As the informationized trend is developing toward intelligentized one in catering industry, the targeted ordering recommendation provide for customer based on the technique of intelligent recommendation has turned into reality. Applying in the scenario of traditional Chinese food service, a heuristic and hybrid model in ordering recommendation is proposed. First, the association rules algorithm was adopted for obtaining the association rules of historically associated dishes combination and calculating their correlation degree. Second, applying the recommended algorithm based on dishes attributes to calculate the similarities of dishes in the database. Third, to calculate comprehensive scores of dishes and create the recommendation rules in accordance with their correlation degree and similarities. Finally, a recommended order list is shaped from both the dishes ordered and the recommendation rules concluded in the former step. The effectiveness and validation of the model and algorithm are being proved by real order data in Chinese restaurants. The data shows the model is better than traditional association rules in the aspects of recommendation precision and coverage when the dishes ordered reaching to certain amount.
[1]
Simone Daminelli,et al.
Common neighbours and the local-community-paradigm for topological link prediction in bipartite networks
,
2015,
ArXiv.
[2]
Anton Nijholt,et al.
Case-Based Reasoning as a Prediction Strategy for Hybrid Recommender Systems
,
2004,
AWIC.
[3]
Jingbo Yuan,et al.
Research and Improvement on Association Rule Algorithm Based on FP-Growth
,
2012,
WISM.
[4]
Paul Resnick,et al.
Recommender systems
,
1997,
CACM.
[5]
Michael J. Pazzani,et al.
Content-Based Recommendation Systems
,
2007,
The Adaptive Web.
[6]
Philip Sedgwick,et al.
Pearson’s correlation coefficient
,
2012,
BMJ : British Medical Journal.
[7]
Robin D. Burke,et al.
Hybrid Recommender Systems: Survey and Experiments
,
2002,
User Modeling and User-Adapted Interaction.
[8]
Dan Frankowski,et al.
Collaborative Filtering Recommender Systems
,
2007,
The Adaptive Web.
[9]
Mohammadreza Radmanesh,et al.
Hybrid Recommender Systems based on Content Feature Relationship
,
2016
.