Skip to main content
SpringerLink
Log in

Automated and personalized meal plan generation and relevance scoring using a multi-factor adaptation of the transportation problem

  • Application of soft computing
  • Published:
Soft Computing Aims and scope Submit manuscript

Abstract

Establishing a healthy lifestyle has become a very important aspect in people’s lives. The latter requires maintaining a healthy nutrition by considering the nature and quantity of foods being consumed, allowing to regulate one’s intake and consumption of calories and nutrients. As a result, people reach out for nutrition experts which services are costly, time-consuming, and not readily available. While various e-solutions have been developed to perform meal planning, yet most of them lack a completely automated process and require domain expert intervention at different stages of the recommendation process (e.g., identifying macronutrient distribution, providing pre-defined meal plans, or combining recommended foods into meal structures). In addition, most solutions focus on fulfilling the patients’ nutrition requirements (in terms of caloric intake and macronutrients) while disregarding other relevant factors such as patient food preferences, food variety, food-meal compatibility, and inter-food compatibility. Hence, there is a need for an automated solution to produce a full-fledged meal plan from scratch, based on a recommended caloric intake and considering multiple factors. In this study, we introduce a novel solution titled MPG for automated Meal Plan Generation recommendations, designed based on an adaptation of the transportation optimization problem to simulate the “human thought process” involved in generating daily meal plans. MPG allows to: (i) generate plans which fulfill a recommended caloric intake, given a set of available foods, while (ii) personalizing the plans following patient chosen factors (e.g., food preferences, variety, and compatibility), and (iii) evaluating the relevance of the produced plans following patient preferences. We have conducted various experiments involving 9 human testers and 124 meal plans to test the performance of MPG. Results highlight MPG’s effectiveness in producing “healthy” and personalized meal plans while complying with the testers’ preferences.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

taken from our running example in Fig. 3

Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

Data availability

The datasets generated and analyzed during the current study are described in this published thesis report (Salloum G. and Tekli J., 2020). They are also available from the authors on reasonable request.

Code availability

A software demo and an executable version of the prototype are available at the following link: http://sigappfr.acm.org/Projects/PIN/.

Notes

  1. The Body Fat Percentage (BFP) is computed as the ratio of the patient’s body fat weight over the total body weight. It is a common and expressive metric used in nutrition health practice (cf. Section 2).

  2. Our solution supports any number of daily meals. In the current study, we adopt a 5-meal plan which is typically adopted in health nutrition literature.

  3. Adopted based on the Diabetic Exchange List suggested by the American Dietetic Association (Kathleen and Janice 2017).

  4. Type-2 FL is an extension of the original FL paradigm, referred to as type-1 FL, where every truth degree has an uncertainty degree associated with it (e.g., a person is considered 0.3 underweight with 0.9 certainty, i.e., we are 90% certain that the person is 30% overweight). If there is no uncertainty, then a type-2 fuzzy set is reduced to a type-1 fuzzy set (Karnik and Mendel 2001).

  5. With respect to.

  6. Adopted based on the Diabetic Exchange List suggested by the American Dietetic Association (Kathleen and Janice 2017).

  7. An early version of MPG’s transportation optimization solution is mentioned in (Salloum et al. 2018), where the authors consider: i) basic food items only (the current study introduces a new model integrating both basic and composite foods, cf. Sect. 4.4), ii) the traditional transportation problem only (the present paper introduces a new multi-factor adaptation of the transportation problem, specifically designed to handle composite foods, cf. Sect. 4), iii) pre-defined static food-meal cost values defined by experts (the present paper introduces a dynamic approach consisting of a battery of novel mathematical cost functions – cf. Sect. 4.4.3 – and meal plan evaluation functions – cf. Sect. 4.5).

  8. Compared with typical transportation problem formulations where demands are represented as 1-dimentional scalar values.

  9. Compared with traditional transportation problem formulations where every demand center has one single requirement from a given supply center.

  10. In the formula, we represent the vector as its transpose for ease of presentation.

  11. The food compatibility graph was developed with the help of Dr. Maya Bassil (Associate Professor of Human Nutrition in the Department of Natural Sciences at LAU) and Ms. Eva-Maria Kahwaji (M.Sc. in Sports and Exercise Nutrition at Loughborough University).

  12. Note that Dist(i, i’) = 0 will never occur in our computations since it amounts to comparing a food with itself.

  13. This is different from having static cost values that remain unchanged throughout the whole computation process of typical transportation problem solutions.

