<![CDATA[The woven fabric from South Central Timor (TTS), East Nusa Tenggara Province, holds significant cultural and symbolic value; however, the influx of similar-patterned fabrics from outside the region presents challenges in authenticity identification, particularly among the general public and younger generations. This study aims to develop a mobile application based on the Adaptive Neuro-Fuzzy Inference System (ANFIS) to detect and classify motifs of TTS woven fabric. Texture feature extraction was conducted using the Gray Level Co-occurrence Matrix (GLCM) method, applying six key parameters: contrast, dissimilarity, homogeneity, energy, correlation, and ASM. The ANFIS model was trained for two types of classification: fabric authenticity (authentic vs. non-authentic), achieving an average accuracy of 87.50%, and regional motif classification (Amanatun, Amanuban, and Mollo), with an accuracy of 78.00%. The application was developed using a prototyping method and integrated with the classification system via FastAPI services. Black-box testing confirmed that all application features functioned as designed, while usability testing using the Usability Metric for User Experience (UMUX) yielded a score of 87.92, indicating a high level of user comfort and ease of use. The study concludes that the ANFIS-based mobile application is effective as a supporting tool in preserving TTS woven fabric through the application of intelligent technology. Keywords: Mobile Application; TTS Woven Fabric; ANFIS; GLCM Texture Extraction; Image Classification]]>
Copyrights © 2025
