Understanding student feedback is essential for informing pedagogical strategies and institutional decision-making in higher education. Sentiment analysis offers scalable mechanisms for extracting insights from open-ended student evaluations; however, many existing approaches prioritize technical performance without sufficient consideration of contextual and institutional constraints, particularly in underrepresented regions. This study proposes a context-aware framework for sentiment analysis of student feedback, designed to support educational decision-making within Latin American universities. Rather than introducing new algorithms, the framework systematically evaluates established machine learning and deep learning models through a multi-phase process that includes data preprocessing, Bayesian optimization, threshold calibration, and class balancing. The framework is validated using authentic Spanish-language student feedback collected from a public university in Peru. Experimental results indicate that while advanced models can achieve strong predictive performance, simpler and more interpretable approaches often provide comparable institutional value when deployment feasibility, computational efficiency, and transparency are considered. These findings highlight that marginal performance gains do not necessarily translate into meaningful advantages for routine educational use. Overall, this work contributes a replicable and resource-sensitive framework that bridges learning analytics research and practical educational application. By prioritizing contextual suitability and interpretability, the proposed approach enables higher education institutions to leverage student sentiment data as an actionable input for continuous improvement and evidence-based educational strategies.

The attenuation of resonant frequencies across the entire spectrum of an audio signal is important because it helps to eliminate the harshness, sibilance, clear muddiness, boominess, and proximity effect of any sound source. This paper presents a method for the attenuation of resonant frequencies across the entire spectrum of an audio signal. A spectrum obtained by the Fast Fourier Transform is segmented into bands—one-third octave bands and Equivalent Rectangular Bandwidth-scale bands—in order to obtain the maximum value per band. Additionally, a curve representing the general shape of the spectrum is generated using the standard deviation to create a threshold curve for detecting resonant frequencies. The array with maximum values per bands and the array with the threshold curve are used to detect the resonant frequencies and calculate the attenuation for each filter. Subsequently, the coefficients of a second-order section of IIR-Peak filters are calculated for processing the input signal. Twenty audio files from different sources are utilized to test the algorithm. The output produced is then compared to that produced by the commercially available Soothe2 and RESO plug-ins. The Root Mean Square Level and the Loudness Units Full Scale integrated metrics are reported. The proposed plug-in output is more attenuated than the output from commercial plug-ins under factory conditions. The average RMS attenuation is −2.32 dBFS, while Soothe2 and RESO exhibit −1.27 dBFS and −1.10 dBFS, respectively. The attenuation per octave band over time is calculated using the Wavelet Transform. Finally, an annotator agreement used as a subjective result is made with 40 people related to audio and music in order to verify if the attenuation generated by the present work at resonant frequencies agrees with subjective opinion. The octave band analysis and annotator agreement show that the proposed plug-in performs better on audio from vocal, percussion, and guitar ensembles.

En 2023, se observó un incremento en la incidencia y prevalencia de actos delictivos en México, con variaciones significativas entre las diferentes entidades federativas. El Estado de México destacó particularmente por registrar un aumento en los niveles de eventos delictivos con respecto al año anterior. Los delitos más comunes incluyeron robos perpetrados mediante distintas modalidades, destacando el uso de armas blancas y de fuego, así como agresiones físicas directas. Por otro lado, la violencia en los entornos escolares en México ha mostrado tendencias alarmantes en varios tipos de agresión. De lo antes mencionado, se destaca la importancia de implementar técnicas avanzadas que permitan la identificación oportuna de hechos delictivos en entornos escolares a fin de salvaguardar la integridad de los estudiantes. Este estudio explora el uso del modelo YOLOv5s para la detección de arrebatos, armas blancas y de fuego, específicamente en el Tecnológico de Estudios Superiores de Tianguistenco utilizado imágenes digitales obtenidas a través de cámaras fotográficas. Los resultados obtenidos indican que este modelo logra clasificar incidentes que involucran el uso de armas blancas, armas de fuego y arrebatos, alcanzando un promedio de precisión media del 60 %.