Detecting Transaction-based Tax Evasion Activities on Social Media Platforms Using Multi-modal Deep Neural Networks
This provides a tool for international tax authorities to combat social e-commerce tax evasion, though it is incremental as it applies existing multi-modal methods to a new domain.
The paper tackled the problem of detecting transaction-based tax evasion on social media platforms by developing a multi-modal deep neural network, which achieved an AUC of 0.808 and F1 score of 0.762 on a dataset of Instagram posts.
Social media platforms now serve billions of users by providing convenient means of communication, content sharing and even payment between different users. Due to such convenient and anarchic nature, they have also been used rampantly to promote and conduct business activities between unregistered market participants without paying taxes. Tax authorities worldwide face difficulties in regulating these hidden economy activities by traditional regulatory means. This paper presents a machine learning based Regtech tool for international tax authorities to detect transaction-based tax evasion activities on social media platforms. To build such a tool, we collected a dataset of 58,660 Instagram posts and manually labelled 2,081 sampled posts with multiple properties related to transaction-based tax evasion activities. Based on the dataset, we developed a multi-modal deep neural network to automatically detect suspicious posts. The proposed model combines comments, hashtags and image modalities to produce the final output. As shown by our experiments, the combined model achieved an AUC of 0.808 and F1 score of 0.762, outperforming any single modality models. This tool could help tax authorities to identify audit targets in an efficient and effective manner, and combat social e-commerce tax evasion in scale.