English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Intelligent Transportation Systems are producing tons of hardly manageabletraffic data, which motivates the use of Machine Learning (ML) for data-drivenapplications, such as Traffic Forecasting (TF). TF is gaining relevance due toits ability to mitigate traffic congestion by forecasting future trafficstates. However, TF poses one big challenge to the ML paradigm, known as theModel Selection Problem (MSP): deciding the most suitable combination of datapreprocessing techniques and ML method for traffic data collected underdifferent transportation circumstances. In this context, Automated MachineLearning (AutoML), the automation of the ML workflow from data preprocessing tomodel validation, arises as a promising strategy to deal with the MSP inproblem domains wherein expert ML knowledge is not always an available oraffordable asset, such as TF. Various AutoML frameworks have been used toapproach the MSP in TF. Most are based on online optimisation processes tosearch for the best-performing pipeline on a given dataset. This onlineoptimisation could be complemented with meta-learning to warm-start the searchphase and/or the construction of ensembles using pipelines derived from theoptimisation process. However, given the complexity of the search space and thehigh computational cost of tuning-evaluating pipelines generated, onlineoptimisation is only beneficial when there is a long time to obtain the finalmodel. Thus, we introduce AutoEn, which is a simple and efficient method forautomatically generating multi-classifier ensembles from a predefined set of MLpipelines. We compare AutoEn against Auto-WEKA and Auto-sklearn, two AutoMLmethods commonly used in TF. Experimental results demonstrate that AutoEn canlead to better or more competitive results in the general-purpose domain and inTF.

AutoEn: An AutoML method based on ensembles of predefined Machine Learning pipelines for supervised Traffic Forecasting