  14. Recall that that \(\overrightarrow{{D}_{j}}\) = (d1, d2, d3, d4, d5, d6) represents a 6-dimentional vector where every dimension corresponds to one of the 6 categories of basic foods considered in our study, cat1-to-cat6 (i.e., starch, fruits, milk, vegetables, lean meat, and fat).

  15. MPG can generate multiple different MPMax solutions for the same cost factor, considering the nature of our computation process. Hence, we perform multiple runs for every individual factor separately, compute the score of each produced MPMax solution in each run, and then average them out to produce score(MPMax).

  16. The scores for the best possible meal plans are computed experimentally, considering a pool of 15 experimental runs where the produced score for every individual cost factor is computed as the average of the maximum scores obtained in every experimental run (cf. Sect. 5).

  17. This paper describes MPG, while PIN’s remaining modules are developed in (Salloum G. and Tekli J., 2020).

  18. http://sigappfr.acm.org/Projects/PIN/.

  19. The CI requirement cases considered in this experiment are chosen based on common practices in health nutrition literature (Kathleen and Janice 2017).

  20. Note that the food preference criterion does not reflect meal plan healthiness, and is evaluated by non-expert testers in Experiment 2 (Sect. 5.2).

References

  • Abboud R, Tekli J (2018) MUSE prototype for music sentiment expression.In: IEEE international conference on cognitive computing (ICCC'18), part of the IEEE world congress on services 2018. pp 106–109

  • Abboud R, Tekli J (2019) Integration of non-parametric fuzzy classification with an evolutionary-developmental framework to perform music sentiment-based analysis and composition. Springer Soft Comput. https://doi.org/10.1007/s00500-019-04503-4

    Article  Google Scholar 

  • Ainsworth B, Haskell W, Herrmann S, Meckes BDJN, Tudor-Locke C, Greer VJJ, Whitt-Glover M, Leon A (2011) 2011 compendium of physical activities: a second update of codes and MET values. Med Sci Sports Exerc 43(8):1575–1581

    Article  Google Scholar 

  • Ayoub J, Samra A, Hlais A, Bassil M, Obeid O (2015) Effect of phosphorus supplementation on weight gain and waist circumference of overweight/obese adults: a randomized clinical trial. Nutr Diabetes 5(12):e189

    Article  Google Scholar 

  • Bassi L (1976) The diet problem revisited. Am Econ 20(2):35–39

    Google Scholar 

  • Bianchi L, Dorigo M, Gambardella L, Gutjahr W (2009) A Survey on metaheuristics for stochastic combinatorial optimization. Nat Comput 8(2):239–287

    Article  MathSciNet  Google Scholar 

  • Cormen TH, Leiserson CE, Rivest RL (2009) Introduction to algorithms, 3rd edn. MIT Press and McGraw-Hill, New York

    MATH  Google Scholar 

  • Deb K, Jain H (2014) An evolutionary many-objective optimization algorithm using reference-point-based non-dominated sorting approach, Part I: solving problems with box constraints. IEEE Trans Evol Comput 18(4):577–601

    Article  Google Scholar 

  • El-Ghazi T (2009) Metaheuristics: from design to implementation. Wiley, New Jersey, p 624

    Google Scholar 

  • Evans D (2017) MyFitnessPal. Br J Sports Med 51(14):1101–1102

    Article  Google Scholar 

  • EatThisMuch Inc. (2020) Eat this much. Available: https://www.eatthismuch.com/ Accessed 03 Apr 2020.

  • Fahmi A, Abdullah S, Amin F (2019) Trapezoidal cubic fuzzy number einstein hybrid weighted averaging operators and its application to decision making. Soft Comput 24(14):5753–5783

    Article  Google Scholar 

  • Fister D, Fister RSI (2016) Generating eating plans for athletes using the particle swarm optimization. In: IEEE Int. Symp. Comput. Intell. Informatics (CINTI'16). pp 193–198

  • Fitness Meal Planner (2020) Fitness meal planner. Available: http://www.fitnessmealplanner.com/, Accessed: 03 Ap 2020

  • Foytik J (1981) Devising and using a computerized diet: an exploratory study. J Consum Aff 15(1):158–169

    Article  Google Scholar 

  • Gaál B, Vassányi I, Kozmann G (2005) A novel artificial intelligence methodfor weekly dietary menu planning. Methods Inf Med 44:655–664

    Article  Google Scholar 

  • Hall KD et al (2011) Quantification of the effect of energy imbalance on bodyweight. Lancet 378(9793):826–837

    Article  Google Scholar 

  • Hira D, Gupta P (2014) Operations research. S. Chand & Company Ltd.

