Taxi Booking Mobile App Order Demand Prediction Based on Short-Term Traffic Forecasting

2017 ◽  
Vol 2634 (1) ◽  
pp. 57-68 ◽  
Author(s):  
Yunxuan Li ◽  
Jian Lu ◽  
Lin Zhang ◽  
Yi Zhao
Keyword(s):  
High Frequency ◽  
Demand Forecasting ◽  
Low Frequency ◽  
Original Data ◽  
Mobile App ◽  
Support Vector ◽  
Short Term ◽  
Traffic Demand ◽  
Traffic Forecasting ◽  
Demand Prediction

The Didi Dache app is China’s biggest taxi booking mobile app and is popular in cities. Unsurprisingly, short-term traffic demand forecasting is critical to enabling Didi Dache to maximize use by drivers and ensure that riders can always find a car whenever and wherever they may need a ride. In this paper, a short-term traffic demand forecasting model, Wave SVM, is proposed. It combines the complementary advantages of Daubechies5 wavelets analysis and least squares support vector machine (LS-SVM) models while it overcomes their respective shortcomings. This method includes four stages: in the first stage, original data are preprocessed; in the second stage, these data are decomposed into high-frequency and low-frequency series by wavelet; in the third stage, the prediction stage, the LS-SVM method is applied to train and predict the corresponding high-frequency and low-frequency series; in the last stage, the diverse predicted sequences are reconstructed by wavelet. The real taxi-hailing orders data are applied to evaluate the model’s performance and practicality, and the results are encouraging. The Wave SVM model, compared with the prediction error of state-of-the-art models, not only has the best prediction performance but also appears to be the most capable of capturing the nonstationary characteristics of the short-term traffic dynamic systems.

A Study of a Railway Traffic Demand Forecasting Model Based on Influencing Factor Analysis

2010 ◽  
Author(s):  
M. S. Ma ◽  
J. Liu ◽  
C. Y. Li ◽  
Y. X. Ma
Keyword(s):  
Influencing Factor ◽  
Demand Forecasting ◽  
Real Data ◽  
Short Term ◽  
Traffic Demand ◽  
Demand Prediction ◽  
Railway Traffic ◽  
Passenger Volume ◽  
Induced Flow ◽  
Demand Forecasting Model

This paper introduced a model for the railway traffic demand prediction based on the analysis of the influencing factors over short-term passenger flow. The proposed prediction includes two parts, one for the relatively stable trend, named the basic flow forecasting, and the other for the fluctuations induced by short-term factors, named induced flow forecasting. The prediction model of the overall passenger volume utilizes the summation of the two parts. The experiments with real data demonstrate that the proposed model is effective in improving the accuracy of the prediction.

scholarly journals Small Floating Target Detection Method Based on Chaotic Long Short-Term Memory Network

2021 ◽  
Vol 9 (6) ◽  
pp. 651
Author(s):  
Yan Yan ◽  
Hongyan Xing
Keyword(s):  
High Frequency ◽  
Short Term Memory ◽  
Detection Method ◽  
Low Frequency ◽  
Sea Clutter ◽  
Short Term ◽  
Term Memory ◽  
Memory Network ◽  
Long Short Term Memory ◽  
Small Targets

In order for the detection ability of floating small targets in sea clutter to be improved, on the basis of the complete ensemble empirical mode decomposition (CEEMD) algorithm, the high-frequency parts and low-frequency parts are determined by the energy proportion of the intrinsic mode function (IMF); the high-frequency part is denoised by wavelet packet transform (WPT), whereas the denoised high-frequency IMFs and low-frequency IMFs reconstruct the pure sea clutter signal together. According to the chaotic characteristics of sea clutter, we proposed an adaptive training timesteps strategy. The training timesteps of network were determined by the width of embedded window, and the chaotic long short-term memory network detection was designed. The sea clutter signals after denoising were predicted by chaotic long short-term memory (LSTM) network, and small target signals were detected from the prediction errors. The experimental results showed that the CEEMD-WPT algorithm was consistent with the target distribution characteristics of sea clutter, and the denoising performance was improved by 33.6% on average. The proposed chaotic long- and short-term memory network, which determines the training step length according to the width of embedded window, is a new detection method that can accurately detect small targets submerged in the background of sea clutter.

