IOTA-Based Mobile Crowd Sensing: Detection of Fake Sensing Using Logit-Boosted Machine Learning Algorithms

In the Internet of Things (IoT) era, mobile crowd sensing system (MCS) has become increasingly important. The Auto (IOTA) evolved rapidly in practically every technology field over last decade. IOTA-based is being developed this study using machine learning to detect and prevent users from engaging fake activities. It been determined through testing evaluation that our method effective for both quality estimation incentive allocation. Using IOTA Bottleneck dataset, multiple performance metrics were used demonstrate how well logit-boosted algorithms perform. After applying on dataset classification, Logi-XGB scored 95.7 percent accuracy, while Logi-GBC 90.8 accuracy. As a result this, Logi-ABC had an accuracy rate 89%. Logi-CBC, other hand, got highest 99.8%. Logi-LGBM Logi-HGBC 91.37 which identical. On given Logi-CBC algorithm outperforms earlier Logit-boosted terms new IoTA-Botnet 2020 proposed methodology tested. comparison prior algorithms, model detection