CLASSIFICATION OF FOOD DISTRIBUTION STATUS AND PATTERN MAPPING OF OPTIMAL DISTRIBUTION IN WEST JAVA

• Main Author: Zulkarnain, Handy
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/73908
• Institution: Institut Teknologi Bandung

Inequality in the distribution of food is a problem that is still being faced in Indonesia. This is indicated by the existence of areas that experience surpluses and deficits in the number of certain types of food. Therefore, achieving national food security, which is one of the targets of Indonesian government, is obstructed. One solution that can be applied is to apply clustering, which is a data mining technique to identify and create groupings of several areas with similar characteristics such as the tendencies of receiving food supplies, transporting food supplies, and consuming food supplies. The purpose is to provide information regarding the mapping of certain areas that receive a certain food distribution. There are a number of studies that examine the distribution of a food product. However, the solutions offered are still not applicable to cases of food inequality because these solutions can only be applied in a narrow and limited research environment. In addition, these solutions will require large computational resources due to the size of food distribution dataset. Moreover, this research paper will discuss the utility of polygon objects which form those areas in geographical setting and to compute the distance between areas. This research will utilize the CRISP-DM methodology, and the results of the research will be obtained by clustering the dataset. This study aims to find condition of food distribution by applying clustering, as well as proposing the most optimal way to distribute food based on the distances computed.