10.14429/dsj.72.17039 ◽

2022 ◽

Vol 72 (1) ◽

pp. 49-55

Author(s):

Biji Nair ◽

S. Mary Saira Bhanu

Keyword(s):

Task Scheduling ◽

Linear Complexity ◽

Short Term Memory ◽

Mean Absolute Error ◽

Fog Computing ◽

Absolute Error ◽

Memory Model ◽

Short Term ◽

Forecasting Models ◽

Long Short Term Memory

Fog computing architecture competent to support the mission-oriented network-centric warfare provides the framework for a tactical cloud in this work. The tactical cloud becomes situation-aware of the war from the information relayed by fog nodes (FNs) on the battlefield. This work aims to sustain the network of FNs by maintaining the operational efficiency of the FNs on the battlefield at the tactical edge. The proposed solution monitors and predicts the likely overloading of an FN using the long short-term memory model through a buddy FN at the fog server (FS). This paper also proposes randomised task scheduling (RTS) algorithm to avert the likely overloading of an FN by pre-empting tasks from the FN and scheduling them to another FN. The experimental results demonstrate that RTS with linear complexity has a schedulability measure 8% - 26% higher than that of other base scheduling algorithms. The results show that the LSTM model has low mean absolute error compared to other time-series forecasting models.

Download Full-text

Time Series Forecasting of Agricultural Products’ Sales Volumes Based on Seasonal Long Short-Term Memory

Applied Sciences ◽

10.3390/app10228169 ◽

2020 ◽

Vol 10 (22) ◽

pp. 8169

Author(s):

Tae-Woong Yoo ◽

Il-Seok Oh

Keyword(s):

Time Series ◽

Performance Metrics ◽

Short Term Memory ◽

Mean Absolute Error ◽

Absolute Error ◽

Agricultural Products ◽

Series Data ◽

Short Term ◽

Term Memory ◽

Long Short Term Memory

In this paper, we propose seasonal long short-term memory (SLSTM), which is a method for predicting the sales of agricultural products, to stabilize supply and demand. The SLSTM model is trained using the seasonality attributes of week, month, and quarter as additional inputs to historical time-series data. The seasonality attributes are entered into the SLSTM network model individually or in combination. The performance of the proposed SLSTM model was compared with those of auto_arima, Prophet, and a standard LSTM in terms of three performance metrics (mean absolute error (MAE), root mean squared error (RMSE), and normalization mean absolute error (NMAE)). The experimental results show that the error rate of the proposed SLSTM model is significantly lower than those of other classical methods.

Download Full-text

Comparative Study of Univariate and Multivariate Long Short-Term Memory for Very Short-Term Forecasting of Global Horizontal Irradiance

Symmetry ◽

10.3390/sym13081544 ◽

2021 ◽

Vol 13 (8) ◽

pp. 1544

Author(s):

Ashis Kumar Mandal ◽

Rikta Sen ◽

Saptarsi Goswami ◽

Basabi Chakraborty

Keyword(s):

Time Series ◽

Wind Direction ◽

Power Plants ◽

Prediction Models ◽

Short Term Memory ◽

Meteorological Data ◽

Absolute Error ◽

Short Term ◽

Term Memory ◽

Long Short Term Memory

Accurate global horizontal irradiance (GHI) forecasting is crucial for efficient management and forecasting of the output power of photovoltaic power plants. However, developing a reliable GHI forecasting model is challenging because GHI varies over time, and its variation is affected by changes in weather patterns. Recently, the long short-term memory (LSTM) deep learning network has become a powerful tool for modeling complex time series problems. This work aims to develop and compare univariate and several multivariate LSTM models that can predict GHI in Guntur, India on a very short-term basis. To build the multivariate time series models, we considered all possible combinations of temperature, humidity, and wind direction variables along with GHI as inputs and developed seven multivariate models, while in the univariate model, we considered only GHI variability. We collected the meteorological data for Guntur from 1 January 2016 to 31 December 2016 and built 12 datasets, each containing variability of GHI, temperature, humidity, and wind direction of a month. We then constructed the models, each of which measures up to 2 h ahead of forecasting of GHI. Finally, to measure the symmetry among the models, we evaluated the performances of the prediction models using root mean square error (RMSE) and mean absolute error (MAE). The results indicate that, compared to the univariate method, each multivariate LSTM performs better in the very short-term GHI prediction task. Moreover, among the multivariate LSTM models, the model that incorporates the temperature variable with GHI as input has outweighed others, achieving average RMSE values 0.74 W/m2–1.5 W/m2.

Download Full-text

PREDICTING DAILY PM2.5 USING WEIGHTED LONG SHORT-TERM MEMORY NEURAL NETWORK MODEL

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xliii-b3-2020-1451-2020 ◽

2020 ◽

Vol XLIII-B3-2020 ◽

pp. 1451-1455

Author(s):

H. Fan ◽

M. Yang ◽

F. Xiao ◽

K. Zhao

Keyword(s):

