Data-driven recipe completion using machine learning methods

Abstract Background Completing recipes is a non-trivial task, as the success of ingredient combinations depends on a multitude of factors such as taste, smell and texture. Scope and approach In this article, we illustrate that machine learning methods can be applied for this purpose. Non-negative matrix factorization and two-step regularized least squares are presented as two alternative methods and their ability to build models to complete recipes is evaluated. The former method exploits information captured in existing recipes to complete a recipe, while the latter one is able to also incorporate information on flavor profiles of ingredients. The performance of the resulting models is evaluated on real-life data. Key findings and conclusions The two machine learning methods can be used to build models to complete a recipe. Both models are able to retrieve an eliminated ingredient from a recipe and the two-step RLS model is also capable of completing an ingredient set to create a complete recipe. By applying machine learning methods on existing recipes, it is not necessary to model the complexity of good ingredient combinations to be able to complete a recipe.

[1]  Albert-László Barabási,et al.  Flavor network and the principles of food pairing , 2011, Scientific reports.

[2]  Tapio Salakoski,et al.  A Kernel-Based Framework for Learning Graded Relations From Data , 2011, IEEE Transactions on Fuzzy Systems.

[3]  José Ramón Quevedo,et al.  How to learn consumer preferences from the analysis of sensory data by means of support vector machines (SVM) , 2007 .

[4]  Ricard Boqué,et al.  Application of non-negative matrix factorization combined with Fisher's linear discriminant analysis for classification of olive oil excitation-emission fluorescence spectra , 2006 .

[5]  Frank Westad,et al.  Strategies for consumer segmentation with applications on preference data , 2004 .

[6]  R. Margolskee,et al.  Making sense of taste. , 2001, Scientific American.

[7]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[8]  Nick Craswell.  Mean Reciprocal Rank , 2009, Encyclopedia of Database Systems.

[9]  Conor M. Delahunty,et al.  Mapping consumer preference for the sensory and packaging attributes of Cheddar cheese , 2000 .

[10]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[11]  R. Stevenson.  The Psychology of Flavour , 2009 .

[12]  D. Dubourdieu,et al.  The Color of Odors , 2001, Brain and Language.

[13]  Kush R. Varshney,et al.  Flavor Pairing in Medieval European Cuisine: A Study in Cooking with Dirty Data , 2013, ArXiv.

[14]  Kurt Hornik,et al.  kernlab - An S4 Package for Kernel Methods in R , 2004 .

[15]  Kazuaki Kishida.  Property of average precision and its generalization: An examination of evaluation indicator for information retrieval experiments , 2005 .

[16]  C. Spence,et al.  The multisensory perception of flavor , 2008, Consciousness and Cognition.

[17]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[18]  Janusz Sobecki,et al.  Application of Hybrid Recommendation in Web-Based Cooking Assistant , 2006, KES.

[19]  Mark A. Roe,et al.  Assessing and improving the quality of vitamin data in food composition databases , 2012, Food & nutrition research.

[20]  Charles X. Ling,et al.  Using AUC and accuracy in evaluating learning algorithms , 2005, IEEE Transactions on Knowledge and Data Engineering.

[21]  Patrik O. Hoyer,et al.  Non-negative Matrix Factorization with Sparseness Constraints , 2004, J. Mach. Learn. Res..

[22]  Martha R. Bajec,et al.  Influence of Stimulus Temperature on Orosensory Perception and Variation with Taste Phenotype , 2012, Chemosensory Perception.

[23]  D. Small,et al.  Odor/taste integration and the perception of flavor , 2005, Experimental Brain Research.

[24]  Nicholas Bakalar.  Sensory science: Partners in flavour , 2012, Nature.

[25]  Bart M. ter Haar Romeny,et al.  The Gaussian kernel , 2003 .

[26]  Bernard De Baets,et al.  A Two-Step Learning Approach for Solving Full and Almost Full Cold Start Problems in Dyadic Prediction , 2014, ECML/PKDD.

[27]  Sanjeev Arora,et al.  Computing a nonnegative matrix factorization -- provably , 2011, STOC '12.

[28]  John Prescott,et al.  Psychological processes in flavour perception , 2004 .

[29]  John D. Lafferty,et al.  Diffusion Kernels on Graphs and Other Discrete Input Spaces , 2002, ICML.

[30]  Shlomo Berkovsky,et al.  Intelligent food planning: personalized recipe recommendation , 2010, IUI '10.

[31]  R. Linforth,et al.  Oral shear stress predicts flavour perception in viscous solutions. , 2003, Chemical senses.

[32]  C. Spence,et al.  The multisensory perception of flavor: Assessing the influence of color cues on flavor discrimination responses , 2007 .

[33]  Youri van Pinxteren,et al.  Deriving a recipe similarity measure for recommending healthful meals , 2011, IUI '11.

[34]  Bernard De Baets,et al.  Efficient regularized least-squares algorithms for conditional ranking on relational data , 2012, Machine Learning.

[35]  B. Nilius,et al.  Influence of temperature on taste perception , 2007, Cellular and Molecular Life Sciences.

[36]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[37]  Robert A. Boakes,et al.  Sweet and sour smells : learned synesthesia between the senses of taste and smell , 2004 .

[38]  Max Feinberg,et al.  Extraction of Food Consumption Systems by Nonnegative Matrix Factorization (NMF) for the Assessment of Food Choices , 2011, Biometrics.

[39]  Mu Zhu,et al.  Content-boosted matrix factorization for recommender systems: experiments with recipe recommendation , 2011, RecSys '11.