20.500.12556/DKUM-84450 Development of the web service for grocery store receipt recognition and subsequent analysis of the nutritional value of the products purchased Razvoj spletne storitve za prepoznavo računov in analizo prehranskih vrednosti izdelkov The objective of this study is to tackle the worldwide problem of cardiovascular diseases (CVDs). These diseases represent the foremost reason for mortality on a global scale, accounting for approximately 17.9 million deaths annually. Unhealthy eating habits are one of the most frequent causes for CVD in the world and are addressed in various ways, including digital technologies. The goal of this project i`s to develop a web service that enables a non-intrusive way of tracking food consumption and monitoring the dynamics of nutrient intake compared to the traditional "record-every-meal" type of tracking. We followed a design science research approach to develop the web service. The web service employs semantic similarity algorithms and enables human feedback to improve its performance. It will be developed using the Serverless framework and deployed on the AWS platform. Receipt recognition will be performed using AWS Textract, and the nutritional value of the products will be obtained from publicly available databases. This approach aims to create an innovative and user-friendly method for tracking food consumption and nutrient dynamics. Cilj te študije je soočiti se s svetovno težavo srčno-žilnih bolezni (Cardiovascular diseases - CVD). Te bolezni predstavljajo glavni razlog za smrtnost na globalni ravni, saj povzročijo približno 17,9 milijona smrti letno. Nezdrave prehranjevalne navade so eden najpogostejših vzrokov za CVD na svetu in se soočajo z njimi na različne načine, vključno z digitalnimi tehnologijami. Cilj tega projekta je razviti spletno storitev, ki omogoča nevsiljiv način sledenja prehranjevalnim navadam in spremljanja dinamike vnosa hranil v primerjavi s tradicionalnim načinom sledenja "beleženje vsakega obroka". Sledili smo pristopu oblikovalsko-znanstvenega raziskovanja za razvoj spletne storitve. Spletna storitev uporablja algoritme semantične podobnosti in omogoča človeško povratno informacijo za izboljšanje njegove učinkovitosti. Razvita bo z uporabo okvira Serverless in nameščena na platformi AWS. Prepoznavanje računov bo izvedeno z uporabo AWS Textract, hranilna vrednost izdelkov pa bo pridobljena iz javno dostopnih baz podatkov. Ta pristop si prizadeva ustvariti inovativen in uporabniku prijazen način sledenja prehranjevanju in dinamiki hranil. Healthy eating Grocery store receipt Nutritional analysis Daily intake monitoring machine learning chatbot zdrava prehrana račun živilske trgovine prehranska analiza nadzor dnevnega vnosa strojno učenje klepetalnik true true false Angleški jezik Slovenski jezik Magistrsko delo/naloga 2023-06-09 09:39:00 2023-09-14 13:20:08 2023-09-15 03:06:53 0000-00-00 00:00:00 2023 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 164451843 MAG_Nesterov_Pavel_2023.pdf MAG_Nesterov_Pavel_2023.pdf 1 D445339D1075FEB052542D70CD03FF67 5502b4c7a3ae177fb874af486c5b3ee7dd9b6be862fb8edc7af0482a39292a95 c1c2e8e3-091d-11ee-bc5d-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=167971 Fakulteta za organizacijske vede 0 0 0