Dr. k. Sundara Velrani, Juda Harith, Maria Mary Arlin, Julian Fernando, Manusri R, Melvin k
Abstract
Nutri Tracker is a smart mobile app created to tackle issues in personalized nutrition planning and effective food use. From the receipt text, system shall extract grocery item and be organized in the virtual pantry. By using user-defined health goals like weight management, skin health, hair growth, and PCOD-friendly nutrition, the app creates meal plans, quick recipes, and wise grocery purchase suggestions. This process utilizes the existing food sources efficiently, leads to a varied diet, and cuts down waste.. Nutri Tracker provide an intuitive, easy to use system for daily nutritional control by enabling automated, data-driven recommendations.Another important feature of the app helps users stay committed to their plan by sending personalized notifications, tracking the food they eat each day, and providing nutritional information that can assist them in addressing their health concerns and planning better meals for future days. The app also allows users to connect it with fitness applications, which helps enhance the plan by incorporating data on their physical activities
Keywords
personalized nutrition, virtual pantry, parse
grocery receipt, meal suggestion, PCOD meal plan, health app, health mobile application
Conclusion
Diet chart: – The first part of ensuring a good and a healthy life style is to adopt the right diet chart. A balanced diet will not only improve the body’s immune system but also keep it healthy throughout life. One of the few disadvantages of humans in today’s world is to plan proper nutritional diets for ourselves based on health status and the foods currently in use in the kitchen.
The reasons for this are primarily the rush in daily lives and a lack of awareness in the field of nutrition. To address these issues and to eat nutritional food and not produce any extra food (waste), several nutritional tracking tools and diet applications are available.
But most of these are for you to enter the foods that you eat, and none to suggest the food for you. More and more such tools exist, the only deficiency that the tool users face is that they either don’t support their required nutritional goals or these tools require a lot of input/manually entering foods you eat. In order to overcome this problem, a Smart Nutrition Tracker app, called the ‘Nutri Tracker’, will be designed.
This application will be a mobile platform to analyse grocery receipt text and help users manage their grocery food with the concept of a virtual pantry; in order to create personalised meals on the basis of desired nutritional goals and objectives (e.g., weight loss, healthy skin, hair growth, PCOD diet, etc).
The main purpose of the Nutri Tracker will be to convert grocery products purchased by users into a way where it can translate those into the users desired Nutritional Food and provide him with nutritional guidance and a meal plan that is easily manageable by any individual.
The primary objective of the proposed Nutri Tracker system will be to help users turn grocery information from receipt text to nutritional knowledge and simple, doable meal plans for achieving personal health objectives. The input for the proposed system will consist of the user’s grocery receipt text, and output will be based on the output for the personal health goals desired. For example, by taking into account receipt text, the system can establish a user’s virtual pantry; the user can choose his desired personal health objectives from various options like weight management, healthy skin, hair growth, etc. and also the PCOD diet.
After defining a desired objective the Nutri Tracker will first obtain the detailed nutritional information of all items purchased by the user from his grocery receipt text using a Structured Nutritional Database. Then it can generate various meal recommendations on the basis of the defined user objectives by using a developed Recommendation Engine that is built upon the analysis of a wide variety of nutritional criteria such as dietary restrictions, nutrient values, availability of ingredients from the user’s pantry, user’s taste preferences (optional) and more importantly defined health objectives. Apart from suggesting and recommending meals, application could also generate fast and simple recipes for e.g. Fast breakfast, fast salads, fast smoothie, fast lunch etc. Recipes have to be simple and fast to prepare so that user could prepare without requiring many time or cooking knowledge.
Another major novelty introduced by the present system will be its smart grocery shopping recommendation module that analyses potential dietary gaps based on the nutritional requirement criteria set for personal health objectives, based on the items available in the user’s pantry and provide recommendations on additional items required in the next grocery shopping trip for an optimized diet plan, thereby also avoiding any sort of unnecessary spending of money on ingredients which would then later be wasted due to non-use and unsuitability.