  • Husain W, Wei L, Cheng ZNS (2011) Application of data mining techniques in a personalized diet recommendation system for cancer patients. In: 2011 IEEE colloquium on humanities, science and engineering. IEEE, pp 239–244

  • MakeMyPlate Inc. (2020) Make my plate. Available: http://www.makemyplate.co/ Accessed 01 Apr 2020

  • MyNetDiary Inc. (2020). MyNetDiary. Available: ttps://www.mynetdiary.com/ Accessed 01 Apr 2020

  • Kaldirim E, Kose Z (2006) Application of a multi-objective genetic algorithm to the modified diet problem. In: Genetic and evolutionary computation conf. pp 10–13

  • Karnik N, Mendel J (2001) Operations on type-2 fuzzy sets. Fuzzy Sets Syst 122:327–348

    Article  MathSciNet  Google Scholar 

  • Kathleen M, Janice R (2017) Food & the nutrition care process, 14th edn. Saunders, p 1152

  • Khan AS, Hoffmann A (2003a) An advanced artificial intelligence tool for menu design. Nutr Health J 17(1):43–53

    Article  Google Scholar 

  • Khan AS, Hoffmann A (2003b) Building a case-based diet recommendation system without a knowledge engineer. Elsevier J Artif Intell Med 27(2):155–179

    Article  Google Scholar 

  • Livestrong Foundation (2020) Myplate. Available: https://www.livestrong.com/myplate/ Accessed 1 Apr 2020

  • Lancaster L (1992) The history of the application of mathematical programming to menu planning. Eur J Oper Res 57(3):339–347

    Article  MathSciNet  Google Scholar 

  • Lee C, Lan S (2015) Adaptive personalized diet linguisti recommendation mechanism based on type-2 fuzzy sets and genetic fuzzy markup language. EEE Trans Fuzzy Syst 23(5):1777–1802

    Article  Google Scholar 

  • Lee C, Wang M, Hagras H (2010a) A type-2 fuzzy ontology and its application to personal diabetic-diet recommendation. IEEE Trans Fuzzy Syst 18(2):374–395

    Google Scholar 

  • Lee C, Wang M, Habras H, Chen Z, Lan S, Hsu C, Kuo S, Kuo H, Cheng H (2012) A novel genetic fuzzy markup language and its application to healthy diet assessment. Int J Uncertain Fuzziness Knowl-Based Syst 20:247–278

    Article  Google Scholar 

  • Lee C, Wang M, Li H, Chen W (2008) Intelligent ontological agent for diabetic food recommendation. In: IEEE international conference on fuzzy systems. pp 1803–1810

  • Lee C, Wang M, Hsu C, Hagras H (2009) A novel type-2 fuzzy ontology and its application to diet assessment. In: IEEE/WIC/ACM international joint conference on web intelligence and intelligent agent technology, vol 3, pp 417–420

  • Lee CS, Wang MH, Acampora G, Hsu CY, Hagras H (2010). Diet assessment based on type-2 fuzzy ontology and fuzzy markup language. international journal of intelligent systems. pp 1187–1216

  • Lenstra H (1983) Integer programming with a fixed number of variables. Math Oper Res 8(4):538–548

    Article  MathSciNet  Google Scholar 

  • Mahan L, Escott-Stump S, Raymond L (2012) Krause’s food & the nutrition care process. Saunders, Philadelphia, p 1248

    Google Scholar 

  • Mattar L, Zeeni N, Bassil M (2015) Effect of movie violence on mood, stress, appetite perception and food preferences in a random population. Eur J Clin Nutr 69(8):972

    Article  Google Scholar 

  • McCann-Rugg M, White G, Endres J (1983) Using goal programming to improve the calculation of diabetic diets. Comput Oper Res 10(4):365–373

    Article  Google Scholar 

  • Noor S, Mohd A, Ruhana K, Mahamud K, Arbin N (2018) Self-adaptive hybrid genetic algorithm (SHGA). Far East J Math Sci 103:171–190

    Google Scholar 

  • Petot GJ, Marling C, Sterling L (1998) An artificial intelligence system for computer-assisted menu planning. J Am Diet Assoc 98(9):1009–1014

    Article  Google Scholar 

  • Rolls B (2009) The relationship between dietary energy density and energy intake. Physiol Behav 97(5):609–615

    Article  Google Scholar 

  • Salameh K, Tekli J, Chbeir R (2014) SVG-to-RDF Image Semantization. In: International conference on similarity search and applications (SISAP'14). pp 214–228

