Machine Learning Frameworks for Parking Guidance

A wide variety of open-source libraries and platforms support the creation of machine learning models that provide ready-to-use implementations of many popular algorithms available under various licenses. Such software frameworks reduce the time and effort of researchers and developers engaged in different projects where there is a need to use specific machine learning techniques. Case-based reasoning (CBR) involved in Parking Guidance Information System (PGIS) uses k-nearest neighbor (kNN) classification as well as several other machine learning methods in the various stages of building the action plan. In this analysis, we explore free toolkit options that can meet the needs of a specific input data system. We also consider additional requirements for efficient indexing and optimization of use cases, aiming to reduce the need for multiple programming languages and platforms and coding algorithms from scratch.