Su búsqueda

En autores o colaboradores
  • Artificial neural networks have a wide use in the prediction and classification of different variables, but their application in the area of educational psychology is still relatively rare. The aim of this study was to examine the accuracy of artificial neural networks in predicting students’ general giftedness. The participants were 221 fourth grade students from one Croatian elementary school. The input variables for artificial neural networks were teachers’ and peers’ nominations, school grades, earlier school readiness assessment and parents’ education. The output variable was the result on the Standard Progressive Matrices (Raven, 1994), according to which students were classified as gifted or non-gifted. We tested two artificial neural networks’ algorithms: multilayer perception and radial basis function. Within each algorithm, a number of different types of activation functions were tested. 80% of the sample was used for training the network and the remaining 20% to test the network. For a criterion according to which students were classified as gifted if their result on the Standard Progressive Matrices was in the 95 th centile or above, the best model was obtained by the hyperbolic tangent multilayer perception, which had a high accuracy of 100% of correctly classified non-gifted students and 75% correctly classified gifted students in the test sample. When the criterion was the 90 th centile or above, the best model was also obtained by the hyperbolic tangent multilayer perception, but the accuracy was lower: 94.7% in the classification of non-gifted students and 66.7% in the classification of gifted students. The study has shown artificial neural networks’ potential in this area, which should be further explored. (PsycInfo Database Record (c) 2024 APA, all rights reserved) (Source: journal abstract) Resumen en idioma originalUmjetne neuronske mreže imaju široku upotrebu u predikciji i klasifikaciji različitih varijabli, no njihova primjena u području psihologije obrazovanja je još uvijek relativno rijetka. Cilj ovog istraživanja bio je ispitati točnost umjetnih neuronskih mreža u predviđanju nadarenosti učenika. U ispitivanju je sudjelovao 221 učenik 4. razreda jedne hrvatske osnovne škole. Kao ulazne varijable za umjetne neuronske mreže korištene su nominacije učitelja i drugih učenika, ocjene, ranija procjena spremnosti učenika za školu i obrazovanje roditelja. Kao izlazna varijabla korišten je rezultat učenika na Standardnim progresivnim matricama (Raven, 1994), prema kojem su učenici svrstani u darovite ili nedarovite. Testirali smo dva algoritma umjetnih neuronskih mreža: mrežu s radijalno zasnovanom funkcijom i višeslojni perceptron. Unutar svakog algoritma, testirano je više aktivacijskih funkcija. 80% uzorka korišteno je za uvježbavanje mreža, 20% za testiranje njihove uspješnosti. Za kriterij prema kojem su u darovite učenike svrstani oni koji postižu rezultat na Standardnim progresivnim matricama u 95. centilu ili više, najuspješnijom se pokazala mreža višeslojnog perceptrona s funkcijom tangens hiperbolni, koja je na testnom uzorku postigla visoku točnost od 100% u klasifikaciji nedarovitih učenika i 75% u klasifikaciji darovitih učenika. Kada je kriterij bio rezultat u 90. centilu ili više, najuspješnija je bila također mreža višeslojnog perceptrona s funkcijom tangens hiperbolni, no točnost je bila niža: 94,7% u klasifikaciji nedarovitih učenika i 66,7% u klasifikaciji darovitih učenika. Istraživanje je pokazalo potencijal umjetnih neuronskih mreža u ovom području, koji treba dalje istražiti. (PsycInfo Database Record (c) 2024 APA, all rights reserved) (Source: journal abstract)

Última actualización de la base de datos: 4/10/26, 6:08 (CEST)

Explorar

Tipo de recurso

Año de publicación