Neural Network ◽

Air Pollution ◽

Short Term Memory ◽

Meteorological Data ◽

Series Data ◽

Wind Condition ◽

Short Term ◽

Term Memory ◽

Long Short Term Memory ◽

Pm2.5 Concentration

Abstract. Over the past few decades, air pollution has caused serious damage on public health, thus making accurate predictions of PM2.5 crucial. Due to the transportation of air pollutants among areas, the PM2.5 concentration is strongly spatiotemporal correlated. However, the distribution of air pollution monitoring sites is not even, making the spatiotemporal correlation between the central site and surrounding sites varies with different density of sites, and this was neglected by most existing methods. To tackle this problem, this study proposed a weighted long short-term memory neural network extended model (WLSTME), which addressed the issue that how to consider the effect of the density of sites and wind condition on the spatiotemporal correlation of air pollution concentration. First, several the nearest surrounding sites were chosen as the neighbour sites to the central station, and their distance as well as their air pollution concentration and wind condition were input to multi-layer perception (MLP) to generate weighted historical PM2.5 time series data. Second, historical PM2.5 concentration of the central site and weighted PM2.5 series data of neighbour sites were input into LSTM to address spatiotemporal dependency simultaneously and extract spatiotemporal features. Finally, another MLP was utilized to integrate spatiotemporal features extracted above with the meteorological data of central site to generate the forecasts future PM_2.5 concentration of the central site. Daily PM_2.5 concentration and meteorological data on Beijing–Tianjin–Hebei from 2015 to 2017 were collected to train models and evaluate the performance. Experimental results with 3 other methods showed that the proposed WLSTME model has the lowest RMSE (40.67) and MAE (26.10) and the highest p (0.59). This finding confirms that WLSTME can significantly improve the PM2.5 prediction accuracy.

Download Full-text

Hearing the Beat

Music Perception An Interdisciplinary Journal ◽

10.1525/mp.2016.34.1.56 ◽

2016 ◽

Vol 34 (1) ◽

pp. 56-70 ◽

Cited By ~ 9

Author(s):

Kathleen M. Einarson ◽

Laurel J. Trainor

Keyword(s):

Cognitive Skills ◽

Short Term Memory ◽

Superior Performance ◽

Auditory Stimuli ◽

Perceptual Sensitivity ◽

Short Term ◽

Term Memory ◽

Phase Errors ◽

The Relationship ◽

Better Than

Adults can extract the underlying beat from music, and entrain their movements with that beat. Although infants and children are poor at synchronizing their movements to auditory stimuli, recent findings suggest they are perceptually sensitive to the beat. We examined five-year-old children’s perceptual sensitivity to musical beat alignment (adapting the adult task of Iversen & Patel, 2008). We also examined whether sensitivity is affected by metric complexity, and whether perceptual sensitivity correlates with cognitive skills. On each trial of the complex Beat Alignment Test (cBAT) children were presented with two successive videos of puppets drumming to music with simple or complex meter. One puppet’s drumming was synchronized with the beat of the music while the other had either incorrect tempo or incorrect phase, and children were asked to select the better drummer. In two experiments, five-year-olds were able to detect beat misalignments in simple meter music significantly better than beat misalignments in complex meter music for both phase errors and tempo errors, with performance for complex meter music at chance levels. Although cBAT performance correlated with short-term memory in Experiment One, the relationship held for both simple and complex meter, so cannot explain the superior performance for culturally typical meters.

Download Full-text

Developing Deep Survival Model for Remaining Useful Life Estimation Based on Convolutional and Long Short-Term Memory Neural Networks

Wireless Communications and Mobile Computing ◽

10.1155/2020/8814658 ◽

2020 ◽

Vol 2020 ◽

pp. 1-12

Author(s):

Chia-Hua Chu ◽

Chia-Jung Lee ◽

Hsiang-Yuan Yeh

Keyword(s):

Neural Networks ◽

Short Term Memory ◽

Survival Model ◽

Remaining Useful Life ◽

Superior Performance ◽

Short Term ◽

Term Memory ◽

Useful Life ◽

Long Short Term Memory ◽

Prediction Approach

The application of mechanical equipment in manufacturing is becoming more and more complicated with technology development and adoption. In order to keep the high reliability and stability of the production line, reducing the downtime to repair and the frequency of routine maintenance is necessary. Since machine and components’ degradations are inevitable, accurately estimating the remaining useful life of them is crucial. We propose an integrated deep learning approach with convolutional neural networks and long short-term memory networks to learn the latent features and estimate remaining useful life value with deep survival model based on the discrete Weibull distribution. We conduct the turbofan engine degradation simulation dataset from Commercial Modular Aero-Propulsion System Simulation dataset provided by NASA to validate our approach. The improved results have proven that our proposed model can capture the degradation trend of a fault and has superior performance under complex conditions compared with existing state-of-the-art methods. Our study provides an efficient feature extraction scheme and offers a promising prediction approach to make better maintenance strategies.

Download Full-text

Deep Belief Neural Networks and Bidirectional Long-Short Term Memory Hybrid for Speech Recognition

Archives of Acoustics ◽

10.1515/aoa-2015-0021 ◽

2015 ◽

Vol 40 (2) ◽

pp. 191-195 ◽

Cited By ~ 10

Author(s):

Łukasz Brocki ◽

Krzysztof Marasek

Keyword(s):

Neural Networks ◽

Speech Recognition ◽

Short Term Memory ◽

Recognition Accuracy ◽

Superior Performance ◽

Acoustic Model ◽

Short Term ◽

Term Memory ◽

Internal Memory ◽

Long Short Term Memory

Abstract This paper describes a Deep Belief Neural Network (DBNN) and Bidirectional Long-Short Term Memory (LSTM) hybrid used as an acoustic model for Speech Recognition. It was demonstrated by many independent researchers that DBNNs exhibit superior performance to other known machine learning frameworks in terms of speech recognition accuracy. Their superiority comes from the fact that these are deep learning networks. However, a trained DBNN is simply a feed-forward network with no internal memory, unlike Recurrent Neural Networks (RNNs) which are Turing complete and do posses internal memory, thus allowing them to make use of longer context. In this paper, an experiment is performed to make a hybrid of a DBNN with an advanced bidirectional RNN used to process its output. Results show that the use of the new DBNN-BLSTM hybrid as the acoustic model for the Large Vocabulary Continuous Speech Recognition (LVCSR) increases word recognition accuracy. However, the new model has many parameters and in some cases it may suffer performance issues in real-time applications.

Download Full-text