The invention provides a multi-source heterogeneous data truth value determination method and device. Heterogeneous conflict data can be subjected to joint processing, and the truth value discovery accuracy is improved. The method comprises the following steps: S1, acquiring heterogeneous conflict data from different data sources; S2, for conflict data describing the same object, aiming at each object and all objects of all data sources, respectively constructing a target function G and an optimization model F which take maximization of a declaration value credibility weighted sum as a target;S3, for each object, updating the weights of all the data sources by adopting a truth value selection strategy based on an exhaustion method; s4, calculating an F value according to the updated weights of all the data sources, judging whether the optimization model F is converged or not according to the obtained F value, and if not, returning to S3 to continue to execute; and if so, forming an optimal truth value set by the obtained optimal truth values of all the objects. The invention relates to the field of data mining.