DATA MINING FOR RANKING COOKING RECIPES

Abstract

CULINARY RESEARCH IS USUALLY CARRIED OUT WITH PRIOR KNOWLEDGE OF THE RECIPE, HOWEVER, THE SEARCH BASED ON INGREDIENTS BECOMES A NECESSITY SO THAT THE USER HAS MORE OPTIONS THAT HE DOESN'T KNOW, BUT WHICH ARE IN ACCORDANCE WITH WHAT IS WITHIN HIS REACH. THIS DOCUMENT AIMS TO DESCRIBE THE PROCESS OF DEVELOPING AN APPLICATION IN THE PWA PATTERN (PROGRESSIVE WEB APP), WHICH SUGGESTS RECIPES RANKED BY RELEVANCE. DATA MINING IS A METHOD USED TO EXPLORE LARGE AMOUNTS OF DATA AND EXTRACT INFORMATION, AND IN THE CONTEXT OF THIS WORK, IT WILL BE USED TO OBTAIN AND CLASSIFY RECIPES EXTRACTED FROM CULINARY SITES. SCRUM AND LEAN INCEPTION METHODOLOGIES WERE USED, ALONG WITH KANBAN THROUGH THE TRELLO TOOL THAT WAS CHOSEN FOR PROJECT MANAGEMENT, WHICH ALLOWS THE VISUALIZATION OF THE FLOW AND STATUS OF ACTIVITIES, IN ADDITION TO THE ASSIGNMENT OF RESPONSIBILITIES AND DEADLINES. C# AND .NET CORE TECHNOLOGIES WERE USED FOR THE BACK-END PROGRAMMING, TYPESCRIPT NEXT JS REACT FOR THE FRONT-END, NODE JS FOR THE CRAWLER AND MONGODB (NOSQL) FOR THE DATABASE. THE ELEMENTS PRODUCED DURING THE PROCESS WERE: PROJECT MANAGEMENT DOCUMENTS (SUCH AS THE PM CANVA AND DEVELOPMENT MODEL), MODELING AND PRODUCT DEFINITION ARTIFACTS (SUCH AS PROTOTYPING AND PERSONA DEFINITION), WEB MINING ALGORITHM, AND THE DEVELOPMENT OF THE MVP (MINIMUM VIABLE PRODUCT) WHICH WAS MADE AVAILABLE ON THE WEB ADDRESS �€ŒHTTPS://CHELF.APP�€ WITH AROUND 500 RECIPES AVAILABLE FOR IMMEDIATE RESEARCH. DURING THE DEVELOPMENT OF THE PROJECT, IT WAS POSSIBLE TO OBSERVE SEVERAL PATTERNS AND CLAIMS THAT WERE ANALYZED TO IMPROVE THE QUALITY OF THE FINAL PRODUCT. AMONG THEM, IT WAS OBSERVED THAT IN THE DATA MINING PROCESS THERE WAS A NEED TO FILTER RECIPES MORE RELEVANT TO THE COMMUNITY, IN VIEW OF THE QUALITY OF THE WRITING AND THE CONTENT OBTAINED.