Amazing technological breakthrough possible @S-Logix

Office Address

  • #5, First Floor, 4th Street Dr. Subbarayan Nagar Kodambakkam, Chennai-600 024 Landmark : Samiyar Madam
  • +91- 81240 01111

Social List

Deep Learning for Traffic Time Series Data Analysis

Deep Learning for Traffic Time Series Data Analysis

Trending PhD Thesis on Deep Learning for Traffic Time Series Data Analysis

Research Area:  Machine Learning


  Time series data in traffic has been playing an important role in intelligent transportation systems (ITS) research and applications. However,because of the sparse,imbalanced, stochastic and highly non-linear natures of traffic time series data, typical parametric methods or machine learning methods are unable to learn the representations of data well.This work aims to develop deep learning methods to gain novel and valuable knowledge from traffic time series analysis for ITS.Specically, deep learning-based methods are developed for three topics, namely taxi destination prediction, anomalous traffic patterns detection,and urban traffic ow prediction.The first method is to predict taxi destination using trajectory data.Accurate and timely destination prediction of taxis is of great importance for location-based service applications.Over the last few decades, popularization of vehicle navigation systems has brought the era of big data for the taxi industry.Existing destination prediction approaches are mainly based on various Markov chain models or trip matching ideas,which require geographical information and may encounter the problem of data sparsity.Other machine learning prediction models are still unsatisfying to provide favourable results.In this work,firstly, we propose a novel and efficient data embedding method for time-related features preprocessing.The key idea behind this is to embed the data into a two-dimensional space before features learning.Secondly,we propose a novel ensemble learning approach for destination prediction.

Name of the Researcher:  Xiaocai Zhang

Name of the Supervisor(s):  Prof. Jinyan Li

Year of Completion:  2020

University:  University of Technology Sydney

Thesis Link:   Home Page Url