Search

The crop recommendation of the rural farmers is important in our country; this research work aim is to increase the profit of the farmers by suggesting the suitable crop recommendation in towns and villages of Tamil Nadu. The agriculture sectors are widespread that requires thorough preparation and judgement. Artificial intelligence and machine learning algorithms are extended practically in every major area, including agriculture. Data on Tamil Nadu's agricultural production were obtained through an open data platform and also from the manual of the Economic and Statistical Department of Tamil Nadu which is published each year. Their main objective was to collect knowledge through data that could be applied to obtain useful predictable results. Hence, to achieve these objectives, fuzzy ant clustering with detection of cluster similarity and cluster combination along with association rule mining is used to provide crop recommendation to farmers depending on the current season and soil type. By evaluating the previous year's agriculture production record, analyse the yield produced in the previous year by various crops and seasons. An algorithm using fuzzy ant clustering with detecting and combining the overlapping nodes to reduce the redundancy and improve the quality of the clusters was developed. The evaluation results show that the fuzzy ant colony with overlapping cluster detection algorithm provides good RS of the crops as the error rate is decreased to 8 percentage and accuracy is increased to 91.9 percentage when compared with results obtained from crop recommendation system with ant colony clustering and association rule mining. [ABSTRACT FROM AUTHOR]