Reduced Time Compression in Big Data Using MapReduce Approach and Hadoop.

Area of Science:

• Data Science
• Artificial Intelligence
• Agricultural Science

Background:

• Exponential growth in real-time data volume necessitates efficient processing techniques.
• Classification of big data is a complex and time-consuming task.
• Integrating climatological and meteorological data with farming decisions presents challenges.

Purpose of the Study:

• To propose a MapReduce-based framework for processing heterogeneous big data.
• To enhance climate classification and prediction using the Co-Effective and Adaptive Neuro-Fuzzy System (Co-EANFS).
• To examine association rule mining for optimizing crop production based on soil and weather conditions.

Main Methods:

• Utilized the MapReduce framework for parallel data processing and distribution.
• Employed the Co-Effective and Adaptive Neuro-Fuzzy System (Co-EANFS) for empirical climate classification and prediction.
• Implemented association rule mining to identify optimal crop production strategies.
• Developed a technique for managing data preprocessing, clustering, classification, and prediction levels.

Main Results:

• The proposed model effectively processes weather datasets, forming seasonal cluster datasets.
• Co-EANFS demonstrated accurate predictions with varying inputs and variables.
• The framework achieved minimal execution time, indicating high efficiency.

Conclusions:

• The integrated MapReduce and Co-EANFS framework offers an efficient solution for big data challenges in agriculture.
• The approach provides valuable insights for crop production by linking climate data with farming decisions.
• The proposed technique successfully manages multiple data processing stages, leading to optimized outcomes.