The Concept of an Adaptive Trainer and Assessing Its Effectiveness in a Mathematical Application
Author(s) -
L.S. Kuravsky,
D.A. Pominov,
G.A. Yuryev,
N.E. Yuryeva,
M.A. Safronova,
Y.D. Kulanin,
S.N. Antipova
Publication year - 2021
Publication title -
modelling and data analysis
Language(s) - English
Resource type - Journals
eISSN - 2311-9454
pISSN - 2219-3758
DOI - 10.17759/mda.2021110401
Subject(s) - trainer , computer science , computerized adaptive testing , task (project management) , test (biology) , selection (genetic algorithm) , adaptation (eye) , adaptive learning , artificial intelligence , curriculum , machine learning , mathematics , psychology , psychometrics , engineering , statistics , pedagogy , paleontology , systems engineering , neuroscience , biology , programming language
Presented is a mathematical model of the self-learning adaptive trainer intended for adaptive learning and providing task selection. The approach in question is an alternative to the adaptive technologies based on the Item Response Theory. Possibility to take into account temporal dynamics of solution ability as well as smaller number of tasks that must be performed by a subject to provide the given results are among the features of the methods in use. To assess the effectiveness of the adaptive trainer concept under consideration, its web-implementation intended for training school students to solve mathematical tasks covered by the school curriculum was employed. The analysis performed revealed both high efficiency of the adaptive trainer (the mean test rating has increased 1.54 times owing to its use) and proven statistically significant influences of the adaptive training factor on the observed mathematical test results.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom