PORDE: Explaining Data Poisoning Attacks Through Visual Analytics with Food Delivery App Reviews

• DOI: 10.1145/3581754.3584128
• Publication Date: 2023-03-26T22:12:25Z
• Authors: Hyunmin Lee; Minki Chun; Hyunggu Jung

Artificial intelligence (AI) gives many benefits to our lives. However, biased AI models created by receiving data poisoning attacks may induce social problems. Therefore, developers must consider carefully whether the training received a poison attack when developing an model. Data visualization is one of methods facilitate analysis required for checking if attack. prior studies did not visualize real-world data. Restaurant reviews in delivery apps are cases poisoned dataset. Restaurants hold review events on encourage customers write positive return certain rewards, thereby creating with bias. In this study, we propose POisoned Real-world Explainer (PORDE) that explains through visual analytics food app reviews. The findings study suggest implications securing safe and less models.