A Self-driven dual reinforcement model with meta heuristic framework to conquer the iot based clustering to enhance agriculture production

Abstract

In agriculture, irrigation is essential for providing water to crops based on the type of soil they are grown in. To achieve success in farming, it is important to evaluate soil fertility, temperature, rainfall and set irrigation schedules. In this work, to enhance agricultural production, an IoT-based hydration system that utilizes soil moisture and humidity sensors to keep an eye on soil conditions and water crops precisely is developed. This system effectively manages water usage in farming, resulting in an efficient conservation of water resources. The proposed model is categorized into four main phases: (a) Pre-processing; (b) Clustering; (c) Feature extraction; (d) Classification. Initially, the collected raw data is pre-processed via a data-cleaning approach. From the pre-processed data, DB scan is used to cluster the data. From the clustered data, the important feature is extracted by using statistical features like mean, standard deviation, kurtosis, and skewness. Subsequently, from the extracted features, the most optimal features are selected via a new hybrid meta-heuristic optimization model referred to as Cuckoo search-based Levy adolescent identity search algorithm (CLAIS). The projected CLAIS model is the conceptual amalgamation of the standard Cuckoo search optimization (CSO) and adolescent identity search algorithm (AISA), respectively. CLAIS-based deep Q network (CLAIS-DQN) classifier is used to classify the optimal features. DQN is an efficient solution to offload the request optimally which improves the overall performance of the network. The proposed model is implemented using the PYTHON platform. The proposed model has recorded the highest detection accuracy as 96%.