Ford Motor Company E-Commerce Development on Social Networks based on HED Feature Extraction, Machine Learning and Alan Mulally’, s Strategy

Q: How can I download an article?

To download an article from SID, first log in to the site, search for the article title, and click on the 'Download Article' option.

Q: How can I download an ISI article?

To download an ISI article on SID, enter the keyword or article title in the search bar, view the relevant results, click on the desired article, and select the 'Download Article' option.

Q: How can I access the SID database?

To access the SID database, visit SID.ir, create an account, and log in to access scientific resources.

Q: Is downloading articles from SID free?

Some articles on SID are available for free, while others require payment. Details are specified on the article's page.

Barshooi Amir Hossein

مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

Seminar Paper

Paper Information

Seminar: INTERNATIONAL CONFERENCE ON WEB RESEARCH
Year:2023 | Issue:9

Download Full-Text

View:

Download:

Information Seminar Paper

Title

Ford Motor Company E-Commerce Development on Social Networks based on HED Feature Extraction, Machine Learning and Alan Mulally’, s Strategy

Author(s)

Barshooi Amir Hossein

Keywords

E-Commerce

Consumer Identification

Ensemble Learning

Classification

Vehicle Part Localization

Heuristic-nested Feature Extraction

Abstract

In this article, a multi-stage E-Commerce platform based on deep learning methods is presented to improve Ford Motor Company's consumer engagement on the Instagram social network. In the first stage, Instagram accounts are crawled and consumers are identified based on comments, hashtags, and followers. With the help of a network built on Darknet as its backbone, the users’,vehicles are detected in the next stage from any shared content. It is in the third stage where MobileNetV2, VGG19, DenseNet121, ShuffleNetV2, and InceptionV3 networks are used to recognize the model and year of each vehicle, and then a voting process takes place to determine the final prediction. In the last step, by applying the provided landmark detection module, all components such as headlights and taillights are localized. In order to improve the robustness and performance of the proposed approach, a Heuristic-nested Feature Extraction block has been embedded at the beginning of each stage and all networks are trained on a dedicated dataset of 16K images in 8 different classes. Based on the results, the vehicles were detected with a mean average precision (mAP) of 72%, and the recognition of each model and year was performed with the accuracy of 98. 14%, precision of 98. 13%, recall of 98. 14%, specificity of 99. 73%, F1-score of 98. 13%, and MCC of 0. 98. The least and the most value of normalized errors (NE) was obtained 0. 0335 and 0. 0504, which corresponds to the hood and right side mirror in the landmark module, respectively.

Cites

No record.

References

No record.

Cite

APA: Copy

Barshooi, Amir Hossein. (). . . SID. https://sid.ir/paper/1046835/en

Vancouver: Copy

Barshooi Amir Hossein. . . Available from: https://sid.ir/paper/1046835/en

IEEE: Copy

Amir Hossein Barshooi, “,” presented at the . , [Online]. Available: https://sid.ir/paper/1046835/en

Related Journal Papers