2024
DEVELOPMENT OF MOBILE APPLICATION FOR CUISINE RECIPE GUIDE USING IMAGE CLASSIFICATION METHOD BASED ON ANDROID
J-PTIIK · Journal Publication
The cooking process is a way to create food by going through sequential stages. Based on a survey conducted to 60 people aged 17 years and over, it was found that 93.3% of respondents have their own desires in cooking. Therefore, 93.3% of respondents feel that they need an application that can find recipes using camera shots based on ready-to-eat food. Therefore, a recipe guide application was developed using the image classification method. This android-based application was developed using SDLC Prototyping and system design consisting of ktor for the server side, android native kotlin on the android side, and tensorflow to develop machine learning models. Application testing with black box testing gives perfect results, namely 100%, which means that all functionality works properly. Usability testing also produces an effectiveness rate of 99.3%. Usability testing is supported by SUS which gets a score of 92 which means the application is acceptable to users, is at grade A, and has the best imaginable rating. Machine learning model testing on 10 food classes gave excellent accuracy results of 94%. Based on these results, it can be concluded that the application can be easily accepted and used by users.
