MAKE MOST OF THE KNOWLEDGE NETWORK, JOIN ACADEMIC RESEARCH FOUNDATION

Odisha is one of the agricultural dependent states of India. Due to Odisha Millet mission millets have been popularized throughout the state. Ragi is now one of the super food that is being consumed by majority of the population which has a scope of growing in a large scale in the future. In this study ARIMA model was compared with hybrid Ensemble Empirical Mode Decomposition (EEMD)-ARIMA to evaluate the past behavior of the time series data regarding the production of Ragi, in order to make inference about its future behavior. The forecasting performance of these models are evaluated and compared by using the Root Mean Square Error, Mean Absolute Percentage Error and Mean absolute error. The findings reveal that superiority of Hybrid EEMD-ARIMA model than the ARIMA model for forecasting.

Keywords: ARIMA, EEMD, Forecasting, Hybrid EEMD-ARIMA, RMSE.

Breast cancer remains one of the leading causes of death among women globally. The BRCA1 (Breast Cancer 1) gene, a tumour suppressor, plays a critical role in DNA repair, cell cycle regulation, and genomic stability. Mutations in BRCA1 significantly increase the risk of hereditary breast and ovarian cancers. In this study, we employed a comprehensive in-silico pipeline incorporating structure modelling, molecular docking, pharmacokinetics (ADMET) profiling, and molecular dynamics simulations to identify potential drug candidates targeting BRCA1. Several lead compounds exhibited strong binding affinities and favourable drug-like properties, supporting their further investigation as targeted therapies.

Keywords: BRCA1, In-silico, Drug Discovery, Molecular Docking, ADMET, Molecular Dynamics, Phytochemicals.

Accurate forecasting of flower crop production is essential for data-driven decision-making and sustainable planning in the floriculture sector. Marigold, a commercially valuable crop in the Chitraduga district of Karnataka, is highly influenced by fluctuating agro-climatic conditions, making production forecasting a vital tool for growers and policymakers. This study presents a comparative analysis of two machine learning algorithms Support Vector Regression (SVR) and k-Nearest Neighbors (KNN) for forecasting Marigold production using historical production data. Model performance was assessed using key evaluation metrics: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Squared Logarithmic Error (MSLE), and the Coefficient of Determination (R²). Empirical results demonstrated that the KNN model outperformed SVR, achieving a lower MAE (408.9535), RMSE (714.1817), and MSLE (0.2340), alongside a substantially higher R² value (0.4282), indicating robust predictive accuracy. In comparison, the SVR model recorded higher MAE (411.8339), RMSE (775.2162), MSLE (0.2986), and a lower R² (0.3262), reflecting a relatively weaker fit. The superior performance of KNN may be attributed to its ability to capture local nonlinear variations more effectively, which is crucial in heterogeneous agricultural environments. This study underscores the importance of algorithm selection in crop production modeling and highlights the potential of KNN as a reliable tool for forecasting Marigold production. These findings offer practical implications for improving precision agriculture practices and call for further validation of machine learning models across varying agro-ecological contexts to support resilient and informed floricultural development.

Keywords: Chitradurga, Marigold, Support Vector Regression, k-Nearest Neighbors, Forecast, Coefficient of Determination.

Accurate rainfall prediction is critical for agricultural planning and water resource management in agrarian regions. This study presents a comparative analysis of three machine learning models—Decision Tree (DT), SVM, and K-Nearest Neighbors (KNN)—for predicting monthly rainfall in the Rajnandgaon district of Chhattisgarh, India. Historical meteorological data was sourced from the NASA POWER repository and preprocessed to handle missing values, outliers, and feature scaling. The models were implemented in Python and evaluated using standard metrics: Coefficient of Determination (R²), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE). Results demonstrated that the KNN model outperformed the others, achieving the highest R² value (0.81) and the lowest errors (RMSE: 61.30 mm, MAE: 38.16 mm). Based on its superior performance, the KNN model was employed to forecast monthly rainfall for the period 2023–2028, revealing significant inter-annual variability. The findings indicate that KNN is a highly effective tool for rainfall prediction in this region, offering valuable insights for stakeholders in agriculture and water management to enhance climate resilience and decision-making. And KNN is used to forecast Average monthly rainfall from 2023 to 2028.

Keywords: Rainfall Prediction, Rajnangaon, Machine Learning, KNN, Support Vector Machine, Decision Tree.

This research investigates decadal trends in the Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST) across Raipur District, India, from 2015 to 2025, utilizing Landsat- 08 and Landsat-09 satellite imagery analyzed through remote sensing and GIS methods. The study identifies notable spatiotemporal variations in vegetation health and surface temperature over the ten-year period. NDVI values indicate a rise in average vegetation indices from 0.163 in 2015 to 0.410 in 2020, followed by a slight decrease to 0.396 by 2025, reflecting dynamic shifts in vegetation cover. Meanwhile, average LST shows a modest decline from 38.42°C in 2015 to 38.37°C in 2020, and a more significant drop to 32.76°C by 2025, with maximum temperatures also decreasing substantially by 2025. An inverse correlation between NDVI and LST highlights that regions with denser vegetation typically exhibit cooler surface temperatures, underscoring vegetation’s role in temperature regulation. The study highlights how urbanization and land cover changes affect Raipur’s climate and vegetation, offering insights for sustainable land management. Continuous monitoring via remote sensing and GIS is vital for climate adaptation in rapidly developing areas.

Marigold, a fragrant ornamental flower crop, holds significant economic and aesthetic value in Karnataka, particularly in Tumkur district. Known for its use in garland making, perfume extraction, and decorative arrangements, Marigold cultivation supports both livelihoods and regional floriculture markets. In this study, secondary data on Marigold production in Tumkur district from 1981 to 2023 were obtained. A comprehensive trend analysis was performed using various statistical models, including Linear, Quadratic, Cubic, Exponential, and Sinusoidal models, to examine long-term changes in production and productivity. The models were evaluated based on R-squared (R²) values, which indicate the proportion of variance explained, and Root Mean Square Error (RMSE), which reflects prediction accuracy. Among the models assessed, the cubic model provided the best fit for Marigold production and linear model provided the best for Marigold productivity, exhibiting the highest R² and the lowest RMSE values, signifying its superior ability to capture the underlying pattern in the data. These results suggest that Marigold production in Tumkur has undergone non-linear changes over the years, possibly influenced by varying agro-climatic conditions, market demand, and farmer adoption of improved varieties.

Keywords: Linear, Quadratic, Cubic, Exponential, sinusoidal, Root Mean Sum of Squared R Squared.