Paper Title: Anomaly Perception Data and Computation For Further Utilization

Author's name: Angaleswary R, Logesh T

The outlier identification method has a lot of applications and has recently attracted a lot of attention. Applications for the usage of outlier identification techniques have included clinical trials, voting irregularity analysis, data purification, network intrusion, severe weather prediction, geographic information systems, athlete performance analysis, and other data-mining activities. Outlier identification is an essential task in many safety-critical settings since an outlier flags aberrant operating conditions that may result in considerable performance loss, such as an aviation engine rotation failure or a pipeline flow issue. An outlier is a strange object in a picture, like a land mine. Early detection is essential because an anomaly may reveal a malicious breach into a system. In order to identify aberrant data and use it efficiently for production, this paper explores research on outliers in the automotive industry. Developers should choose an outlier identification method that is acceptable for their data collection in terms of the correct distribution model, the relevant attribute types, scalability, speed, and any incremental capabilities to enable the saving of more exemplars. With less computational complexity and experiments using data from various reporters as well as a synthesis of processed data from deep analysis and forecasting the acquired data for further acknowledgement, our project proposal can cost-effectively identify outliers in large-scale datasets from a variety of data views.

Download the paper