Sequential fuzzy association rule mining algorithm for plants environment classification using internet of things

Abstract

Every plants need an environment which is in accordance with its adaptability. However, the environment always changes so that it requires regular analysis to maintain plant fertility. Environmental data such as temperature, humidity from soil and air, rainfall, light intensity can be processed using computer algorithms in the form of a sequential (time series) matrix. The paper proposed a sequential Fuzzy Association Rule Mining (FARM) for plants environment classification using Internet of Things (IoT). FARM is used to extract association rule in form of fuzzy memberships from plant environment data that collected from sensors of IoT. Fuzzy is used to facilitate grouping of sensor data and detect changes in the environment when the degree of membership becomes irrelevant. Fuzzy membership degrees are also mapped based on time series to interpret routine environmental changes. The paper showed results of a FARM algorithm for plant environments and prototypes from IoT circuits. FARM algorithm and IoT evaluated using real time data collecting of Aglaonema costatum (Chinese Evergreen). The results shown the FARM capable to extract relevant fuzzy rules with different parameter of tolerance of dependent.