MULTI-LAYER FOOD RECOMMENDATION SYSTEMS
Dr. S. V. G. Reddy1, Sirisha Majji2, K. Jasvanth2, G Jeevan Reddy2, G. Vamshi Yadav2
1Associate Professor, Department of Computer Science and Engineering, GITAM University, Visakhapatnam, Andhra Pradesh, 530045, India.
2Student, Department of Computer Science and Engineering, GITAM University, Visakhapatnam, Andhra Pradesh, 530045, India.
ABSTRACT
Recommendation Systems are one of the most popular applications of Machine Learning technology. They are information filtering systems that provide us with valuable suggestions based on our personal preferences and a variety of other factors. Finding one's favourite food from a variety of food items and dishes has become an important issue as one may not know what food to eat at that particular time or is unsure which item to prefer. So, this food recommendation system uses collaborative filtering and a Natural Language Processing algorithm to assist the user by recommending food items based on their previous preferences and ratings on various food items and suggests food items with low spice. There are numerous machine learning techniques available, such as collaborative-based filtering, content-based filtering, and a combination of content and collaborative based filtering known as hybrid filtering, knowledge-based filtering, and so forth, which can be used to implement the recommendation systems. Among all of these methods or technologies, we will be dealing with User-based filtering, which is a type of Collaborative based filtering in which we calculate the similarity coefficients of different users based on the inputs given, and then predictions are made. After the predictions, we will again make recommendations from the recommended list based on user preferences for the cuisine, and again from the recommended list, we will recommend the food items having low spice to the user. The main or primary aim of this proposed recommender system is to increase user satisfaction, and this application is expected to have a high accuracy of 90%.
Keywords- Recommendation systems, food recommendations, Natural Language Processing, predictions.