cnn-bilstm

In PyTorch Learing Neural Networks Likes CNN、BiLSTM

Performed comparative analysis of BiLSTM, CNN-BiLSTM and CNN-BiLSTM with attention models for forecasting cases.

Transformer for Action Recognition in PyTorch

The model leverages the strengths of both CNNs and BiLSTM networks to effectively capture spatial and temporal patterns in network traffic data. We trained and evaluated the model using a comprehensive dataset of cyber attacks. The model achieved a high accuracy of 99%.

[ICTC'24] - "Voice-Based Age and Gender Recognition: A Comparative Study of LSTM, RezoNet and Hybrid CNNs-BiLSTM Architecture" by Nhut Minh Nguyen, Thanh Trung Nguyen, Hua Hiep Nguyen, Phuong-Nam Tran, Duc Ngoc Minh Dang

doctor_prescription_recognization_using_DeepLearning project for epics

Research repository for nervous smile detection deep learning algorithm. 2 models, a KNN-SVM and CNN-BiLSTM are compared.

AI-powered protein mutation analysis platform. Enter any UniProt ID, mutate a residue, and SERAPH, an original ESM2+CNN+BiLSTM model, predicts exactly which secondary structures change and why.

Comparison of machine learning architectures for neural decoding of self-location | MSc dissertation project

Image-based malware family classification using a hybrid CNN-BiLSTM deep learning model.

Final Year project: SafeSpace is a machine learning and NLP-based framework designed to detect cyberbullying behaviour in real-time, particularly focusing on Hinglish (Hindi + English) text.

Developed a Dysgraphia Detection web app using React, Flask, and TensorFlow to detect dysgraphia from handwriting samples. Built a CNN-BiLSTM model achieving 91% accuracy and provided structured reports for user-friendly diagnosis.

A collaborative machine learning and deep learning project for phishing URL detection, featuring multiple detection pipelines, comparative model evaluation, feature engineering, and a Chrome extension for real-time URL classification.

End-to-end automated sleep stage classification using wearable EEG (Dreem Headband) — benchmarking classical ML, deep learning, and transfer learning (MobileNetV2) on multimodal physiological signals. Best: 82% accuracy with XGBoost, 71.2% with fine-tuned CNN.

Deep learning model to detect cyberbullying in code mixed text (Romanized Hindi, Hindi, English) using a CNN–BiLSTM architecture. Includes preprocessing pipeline and can be integrated into websites, software, or mobile apps for real-time content moderation

🛡️ Intrusion-Aware Adaptive Encryption for IoMT | CNN-BiLSTM IDS + ECDH + AES + Self-Healing

Real-time Bangla Speech Emotion Recognition (BSER) using a Hybrid CNN-BiLSTM model with incremental learning and LLM integration for emotionally adaptive responses.

The project of implementing different deep learning models on doing machine-generated text detection and even the mixed human-machine text detection