Neutrosophic Wavelet Time-Series for University Classroom Teaching Quality Analysis: A Multi-Resolution (T, I, F) Framework with Computations

Abstract

This paper presents a method for analyzing teaching quality using a 
neutrosophic wavelet framework. Weekly scores from three courses are treated as 
time series and decomposed using the Haar wavelet transform. The mean of the 
series represents the underlying trend (truth), the distribution of detail energies 
across scales captures variability (indeterminacy), and the energy of negative 
details reflects performance shortfalls (falsity). These components form a 
neutrosophic triplet (T,I,F) for each course. The method is applied to normalized 
scores over eight instructional weeks. I&P Theory shows strong performance with 
low uncertainty and downside risk, while Calculus I and Data Structures display 
more variability or shortfalls. A scalar index, the Neutrosophic Wavelet Teaching 
Quality Index (NW-TQI), integrates the triplet for comparison. The approach is 
transparent and reproducible, offering more nuanced insight than simple averages 
or standard deviations. It can be extended with real classroom data, richer 
wavelets, and predictive models for educational diagnostics. 

DOI: 10.5281/zenodo.16998644