Speaker:Aravind Kalakuntla

Detecting Anomalies in Probe Data

Over the last decade, technological advancement has significantly influenced traffic management and strict rules implementation, which made the life of traffic enforcers easy and made the world a better place for drivers. With this development, there is a lot of dependency on various sensors used in the weather forecast, speed sensors, traffic detection etc. Due to this dependency, the accuracy and data of these sensors greatly impact the analysis of information, and in turn, the alerts being published to the public are affected. Because of this, the accuracy and completeness of the sensor data are very important, and there should be some metrics to ensure these properties of data.

That greatest importance sensor data quality and completeness have acquired is the fundamental aim of the project. In this project, I am analyzing the probe data to detect anomalies in quality & completeness and published Tableau dashboards, which help the technicians concentrate on the sensors with low data completion and low accuracy (based on an attribute within data). This information is stored in a database and can be used in further analysis and other processes.