Enhancing Online Food Service User Experience Through Advanced Analytics and Hybrid Deep Learning for Comprehensive Evaluation

IEEE Access, Год журнала: 2024, Номер 12, С. 70999 - 71009

Опубликована: Янв. 1, 2024

Язык: Английский

---

Logistics Shipping Based Blockchain Using Smart Contracts

Опубликована: Янв. 1, 2024

Язык: Английский

---

Multimodal Recipe Recommendation with Heterogeneous Graph Neural Networks

Electronics, Год журнала: 2024, Номер 13(16), С. 3283 - 3283

Опубликована: Авг. 19, 2024

Recipe recommendation is the process of recommending suitable recipes to users based on factors such as user preferences and dietary needs. Recipes typically involve multiple modalities, with text images being common, while most typical recipe methods recommend text. Obviously, expressiveness a single modal often not enough, semantic information more abundant. Moreover, it difficult grasp feature fusion granularity different kinds relationship between users. To solve above problem, this paper proposes Multimodal Heterogeneous Graph Neural Network Recommendation (MHGRR) architecture, which aims fully fuse various handle recipes. We use embedding shallow Convolutional Networks (CNNs) extract original image for unifying granularity, GraphSAGE capture complex verify effectiveness our proposed model, we perform some comparative experiments real dataset; show that method outperforms popular methods. Through an ablation experiment, found adding effective, additionally output dimensions increased, performance model varied little.

Язык: Английский

---

Optimization of Delivery Process Based on Machine Learning Support Vector Regression SVR Algorithm

Machine Learning Theory and Practice, Год журнала: 2020, Номер 1(2)

Опубликована: Июнь 16, 2020

In the current post-pandemic era, takeout delivery still plays an important role in our daily life.At same time, new challenge facing O2O is how to ensure safe and efficient of order within specified time process delivery.This paper mainly studies optimization based on machine learning support vector regression SVR algorithm.This proposes a regional demand prediction model algorithm.The can effectively predict each business area next hour, which provides basis for intelligent scheduling system platform.The data platform Dalian are used verify model, results compared with BP neural network GA experimental show that algorithm have better fitting effect, region.

Язык: Английский

---

Correction: Zhang, L.; Kim, D. A Peer-to-Peer Smart Food Delivery Platform Based on Smart Contract. Electronics 2022, 11, 1806

Electronics, Год журнала: 2022, Номер 11(17), С. 2673 - 2673

Опубликована: Авг. 26, 2022

In the original publication [...]

Язык: Английский

---