Related Experiment Video
Updated: Jul 29, 2026
09:34
A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
Published on: September 25, 2021
Classification of user expertise level by neural networks
International Journal of Neural Systems
|April 1, 1997
Summary
This study introduces a neural network model to automatically assess user expertise in text editing. The neurofuzzy system accurately classifies users into five levels, outperforming other expert systems.
Area of Science:
- Artificial Intelligence
- Human-Computer Interaction
- Machine Learning
Background:
- Low-level user modeling is crucial for adaptive systems.
- Expertise assessment in text editing tasks has been challenging.
- Previous methods lacked accuracy and efficiency.
Purpose of the Study:
- To develop an automated neural network approach for low-level user modeling.
- To classify users into five distinct expertise levels based on text editing behavior.
- To evaluate the performance of the neurofuzzy system against existing methods.
Main Methods:
- Utilized a Multi-Layer Perceptron (MLP) classifier with rprop learning.
- Incorporated output data fuzzification for enhanced classification.
- Collected user interaction data from the Jove text editor over two weeks.
Main Results:
- Achieved approximately 80% accuracy in classifying user expertise levels.
- Misclassifications were limited to adjacent expertise levels.
- The neurofuzzy system outperformed binary classifiers, production rule systems, and inductive expert systems.
Conclusions:
- The developed neurofuzzy system offers an effective automated approach to user expertise modeling.
- This method provides a more accurate and robust classification of user expertise in text editing.
- The findings suggest potential for adaptive interfaces and personalized user experiences.
