Smart Similarity Detection System

1st Dinesh Mishra
Department of Computer Science Engineering Centurion University of Technology and Management Paralakhemundi, Odisha, India
2nd Ashish Kumar Nayak
Department of Computer Science Engineering Centurion University of Technology and Management
Paralakhemundi, Odisha, India
3rd Sunayana Panda
Department of Computer  Science Engineering Centurion University of Technology and Management Paralakhemundi, Odisha, India
230101120096@centurionuniv .edu.in
4th Ram Prasad Satpathy
Department of ComputerScience Engineering Centurion University of  Technology and Management Paralakhemundi, Odisha, India
230101120099@centurionuniv .edu.in
230101120076@centurionuniv .edu.in
5th Dhawaleswar Rao CH
Head Of The DepartmentDepartment of Computer Science Engineering Centurion University of Technology and Management Paralakhemundi, Odisha, India
dhawaleswar.rao@cutm.ac.in

Abstract-In the digital era, the rapid growth of online information sharing has significantly increased concerns regardingplagiarism and  content duplication. Conventional plagiarism detection systems primarilyrely on keyword matching and string-based comparison,which often fail to identify paraphrased o r emantically
similar content. This paper presents a Smart Similarity Detection System, an Artificial Intelligence (AI) driven web-based solution that evaluates semanticsimilarity between text documents using Natural Language Processing (NLP) and deep learning techniques.  230101120097@centurionuniv .edu.in content with higheraccuracy than traditional methods. This approach provides a reliable and scalable solution for academic integrity verification, content originality checking, and research document analysis.
Keywords: Semantic Similarity, Plagiarism Detection, Natural Language Processing, Deep Learning, Sentence Transformers, MPNet