2022
DOI: 10.33395/sinkron.v7i2.11383
|View full text |Cite
|
Sign up to set email alerts
|

Compare VGG19, ResNet50, Inception-V3 for Review Food Rating

Abstract: The food industry is undergoing a phase of very good improvement, where business actors are experiencing very rapid growth. Creative ideas are many and creative on several social media. When an online business is growing rapidly, many managers in the food sector market their products through online media. So it is quite easy for customers to place orders via mobile. Especially during the COVID-19 pandemic, where a ban on gatherings has become a government recommendation for many food business actors to sell on… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
7
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 8 publications
(7 citation statements)
references
References 25 publications
0
7
0
Order By: Relevance
“…ResNet50, short for residual networks, is a 50-layer model that uses the concept of skipping levels to reduce overfitting, solve the problem of vanishing gradient, and ensure that higher layers function as well as lower ones [22]. Figure 3 shows the detailed architecture of the ResNet50 [23]. MobileNetV2 incorporates a unique layer module, the inverted residual with linear bottleneck, greatly decreasing the amount of memory required for processing [26].…”
Section: Model Architecturementioning
confidence: 99%
“…ResNet50, short for residual networks, is a 50-layer model that uses the concept of skipping levels to reduce overfitting, solve the problem of vanishing gradient, and ensure that higher layers function as well as lower ones [22]. Figure 3 shows the detailed architecture of the ResNet50 [23]. MobileNetV2 incorporates a unique layer module, the inverted residual with linear bottleneck, greatly decreasing the amount of memory required for processing [26].…”
Section: Model Architecturementioning
confidence: 99%
“…Due to its success, CNN has been implemented for some purposes. Andrew and Santoso [13] used three models, namely VGG19, ResNet50, and Inception-V3, to analyze the rating of online food orders. Their research found that the accuracy of VGG19 is 96.86, Resnet50 is 97.29, and Inception_v3 is 97.57.…”
Section: Literatures Reviewmentioning
confidence: 99%
“…It is based on the original paper "Rethinking the Inception Architecture for Computer Vision" [11]. The steps of the Inception process are convolution, pooling, dropout, fully connected, and softmax [12,13].…”
Section: Proposed Strategymentioning
confidence: 99%