Intelligent geographic information system (IGIS) is one of the promising topics in GIS field. It aims at making GIS tools more sensitive for large volumes of data stored inside GIS systems by integrating GIS with other computer sciences such as Expert system (ES) Data Warehouse (DW), Decision Support System (DSS), or Knowledge Discovery Database (KDD). One of the main branches of IGIS is the Geographic Knowledge Discovery (GKD) which tries to discover the implicit knowledge in the spatial databases. The main difference between traditional KDD techniques and GKD techniques is hidden in the nature of spatial data sets. In other words in the traditional data set the values of each object are supposed to be independent from other objects in the same data set, whereas the spatial dataset tends to be highly correlated according to the first law of geography. The spatial outlier detection is one of the most popular spatial data mining techniques which is used to detect spatial objects whose non-spatial attributes values are extremely different from those of their neighboring objects. Analyzing the behavior of these objects may produce an interesting knowledge, which has an effective role in the decision-making process. In this thesis, a new definition for the spatial neighborhood relationship by is proposed considering the weights of the most effective parameters of neighboring objects in a given spatial dataset. The spatial parameters taken into our consideration are; distance, cost, and number of direct connections between neighboring objects. A new model to detect spatial outliers is also presented based on the new definition of the spatial neighborhood relationship. This model is adapted to be applied to polygonal objects. The proposed model is applied to an existing project for supporting literacy in Fayoum governorate in Arab Republic of Egypt (ARE).