Modelling of the Effect of Concentration and Temperature on Reaction Time By Means of Artificial Neural Networks


Sarıçayır H. , Unal M., Serhat Sadı B., Üce M.

Proceedings of World Conference on Educational Multimedia, Hypermedia and Telecommunications, Agualva-Cacem, Portekiz, 11 - 12 Ağustos 2011, ss.3202-3207

  • Basıldığı Şehir: Agualva-Cacem
  • Basıldığı Ülke: Portekiz
  • Sayfa Sayıları: ss.3202-3207

Özet

MODELLING OF THE Effect of Concentration and Temperature on Reaction Time BY MEANS OF ARTIFICIAL NEURAL NETWORKS

 

Hakan Saricayir(1) , Muhammet Unal(2), Serhad Sadi Barutcuoglu(1), Musa Uce(1)

 

(1) Marmara University Ataturk Faculty of Education Chemistry Education Department

(2) Marmara University Faculty of Technical Education Electronics and Computer Education Department

 

 

 

ABSTRACT

 

 

An artificial neural network model was developed to predict reaction time. The concentrations of the KIO3 solution and temperature were employed as input to the network; the output of the network was reaction time. The data used this study were collected through our experiments. The multilayer feed-forward networks were trained and tested by 77 and 11 sets of input-output patterns using a back-propagation algorithm. A two-layered network with 20 neurons in the hidden layer gave optimal results. The model gave good predictions of reaction time correlation coefficient (R= 0.94296). With this modeling, reaction time can be accurately predicted in various concentration and temperature values.

MODELLING OF THE Effect of Concentration and Temperature on Reaction Time BY MEANS OF ARTIFICIAL NEURAL NETWORKS

 

Hakan Saricayir(1) , Muhammet Unal(2), Serhad Sadi Barutcuoglu(1), Musa Uce(1)

 

(1) Marmara University Ataturk Faculty of Education Chemistry Education Department

(2) Marmara University Faculty of Technical Education Electronics and Computer Education Department

 

 

 

ABSTRACT

 

 

An artificial neural network model was developed to predict reaction time. The concentrations of the KIO3 solution and temperature were employed as input to the network; the output of the network was reaction time. The data used this study were collected through our experiments. The multilayer feed-forward networks were trained and tested by 77 and 11 sets of input-output patterns using a back-propagation algorithm. A two-layered network with 20 neurons in the hidden layer gave optimal results. The model gave good predictions of reaction time correlation coefficient (R= 0.94296). With this modeling, reaction time can be accurately predicted in various concentration and temperature values.