Optimizing merchant compliance: A system for product specific rule extraction using NLP

Product quality and safety are crucial in the e-commerce market, particularly for regulated sectors like the food industry, for user protection, and it is the responsibility of the merchants who sell their products to ensure these are maintained. Here comes the role of the government, which lays down certain rules and regulations that need to be followed. These rules and regulations are laid down by the central or state government of India. All these rules are enforced by the authorities established under the respective ministries. All these rules and regulations are maintained in the form of government repositories on the websites of the respective government ministries. Accessing data that is present in these repositories is very difficult for users, as these rules, regulations, and amendments are made available in PDF format to the general public and are often in an unstructured format. This initial study validates the system's effectiveness using the specific domain of Indian food safety regulations. Validation of the system on food-specific regulations demonstrated promising performance, achieving a task accuracy between 82 and 89% over 10 runs and an NLP model F-score of 87.17%. The proposed system makes it easy for users to access this data, allowing them to search for rules and regulations related to their product by providing a search engine. The proposed system gathers data scattered across various websites using web scraping. This data is then automatically cleaned and organized using automation scripts before being presented to the users in a structured and informative format. The proposed system website delivers accurate, real-time data with high speed and reliability, ensuring users get up-to-date and precise information instantly. The proposed system provides product-specific output using NLP (Natural Language Processing).•Ensuring product quality and safety in e-commerce is crucial for user protection, and merchants must maintain these standards. The government enforces rules and regulations through various ministries.•Government rules and regulations are often stored in unstructured PDF formats on ministry websites, making it difficult for users to access and interpret the information.•The proposed system uses web scraping to gather regulatory data, cleans it using automated scripts, and stores it in text format. It employs Spacy natural language processing algorithms to provide product-specific search results.