Multi-Field Models in Neural Recipe Ranking - An Early Exploratory Study

Explicitly modelling field interactions and correlations in complex documents structures has recently gained popularity in neural document embedding and retrieval tasks. Although this requires the specification of bespoke task-dependent models, encouraging empirical results are beginning to emerge. We present the first indepth analyses of non-linear multi-field interaction (NL-MFI) ranking in the cooking domain in this work. Our results show that field-weighted factorisation machines models provide a statistically significant improvement over baselines in recipe retrieval tasks. Additionally, we show that sparsely capturing subsets of field interactions offers advantages over exhaustive alternatives. Although field-interaction aware models are more elaborate from an architectural basis, they are often more data-efficient in optimisation and are better suited for explainability due to mirrored document and model factorisation.

[1]  Stephen E. Robertson,et al.  GatfordCentre for Interactive Systems ResearchDepartment of Information , 1996 .

[2]  Donald Eugene. Farrar,et al.  Multicollinearity in Regression Analysis; the Problem Revisited , 2011 .

[3]  W. Bruce Croft,et al.  A Probabilistic Retrieval Model for Semistructured Data , 2009, ECIR.

[4]  Jane Greenberg,et al.  Using BM25F for semantic search , 2010, SEMSEARCH '10.

[5]  James P. Callan,et al.  Combining document representations for known-item search , 2003, SIGIR.

[6]  Vaishali S. Vairale,et al.  Recommendation of Food Items for Thyroid Patients Using Content-Based KNN Method , 2020 .

[7]  Dennis M. Wilkinson,et al.  Large-Scale Parallel Collaborative Filtering for the Netflix Prize , 2008, AAIM.

[8]  Yoav Shoham,et al.  Fab: content-based, collaborative recommendation , 1997, CACM.

[9]  Michael J. Pazzani,et al.  Content-Based Recommendation Systems , 2007, The Adaptive Web.

[10]  Benjamin Piwowarski,et al.  A Machine Learning Model for Information Retrieval with Structured Documents , 2003, MLDM.

[11]  Bhaskar Mitra,et al.  Neural Ranking Models with Multiple Document Fields , 2017, WSDM.

[12]  Larry P. Heck,et al.  Learning deep structured semantic models for web search using clickthrough data , 2013, CIKM.

[13]  R. Logesh,et al.  Exploring Hybrid Recommender Systems for Personalized Travel Applications , 2018, Cognitive Informatics and Soft Computing.

[14]  Christopher J. C. Burges,et al.  A machine learning approach for improved BM25 retrieval , 2009, CIKM.

[15]  Fei-Fei Li,et al.  ImageNet: A large-scale hierarchical image database , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[16]  W. Bruce Croft,et al.  A Field Relevance Model for Structured Document Retrieval , 2012, ECIR.

[17]  Pasquale Lops,et al.  Content-based Recommender Systems: State of the Art and Trends , 2011, Recommender Systems Handbook.

[18]  Yu Sun,et al.  Field-weighted Factorization Machines for Click-Through Rate Prediction in Display Advertising , 2018, WWW.

[19]  J. Shane Culpepper,et al.  Improving Search Effectiveness with Field-based Relevance Modeling , 2018, ADCS.

[20]  Choon Hui Teo,et al.  Semantic Product Search , 2019, KDD.

[21]  Stephen E. Robertson,et al.  Simple BM25 extension to multiple weighted fields , 2004, CIKM '04.

[22]  Thomas Lengauer,et al.  Classification with correlated features: unreliability of feature ranking and solutions , 2011, Bioinform..