INTELLIGENT ASSISTANT FOR DATA PREPARATION AUTOMATION IN URBAN TRANSPORTATION MANAGEMENT SYSTEMS

Collecting and processing a public transportation schedule, traffic sensors, and GPS devices will result in a large amount of data being generated by modern day urban transportation management systems. This data needs to be extensively cleaned, include feature extraction, and normalization in order to make any analytical or decision-based data. While these traditional techniques might work, they are often expensive, subject to inaccuracies, and very tedious - especially when dealing with the ever-changing dynamics of the urban transport system. Real-time, Large scale transportation data needs additional automation in order to make the students efficient and accurate. Within this document, an intelligent assistant done with machine learning algorithm will do all the legwork required to prepare the data for the urban transportation systems. Rather than focusing of specific preprocessing actions like most other intuitive solutions do, the system utilizes a flexible and smart framework which incorporates intelligent automated feature extraction, data optimization, and anomaly detection. Eleven other frameworks guaranteed modularity which, allows for reduced manual effort and computational efficiency make smart city applications possible. A new approach towards more effective controlling and directing surface public transport is proposed which helps cities automate the MDT processing, lowering the costs for monitoring data processing, a more effective controlling, and working with decreased error carefully.