scholarly journals Multi-zone prediction analysis of city-scale travel order demand

PLoS ONE ◽  
2021 ◽  
Vol 16 (3) ◽  
pp. e0248064
Author(s):  
Pengshun Li ◽  
Jiarui Chang ◽  
Yi Zhang ◽  
Yi Zhang
Keyword(s):  
Prediction Model ◽  
Clustering Algorithm ◽  
Intelligent Transportation System ◽  
Spatial Clustering ◽  
Support Vector ◽  
Nearest Neighbour ◽  
Short Term ◽  
Prediction Ability ◽  
Demand Prediction ◽  
Zone Division

Taxi order demand prediction is of tremendous importance for continuous upgrading of an intelligent transportation system to realise city-scale and personalised services. An accurate short-term taxi demand prediction model in both spatial and temporal relations can assist a city pre-allocate its resources and facilitate city-scale taxi operation management in a megacity. To address problems similar to the above, in this study, we proposed a multi-zone order demand prediction model to predict short-term taxi order demand in different zones at city-scale. A two-step methodology was developed, including order zone division and multi-zone order prediction. For the zone division step, the K-means++ spatial clustering algorithm was used, and its parameter k was estimated by the between–within proportion index. For the prediction step, six methods (backpropagation neural network, support vector regression, random forest, average fusion-based method, weighted fusion-based method, and k-nearest neighbour fusion-based method) were used for comparison. To demonstrate the performance, three multi-zone weighted accuracy indictors were proposed to evaluate the order prediction ability at city-scale. These models were implemented and validated on real-world taxi order demand data from a three-month consecutive collection in Shenzhen, China. Experiment on the city-scale taxi demand data demonstrated the superior prediction performance of the multi-zone order demand prediction model with the k-nearest neighbour fusion-based method based on the proposed accuracy indicator.

scholarly journals Can a simple, short-term memory task help to screen dyslexia?

2019 ◽  
Author(s):  
Nick Perham ◽  
Toni Howell ◽  
Andy Watt
Keyword(s):  
High Frequency ◽  
Short Term Memory ◽  
Memory Task ◽  
Compulsory Education ◽  
Low Frequency ◽  
Short Term ◽  
Term Order ◽  
Significant Difference ◽  
Support Students

AbstractFunding to support students with dyslexia in post-compulsory education is under pressure and more efficient assessments may offset some of this shortfall. We tested potential tasks for screening dyslexia: recall of adjective-noun, compared to noun-adjective, pairings (syntax) and recall of high versus low frequency letter pairings (bigrams). Students who reported themselves as dyslexic failed to show a normal syntax effect (greater recall of adjective-noun compared to noun-adjective pairings) and no significant difference in recall between the two types of bigrams whereas students who were not dyslexic showed the syntax effect and a bias towards recalling high frequency bigrams. Findings are consistent with recent explanations of dyslexia suggesting that those affected find it difficult to learn and utilise sequential long-term order information (Szmalec et al. Journal of Experimental Psychology: Learning, Memory & Cognition, 37(5) ,1270-1279, 2011). Further, ROC curve analyses revealed both tasks showed acceptable diagnostic properties as they were able to discriminate between the two groups of participants.

scholarly journals Recognition of Emotional States using Multiscale Information Analysis of High Frequency EEG Oscillations

Entropy ◽  
2019 ◽  
Vol 21 (6) ◽  
pp. 609 ◽  
Author(s):  
Gao ◽  
Cui ◽  
Wan ◽  
Gu
Keyword(s):  
Emotion Recognition ◽  
High Frequency ◽  
Low Frequency ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Emotional States ◽  
Circumplex Model ◽  
Information Analysis ◽  
Eeg Signals ◽  
High Frequency Oscillations

