A classification system using Hybrid KNN–Decision Tree approach to analyze and classify student dependency levels on Artificial Intelligence technology.
This research aims to classify student dependency levels on AI technology in academic activities and daily life using machine learning methods.
Minimal AI usage, more relying on personal abilities
Balanced AI usage as a learning tool
Highly dependent on AI in various activities
With the expanding use of AI such as ChatGPT, Gemini, and others, it is important to understand student dependency levels and their impact on learning.
Using the KNN algorithm that is proven effective in data classification with high accuracy rates.
All respondent data is kept confidential and used only for research purposes.
Research results provide insights for wiser and more productive AI usage.
Machine learning algorithm used for dependency level classification
Calculate the distance between test data and training data:
d(x,y) = √(Σ(xi - yi)²)
x = Test data
y = Training data
d = Euclidean distance
Determine class based on K nearest neighbors:
Class = Mode(K-Nearest Neighbors)
K = Number of nearest neighbors
Mode = Most frequently occurring class
Output = Low, Moderate, or High
Aspects measured in this research
Frequency, duration, and context of AI usage in academic activities.
Student intrinsic and extrinsic motivation in using AI.
Perception of AI usefulness and ease of use in learning.
AI usage behavior and ethical awareness and reflection of students.
Your participation is very valuable for the success of this research.
Time to complete: ±10-15 minutes
Easy steps to participate in this research
Enter your data such as NIM, name, study program, and GPA.
Fill out the questionnaire according to your experience using AI.
Submit your answers and data will be processed for classification.