Short-Term Traffic Forecasting Using Deep Learning
Iren Valova,Natacha Gueorguieva,Sandeep Smudidonga +2 more
- 01 Aug 2021
TL;DR: This research proposes Convolutional Long Short Term Memory (CLSTM) which incorporates spatial and temporary information into the forecasting process and is compared with various deep learning architectures of Gated Recurrent Unit (GRU), Long Shortterm Memory (LSTM), and baseline methods such as Vector Autoregression (VAR) and historical average.
read more
Abstract: Forecasting is one of the key applications of machine learning. The task of forecasting becomes complex when there are spatiotemporal dependencies in the data generating process. Prediction of congestion ahead of time is a very important aspect of transportation system management. Traffic congestion on a road network has a temporal component due to daily and weekly variation in human travel, and also a spatial component due to the connected nature of the road network and traffic flow. Furthermore, the spatial component of traffic congestion is certainly not Euclidean due to directionality of road network, which is not an undirected graph. Congestion prediction falls into the realm of time series data analysis methods which can be mapped onto a neural network-based methods for sequence prediction. In this research we propose Convolutional Long Short Term Memory (CLSTM) which incorporates spatial and temporary information into the forecasting process. To validate the efficiency of the proposed method, the performance is compared with various deep learning architectures of Gated Recurrent Unit (GRU), Long Short Term Memory (LSTM), and baseline methods such as Vector Autoregression (VAR) and historical average. Experiments include the above topologies with varying parameters as number of units per layer, number of layers, optimizers, learning rate and lengths of sequence input. Prediction comparison is demonstrated with tables and graphical representations.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
N-BEATS Deep Learning Architecture for Agricultural Commodity Price Forecasting
G. H. Harish Nayak,Md Wasi Alam,G. Avinash,Kehar Singh,Mrinmoy Ray,Rajeev Ranjan Kumar +5 more
3
Hybrid Models for Predicting Stock Market Performance
Iren Valova,N. Gueorguieva,Thakkar Aayushi,Pulluri Nikitha,Hassan Mohamed +4 more
A Machine Learning Approach to Traffic Congestion Hotspot Identification and Prediction
Manoj K. Jha,Rishav Jaiswal,D. Sai Kiran Varma,Shalini Rankavat,Anil Kumar Bachu +4 more
- 01 Jan 2024
Performance measurement system
L.M.H. van Zuilichem
- 01 Jan 2011
Abstract: Problem: Although performance measurement systems (PMSs) in firms to a large extent regard motivation of the employees, motivation theory is rarely used as framework to study the use and implications of PMS. The starting point of our identified problem is the narrow theoretical framework used in prior research of formulation of PMS in firms, when only relying on agency theory. We consider the present conceptions used as a foundation for the results in theoretical and empirical research of PMS are a barrier for more efficient governance control in firms. Therefore we ask if agency theory alone solid enough as a framework to capture the complex behavioural aspects of PMS? Or does an incorporation of motivation theory extend the understanding of the use of PMS in firms? The answer has got important practical implications for all firms characterised by the separation of ownership and control. This brings us to the question: When evaluating and rewarding employees, which aspects are crucial to consider when formulating PMS in order to maximise the value of the firm? Purpose: The purpose of this thesis is to extend the understanding of the use of PMSs in banks. Contrary to prior research in this field, our purpose is to extend the conceptual framework by incorporating motivation theory extensively when analysing the use of PMS. In addition the results of this study are intended to give practical indications from a firm value maximizing perspective of which aspects that are crucial when formulating a PMS. Method: A qualitative method has been chosen to collect empirical data to our study. We have conducted interviews with our respondents working in the two banks. Moreover from our theoretical and empirical the conclusions are drawn in line with the approach of abduction. Conclusion: From what we have seen in our study there is no absolute answer, since there are many contingencies which affects how a PMS should be adapted to an organisation. However, due to four numbers of reasons we have come to the conclusion that the use of one atomistic measurement system, together with a profit-sharing system, is to prefer where all employees are rewarded equally.
References
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
99K
Learning representations by back-propagating errors
TL;DR: Back-propagation repeatedly adjusts the weights of the connections in the network so as to minimize a measure of the difference between the actual output vector of the net and the desired output vector, which helps to represent important features of the task domain.
30.1K
Learning Phrase Representations using RNN Encoder--Decoder for Statistical Machine Translation
Kyunghyun Cho,Bart van Merriënboer,Caglar Gulcehre,Dzmitry Bahdanau,Fethi Bougares,Holger Schwenk,Yoshua Bengio,Yoshua Bengio,Yoshua Bengio +8 more
- 01 Jan 2014
TL;DR: In this paper, the encoder and decoder of the RNN Encoder-Decoder model are jointly trained to maximize the conditional probability of a target sequence given a source sequence.
Distribution of the Estimators for Autoregressive Time Series with a Unit Root
David A. Dickey,Wayne A. Fuller +1 more
TL;DR: In this article, the limit distributions of the estimator of p and of the regression t test are derived under the assumption that p = ± 1, where p is a fixed constant and t is a sequence of independent normal random variables.
26.7K