MOHAMMED NOWFAL

ABSTRACT

Developed a machine learning-based system to automate film censorship by detecting profanity, violence, and Sensitive content in movies. The solution integrates CNN for image frame analysis, LSTM for sequence and language modeling, and NLP for subtitle/text analysis. The system identifies objectionable content and provides timestamp outputs to assist in automated moderation.

TECHNOLOGY USED

• CNN – Detects visual elements like blood or violence in video frames
• LSTM – Handles audio/text sequences and detects offensive language
• NLP – Analyzes subtitles and transcribed audio
• TensorFlow, Keras– Model development and training

FINAL OUTCOME

• Automated detection of visual, audio, and textual violations
• Timestamped censorship support for manual or automatic review
• Useful for OTT platforms, live streaming, and content moderation
• Reduced manual workload for censorship boards

ABSTRACT

Developed a Faculty Recommendation System using XGboost, a machine learning algorithm, to match students with teachers based on preferences like subject, time slot, and teaching style. The system analyzes historical data to improve accuracy and provides personalized suggestions through a web interface.

TECHNOLOGY USED

• XGBoost – For accurate faculty recommendations
• Machine Learning – To learn from student-teacher interactions
• Web Interface – For input and displaying suggestions

FINAL OUTCOME

• Personalized faculty suggestions based on student preferences
• Improved academic engagement and satisfaction
• Real-time recommendations with feedback-based learning