Abstarct: Anomaly detection in a key step towards exploring new mineral deposits. There are diverse methods to find anomalies. In this thesis, geochemical sample’s grade analysis of Gold, Silver, Copper, Zinc and Lead in Latala region's boreholes and trenches, located in North West of Shahrebabak, Kerman province, Iran, were analyzed by applying Dempster-Shafer and Bayesian Information fusion methods. First, Latala region and samples locations were introduced. Then, an outline of Bayesian and Dempster-Shafer information fusion theory, functions and some well-known quantites such as priors, likelihood and posteriors were discussed and defined in this geoscience subject area. Investigations revealed that these theories results are sensitive to some parameters that used in functions, so the sensivity analysis were done for them. Finally, by applying methods to the databaxse and determining the same threshold for approached results and normalized amount of grades, outcomes has been compared by some classic geochemical technique and evaluated using grade values. Results shows that, Bayesian information fusion method by increasing the result's range, offers an acceptable separation ability for diverse statistical population. Moreover, it introduced more anomal samples than normalized amount of grades. Furthermore, Dempster-Shafer data fusion by concerning an 'unknown' subspace alongside 'anomal' and 'non-anomal' subspaces, presented more accurate results than normalized amount of grades and shows it’s capability in separating dissimilar statistical population. Like Bayesian method, it introduced more anomal samples too. Furthermore, merits and demerits of applying dempster-shafer and Bayesian information fusion method to this specific type of deposits were discussed and some suggestions for future investigations deliberated.