  • Salloum G, Tekli J (2020) Automated knowledge-based nutrition health assessment, recommendation, progress evaluation, and meal planning. Master's Thesis Report, LAU Lib., https://laur.lau.edu.lb:8443/xmlui/bitstream/handle/10725/10468/George_Kamil_Salloum_Thesis_Redacted.pdf

  • Salloum G, Tekli J (2021) Automated and personalized nutrition health assessment, recommendation, and progress evaluation using fuzzy reasoning. Elsevier Int J Human-Comput Studs (IJHCS) 151:102610

    Article  Google Scholar 

  • Salloum G, Semaan E, Tekli J (2018) PIN prototype for intelligent nutrition assessment and meal planning. In: IEEE internaitonal conference on cognitive computing (ICCC'18) - Part of the IEEE congress on services. pp 110–113

  • Schniederjans M (1995) Goal programming: methodology and applications, 1st edn. Springer, New York

    Book  Google Scholar 

  • Seljak B (2009) Computer-based dietary menu planning. J Food Compos Anal 22(5):414–420

    Article  Google Scholar 

  • Silberberg E (1985) Nutrition and the demand for tastes. J Polit Econ 93(5):881–900

    Article  Google Scholar 

  • SparkPeople Inc. (2020). SparkPeople. Available https://www.sparkpeople.com/resource/mix-and-match-meal-planner.asp Accessed 01 Apr 2020

  • Stigler G (1945) The cost of subsistence. J Farm Econ 27(2):303–314

    Article  Google Scholar 

  • Suksom N, Buranarach M (2010) A knowledge-based framework for development of personalized food recommender system. In: Proc. of the 5th Int. Conf. on knowledge, information and creativity support systems. p 2

  • Türkmenoglu C, Etaner-Uyar AS, Kiraz B (2021) Recommending healthy meal plans by optimising nature-inspired many- objective diet problem. Health Inf J 2127(1):146045822097671

    Article  Google Scholar 

  • Valdez-Pena H, Martinez-Alfaro H (2003) Menu planning using the exchange diet system. In: Int. Conf. on Systems, Man and Cybernetics, vol 3, pp 3044–3049

  • Wang MH (2009) Intelligent ontological multi-agent for healthy diet planning. In: IEEE Inter. Conf. on Fuzzy Systems doi:https://doi.org/10.1109/FUZZY.2009.5277049

  • Winston W, Venkataramanan M (2003) Introduction to mathematical programming: operations research volume. Thomson Learning, 4th Edn. p. 924

  • Yang L, Hsieh CK, Yang H, Dell N, Belongie S, Cole C, Estrin D (2017) Yum-Me: a personalized nutrient-based meal recommender system. ACM Trans Inf Syst 36(1):7

    Article  Google Scholar 

Download references

Acknowledgements

We would like to thank all nutritionists who volunteered to participate in this study, namely: Dr. Maya Bassil (Associate Professor of Human Nutrition in the Department of Natural Sciences at LAU), and Ms. Eva-Maria Kahwaji (M.Sc. in Sports and Exercise Nutrition at Loughborough University), for their help in preparing the food compatibility graph, as well as Ms. Haneen Boughanem (Licensed Dietitian), Mr. Omar Makki (Research Assistant in Nutrition and Dietetics Coordinated Program, Natural Sciences Department, LAU), Ms. Fatima Kawtharani (M.Sc. in Human Nutrition and UNICEF field worker), and Ms. Rym Kalo (Licensed dietitian) for participating in the meal plan assessment tests.

Funding

This study is partly funded by the National Council for Scientific Research (CNRS-L) – Lebanon, and the Lebanese American University (LAU).

Author information

Authors and Affiliations

  1. E.C.E. Department, School of Engineering, Lebanese American University (LAU), Byblos Campus, 36, Byblos, Lebanon

    George Salloum & Joe Tekli

Authors
  1. George Salloum
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Joe Tekli
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Joe Tekli.

Ethics declarations

Conflict of interest

Not Applicable.

Human or animal rights

Statement of human rights: Ethical approval: For this type of study formal consent is not required. Statement on the Welfare of Animals: Ethical approval: This article does not contain any studies with animals performed by any of the authors.

Informed consent

Additional informed consent was obtained from all individual participants for whom identifying information is included in this article.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Salloum, G., Tekli, J. Automated and personalized meal plan generation and relevance scoring using a multi-factor adaptation of the transportation problem. Soft Comput 26, 2561–2585 (2022). https://doi.org/10.1007/s00500-021-06400-1

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00500-021-06400-1

Keywords

Search

Navigation