Bibliographic Details
| Title: |
A comprehensive approach for determining sentence similarity using tokenization and parsing. |
| Authors: |
Vayadande, Kuldeep1 (AUTHOR) kuldeep.vayadande1@vit.edu, Patle, Parul1 (AUTHOR) parul.patle20@vit.edu, Fole, Triveni1 (AUTHOR) triveni.fole21@vit.edu, Mandot, Pavitra1 (AUTHOR) pavitra.mandot20@vit.edu, Mukhane, Dhiraj1 (AUTHOR) dhiraj.mukhane20@vit.edu, Narang, Rohant1 (AUTHOR) rohant.narang20@vit.edu |
| Source: |
AIP Conference Proceedings. 2026, Vol. 3202 Issue 1, p1-6. 6p. |
| Subject Terms: |
*Natural language processing, *Text processing (Computer science), *Plagiarism prevention, *Semantic computing, *Linguistic analysis, *Resemblance (Philosophy), *Parsing (Grammar) |
| Abstract: |
This text similarity checker is a tool that uses advanced natural language processing to assess how similar two sentences are to one another. The tool prompts the user to input two sentences and pre-processes them using a tokenizer. It also performs various tasks such as removing stop words, punctuation, and numbers, converting all text to lowercase, and lemmatizing the text. After pre-processing, the tool uses a parsing module to analyse the structure of the sentences and identify their meanings. The two sentences are compared for resemblance using the Jaccard similarity coefficient and the similarity of the sentences' meanings using synset retrieval. Calculation of the ultimate score for similarity as a weighted average of both measures and presented to the user as a percentage. The tool provides a thorough process for determining how similar two sentences are, considering both their surface-level and semantic features. It is useful for various applications, including plagiarism detection, essay grading, and document analysis. [ABSTRACT FROM AUTHOR] |
| Database: |
Academic Search Index |