Significance of Measuring the Accuracy of Cellular Automata Markov Chain for Land Use Projections in District Gurgaon in Haryana, India

The significance of measuring the accuracy of Cellular Automata Markov Chain (CA MC) for land use projections lies in urban planning, environmental management, and sustainable development. The reliability of the CA MC model in predicting Land Use (LU) changes requires the accuracy and robustness of the model that uses data from a study area to compare the land cover (LC) changes over a period. The model with appropriate transition rules compares the significant LC changes for LU prediction. The reliability metrics assess the prediction model using the Kappa coefficient, overall accuracy, user accuracy, and producer accuracy to measure the predicted changes for chance agreement. The study reassigns the non-diagonal elements of the state transition matrix, derived from the confusion matrix, by interpreting them as TRUE rather than considering FALSE and discarding them to provide improved measured overall accuracy. These reassigned changes provide realistic insights into district Gurgaon's predicted changed map in Haryana, India, which can help policymakers, urban planners, and stakeholders make informed decisions about land management, infrastructure development, and resource allocation. An overall model accuracy of 81.33% for predicted LC data supports policymakers in developing plans and policies to assess LU patterns and trends for sustainable practices aligned with environmental conservation and economic needs.