• PATHI JEJI DURGA
• Dr.EEDARA ASWANI KUMAR
• NIRJOGI LIKHITHA
• KARIBANDI NAVYA RAGHURAM
• CHILUKURI SATYA SAI DHEERAJ

Road traffic accidents remain an important global problem, causing numerous deaths and serious injuries every year. Proper prediction of the severity of accidents is significant for enhancing emergency response strategies and forming effective policies on road safety. This review will cover a broad overview of AI techniques applied for predicting the severity of road accidents, pointing at approaches centered on ML and DL. The traditional ML approaches considered are decision trees, random forests, SVMs, and logistic regression, which are expected to be effective in dealing with structured data while highlighting key predictors. These models will be a better fit in low-data scenarios as they are interpretive, which is necessary for stakeholders making the policy decisions. Ensemble methods, such as gradient boosting and bagging, are also explored because they can improve predictions by aggregating weaker individual models. On the other hand, DL techniques, such as CNNs, RNNs, and LSTMs, can be very efficient for processing unstructured data like images from traffic cameras or time series generated from sensors. Those models accurately capture complex nonlinear relationships within large datasets; they are also more accurate than traditional methods. However, while DL models deliver higher quality models, ML models are more advantageous when used in settings of limited resource, where interpretability and ease of implementation are most paramount. Hybrid models that make a blend of ML and DL techniques offer improved performance, enabling more accurate predictions and better policy decisions, which end up contributing to safer road environments.

Road crash injury severity prediction, Machine Learning techniques, Deep Learning techniques, Road safety.

"PREDICTION OF ROAD ACCIDENT SEVERITY USING DEEP LEARNING TECHNIQUE FOR PROACTIVE EMERGENCY MANAGEMENT", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 12, page no.e532-e552, December-2024, Available :http://www.jetir.org/papers/JETIR2412461.pdf

"PREDICTION OF ROAD ACCIDENT SEVERITY USING DEEP LEARNING TECHNIQUE FOR PROACTIVE EMERGENCY MANAGEMENT", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 12, page no. ppe532-e552, December-2024, Available at : http://www.jetir.org/papers/JETIR2412461.pdf