Exploring the manifestation of emotion in electroencephalogram (EEG) signals is helpful for improving the accuracy of emotion recognition. This paper introduced the novel features based on the multiscale information analysis (MIA) of EEG signals for distinguishing emotional states in four dimensions based on Russell's circumplex model. The algorithms were applied to extract features on the DEAP database, which included multiscale EEG complexity index in the time domain, and ensemble empirical mode decomposition enhanced energy and fuzzy entropy in the frequency domain. The support vector machine and cross validation method were applied to assess classification accuracy. The classification performance of MIA methods (accuracy = 62.01%, precision = 62.03%, recall/sensitivity = 60.51%, and specificity = 82.80%) was much higher than classical methods (accuracy = 43.98%, precision = 43.81%, recall/sensitivity = 41.86%, and specificity = 70.50%), which extracted features contain similar energy based on a discrete wavelet transform, fractal dimension, and sample entropy. In this study, we found that emotion recognition is more associated with high frequency oscillations (51–100Hz) of EEG signals rather than low frequency oscillations (0.3–49Hz), and the significance of the frontal and temporal regions are higher than other regions. Such information has predictive power and may provide more insights into analyzing the multiscale information of high frequency oscillations in EEG signals.

scholarly journals Prediction of Transient NOx Emission from Diesel Vehicles Based on Deep-Learning Differentiation Model with Double Noise Reduction

Atmosphere ◽  
2021 ◽  
Vol 12 (12) ◽  
pp. 1702
Author(s):  
Jiaqiang Li ◽  
Yang Yu ◽  
Yanyan Wang ◽  
Longqing Zhao ◽  
Chao He
Keyword(s):  
Time Series ◽  
Diesel Engines ◽  
High Frequency ◽  
Singular Spectrum Analysis ◽  
Low Frequency ◽  
Ensemble Empirical Mode Decomposition ◽  
Support Vector ◽  
Nox Emissions ◽  
Diesel Vehicles ◽  
Original Time

For diesel engines, accurate prediction of NOx (Nitrogen Oxides) emission plays an essential role in virtual NOx sensor development and engine design under situations of actual road driving. However, due to the randomness and uncertainty in the driving process of diesel vehicles, it is difficult to make predictions about NOx emissions. In order to solve this problem, this paper proposes differential models for noise reductions of NOx emissions in time series. First, according to the internal fluctuation of time series, use SSA (Singular Spectrum Analysis) to reduce the noises of the original time series; second, use ICEEMDAN (Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise) to decompose the noise-reducing data into several relatively stable subsequences; third, use the sample entropy to calculate the complexity of each subsequence, and divide the sequences into high-frequency ones and low-frequency ones; finally, use GRU (Gated Recurrent Unit) to complete the prediction of high-frequency sequences and SVR (Support Vector Regression) for the prediction of low-frequency sequences. To obtain the final models, integrate the prediction results of the subsequences. Make comparisons with five single models, SSA single-processing models, and ICEEMDAN single-processing models. The experimental results show that the proposed model can predict the instantaneous NOx emissions of diesel engines better than the single model and the model processed by SSA, and the differentiated model can effectively improve the execution speed of the model.

Fault Detection and Classification in Smart Grids Using Wavelet Analysis

2020 ◽  
Author(s):  
Muhammad Ibrahim Munir ◽  
Sajid Hussain ◽  
Ali Al-Alili ◽  
Reem Al Ameri ◽  
Ehab El-Sadaany
Keyword(s):  
Wavelet Transform ◽  
High Frequency ◽  
Smart Grids ◽  
Short Circuit ◽  
Low Frequency ◽  
Evaluation Study ◽  
Open Circuit ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Transmission Grid

