This project implements Latent Semantic Indexing (LSI) to effectively match items in transaction records to items listed in a pricing list. By leveraging the power of LSI, the project aims to reduce the ambiguity and enhance the accuracy of item identification, ensuring that transaction items are correctly priced according to the most relevant pricing list entries.
- Automated Item Matching: Utilizes LSI to automate the matching process, reducing manual effort and errors.
- High Accuracy: Improves the matching accuracy by understanding the semantic context of item descriptions. Adopt common techniques such as Named Entity Recognition, Chunking and parsing and Stemming and lemmatization.
- Customizable: Allows users to adjust the sensitivity of the matching algorithm; adjust threshold of similarity; hard code edge cases.
- Python 3.11 or higher
- Pip for installing dependencies