INFLUENCE OF DATA NORMALIZATION ON THE EFFECTIVENESS OF NEURAL NETWORKS APPLIED TO CLASSIFICATION OF PAVEMENT CONDITIONS – CASE STUDY

Abstrakt

In recent years automatic classification employing machine learning seems to be in high demand for tele-informatic-based solutions. An example of such solutions are intelligent transportation systems (ITS), in which various factors are taken into account. The subject of the study presented is the impact of data pre-processing and normalization on the accuracy and training effectiveness of artificial neural networks in the case of pavement condition classification. First, audio parametrization process is shortly described and then the most commonly used methods of data normalization are recalled. Examples of analyses are shown, along with conclusions on application of neural networks to pavement moisture condition classification. A neural network based on the Java Neuroph library was designed. Training time and the network evaluation efficiency of the data without and with normalization performed were shown and analyzed. As it turns out, the Z-score normalization is the most accurate, and also the fastest one for the dataset gathered.

Informacje szczegółowe

Kategoria: Publikacja w czasopiśmie
Typ: artykuły w czasopismach recenzowanych i innych wydawnictwach ciągłych
Opublikowano w: Zeszyty Naukowe Wydziału ETI Politechniki Gdańskiej. Technologie Informacyjne nr 23, wydanie 1, strony 5 - 12,
ISSN: 1732-1166
Język: angielski
Rok wydania: 2018
Opis bibliograficzny: Marciniuk K., Kostek B.: INFLUENCE OF DATA NORMALIZATION ON THE EFFECTIVENESS OF NEURAL NETWORKS APPLIED TO CLASSIFICATION OF PAVEMENT CONDITIONS – CASE STUDY// Zeszyty Naukowe Wydziału ETI Politechniki Gdańskiej. Technologie Informacyjne. -Vol. 23., iss. 1 (2018), s.5-12
wyświetlono 4 razy
Meta Tagi