Abstract One of the core features of the smart grid deemed essential for smooth grid operation is the detection and diagnosis of system failures. For a utility transmission grid system, these failures could manifest in the form of short circuit faults and open circuit faults. Due to the advent of the digital age, the traditional grid has also undergone a massive transition to digital equipment and modern sensors which are capable of generating large volumes of data. The challenge is to preprocess this data such that it can be utilized for the detection of transients and grid failures. This paper presents the incorporation of artificial intelligence techniques such as Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) to detect and comprehensively classify the most common fault transients within a reasonable range of accuracy. For gauging the effectiveness of the proposed scheme, a thorough evaluation study is conducted on a modified IEEE-39 bus system. Bus voltage and line current measurements are taken for a range of fault scenarios which result in high-frequency transient signals. These signals are analyzed using continuous wavelet transform (CWT). The measured signals are afterward preprocessed using Discrete Wavelet Transform (DWT) employing Daubechies four (Db4) mother wavelet in order to decompose the high-frequency components of the faulty signals. DWT results in a range of high and low-frequency detail and approximate coefficients, from which a range of statistical features are extracted and used as inputs for training and testing the classification algorithms. The results demonstrate that the trained models can be successfully employed to detect and classify faults on the transmission system with acceptable accuracy.

scholarly journals An improved EEMD-based hybrid approach for the short-term forecasting of hog price in China

2017 ◽  
Vol 63 (No. 3) ◽  
pp. 136-148 ◽  
Author(s):  
Xiong Tao ◽  
Li Chongguang ◽  
Bao Yukun
Keyword(s):  
High Frequency ◽  
Hybrid Approach ◽  
Low Frequency ◽  
Frequency Fluctuation ◽  
Price Forecasting ◽  
Short Term ◽  
Price Series ◽  
Short Term Forecasting ◽  
High Frequency Fluctuation ◽  
Improved Eemd

Short-term forecasting of hog price, which forms the basis for the decision making, is challenging and of great interest for hog producers and market participants. This study develops improved ensemble empirical mode decomposition (EEMD)-based hybrid approach for the short-term hog price forecasting. Specifically, the EEMD is first used to decompose the original hog price series into several intrinsic-mode functions (IMF) and one residue. The fine-to-coarse reconstruction algorithm is then applied to compose the obtained IMFs and residue into the high-frequency fluctuation, the low-frequency fluctuation, and the trend terms which can highlight new features of the hog price fluctuations. Afterwards, the extreme learning machine (ELM) is employed to model the low-frequency fluctuation, while the autoregressive integrated moving average (ARIMA) and the polynomial function are used to fit the high-frequency fluctuation and trend term, respectively, in a multistep-ahead fashion. The commonly used iterated prediction strategy is adopted for the implementation of the multistep-ahead forecasting. The monthly hog price series from January 2000 to May 2015 in China is employed to evaluate the forecasting performance of the proposed approach with the selected counterparts. The numerical results indicate that the improved EEMD-based hybrid approach is a promising alternative for the short-term hog price forecasting.  

scholarly journals Deep Learning Approach to Power Demand Forecasting in Polish Power System

Energies ◽  
2020 ◽  
Vol 13 (22) ◽  
pp. 6154
Author(s):  
Tomasz Ciechulski ◽  
Stanisław Osowski
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Power System ◽  
Radial Basis Function Network ◽  
Demand Forecasting ◽  
Support Vector ◽  
Learning Approach ◽  
Power Demand ◽  
Diagnostic Features ◽  
Demand Prediction

The paper presents a new approach to predicting the 24-h electricity power demand in the Polish Power System (PPS, or Krajowy System Elektroenergetyczny—KSE) using the deep learning approach. The prediction system uses a deep multilayer autoencoder to generate diagnostic features and an ensemble of two neural networks: multilayer perceptron and radial basis function network and support vector machine in regression model, for final 24-h forecast one-week advance. The period of the data that is the subject of the experiments is 2014–2019, which has been divided into two parts: Learning data (2014–2018), and test data (2019). The numerical experiments have shown the advantage of deep learning over classical approaches of neural networks for the problem of power demand prediction.

Sign in / Sign up

Export Citation Format

Share Document