Major benefits of NutriTracker: – It provides useful practical outputs which will help the user to understand his/her body and nutrition for improving diet choices and making healthy eating simple. – It reduces the need of continuous user input as it identifies the food products automatically from the scanned grocery receipts.
Effective usage of all the groceries purchased. Decreases the loss of grocery items, and increases the grocery bill efficiency. Personalized recommendations for diet and meal plans and the provision of easy and quick recipes with nutritional details increase user engagement and facilitate adherence. The experimental evaluation proved the suitability of the system to extract useful data from grocery receipt text and utilize this information for generating tailored meal recommendations that complement varied personal health objectives. The system offers intuitive user interface and detailed dietary guidelines and recipes.
For future development, smart technologies like Optical Character Recognition (OCR) may be incorporated for fully automated scanning of grocery receipts, and further development of machine learning algorithms can be used to obtain more efficient recommendation models and enable real-time monitoring of user diet and physical activity for more advanced health and fitness tracking. In a Nutshell: The Smart Nutrition Tracker is the latest app of this kind which addresses several limitations of previously developed dietary management systems. Its capability of identifying grocery items directly from receipt text and providing users with the option to establish personal health goals and receive appropriate food recommendations makes it a convenient tool for personalized nutrition management..
References
[1]A. Fadhil and A. Schiavo, “Designing for health: A diet recommendation system for healthy eating,” IEEE International Conference on Digital Health, 2018.
[2]C. BindW. Min, S. Jiang, and R. Jain, “Food recommendation: framework, solutions and problems, in 2016 IEEE international conference on multimedia and expo (ICME), No. 10, PP. 2659-2671, 2020.
[3]D. Elsweiler, M. Harvey, and B. Ludwig, “Towards personalized nutrition recommendation systems,” Proceedings of the ACM International Conference on Intelligent User Interfaces, 2015.
[4]H. Chi, L. Chen, and Y. Wang, “A personalized nutrition recommendation system using health data and food preferences,” IEEE Access, 2020.
[5]H. Kagaya, K. Aizawa, and M. Ogawa, “Food detection and recognition using convolutional neural networks,” ACM Multimedia Conference, 2014.
[6]J. Yang, D. Nguyen, P. San, X. Li, and S. Krishnaswamy, “DeepFood: Deep learning-based food image recognition for computer-aided dietary assessment,” IEEE International Conference on Multimedia and Expo, 2016.
[7]K. Yanai and Y. Kawano, “Food image recognition using deep convolutional networks with pre-training and fine-tuning,” IEEE International Conference on Multimedia and Expo Workshops, 2015.
[8]L. Meyers, A. Johnston, and J. Rathod, “Im2Calories: Towards automated mobile vision food dietary assessment,” IEEE International Conference on Computer Vision, 2015.
[9]M. Harvey and D. Elsweiler, “Context-aware food recommendation systems,” Proceedings of the ACM Conference on Recommender Systems Workshops, 2016.
[10]N. Bossard, M. Guillaumin, and L. Van Gool, “Food-101: Mining discriminative components with random forests,” European Conference on Computer Vision, 2014.
[11]R. Zhu, J. Yu, and H. Li, “Smart nutrition recommendation system based on user health data,” IEEE Access, 2020.
[12]S. Achananuparp, E. P. Lim, and J. Jiang, “Health recommendation systems: A survey,” IEEE Transactions on Knowledge and Data Engineering, 2016.
[13]S. S. Ge, C. Chen, and Y. Chen, “Mobile-based diet recommendation system for personalized health management,” IEEE Access, 2019.
[14]S. Subramanian and R. Subramanian, “Machine learning approaches for personalized diet recommendation systems,” IEEE International Conference on Smart Computing, 2019.
[15]Y. Kawano and K. Yanai, “Food image recognition with deep convolutional features,” Proceedings of the ACM International Conference on Multimedia, 2014.
How to Cite This Paper
Dr. k. Sundara Velrani, Juda Harith, Maria Mary Arlin, Julian Fernando, Manusri R, Melvin k (2026). Nutri-tracker: A Personalized Nutrition Advisory System. International Journal of Computer Techniques, 13(3). ISSN: 2394-2231.
