عنوان مقاله [English]
Usually, geographical features have a kind of adhesion to each another, and a feature attracts other features, for example, near to a hospital or clinic, there are set items of doctors' offices, pharmacies, radiology, laboratories, restaurants, food, flower shops, supermarkets. These item set are placed near to each other so that as a result of accumulation and establishment in the neighbor they complement each other and meet the needs of customers. In this article, we explore the frequent pattern of urban features that are Frequent repeated in different parts of Tehran. To perform research, we have used the location of branches of banks and financial and credit institutions in Tehran (3773 branches) as a case study. Urban feature included 56 classes and in total, 81066 features were used for processing. Apriori algorithm has been used to discover patterns. Apriori is an algorithm for frequent item set mining and association rule learning. The features located within a 200-meter radius of bank branches have been extracted. Then they have been converted into geographic binary matrix and transferred to the Python environment. To run the model, we consider the minimum support level of 0.05 (5%), in other words, only items that occur at least 5% will be used in the model. After processing the information, the association rules were mining. The most frequent urban items was insurance and restaurant which played a greater role in the construction of frequent urban itemset.