Performance comparison of different model combinations, including CNN, LSTM, CLIP and their hybrid architectures, evaluated using precision, recall, F1 score and AUC. The table demonstrates how different combinations impact performance, with models like CLIP + LSTM showing lower scores, whereas CNN + LSTM and CLIP + CNN provide improvements. The proposed HIMP model outperforms all others, achieving the highest scores across all metrics, particularly with an F1 score of 91.20% and an AUC of 95.36%. These results emphasize the robustness of HIMP compared to traditional and hybrid architectures