Description The goal of our joint Asian Pacific Telecommunity (APT) and CyberBrain Mass Agriculture Alarm Acquisition and Analysis (CBMA4) project is to develop a collective intelligence-based platform that accommodates a decision making process related to agriculture early warning among communities of farmers. The idea is to translate tacit knowledge from local farmers as well as fundamental knowledge from universities, research institutes, and government agencies into a computerized system. The system alerts and suggests solution for farmers based on two sources of input: Weather data observed automatically from weather stations and contextual data collected by farmers. We also propose an observation object named WarnCon (Warning Content) and an algebraic model to reduce the semantic gap in agriculture early warning management. The results are collected from a real experimental farm in Thailand where the local farmers are currently use our system to reduce risk in plant disease.