Indigenous farmers often face challenges in determining fair prices for agricultural commodities due to limited access to timely and reliable market information, weakening their bargaining power and income stability. In this study, we aimed to: (1) identify challenges facing indigenous farmers when determining accurate prices for agricultural commodities, (2) determine how the features of IndiGenius Mart address these issues, and (3) evaluate the acceptability of the proposed system based on the ISO/IEC 25010 software quality model in terms of functional suitability, performance efficiency, usability, reliability, and portability. We conducted a questionnaire survey among 50 indigenous farmers using purposive sampling. Results indicated a strong consensus (weighted mean = 4.47) that inadequate price information leads to unfair pricing, financial losses, and inefficient selling schedules. To address these concerns, we developed the IndiGenius Mart, an Al-based, offline-capable mobile application for crop price forecasting that integrates time-series analysis and ensemble machine learning models. The model achieved a high coefficient of determination (R2 = 0.982) and demonstrated consistent prediction accuracy on unseen data (96-98%). Reliability analysis showed excellent internal consistency (Cronbach’s a = 0.93-0.96; overall a = 0.95). System evaluation yielded acceptable ratings across all ISO/IEC 25010 criteria, with a weighted overall mean of 3.90 (SD = 0.29), interpreted as Acceptable. We conclude that accessible and localized Al-driven crop price forecasting tools enhance price transparency and improve decision-making among indigenous farmers, consistent with prior data-driven agriculture studies. Future research should focus on expanding localized datasets and evaluating long-term real-world deployment across diverse regions.

Keywords: Indigenous farmers, crop price forecasting, ensemble learning, time-series analysis, machine learning, agricultural markets, ISO/IEC 25010, decision support system