Improving customers' service of a transit bus system with the help of machine learning and artificial intelligence

在这项研究中，目的是分析与公交系统相关的数据，并从这些数据中提取模式，以及异常情况。然后将这些模式和异常通知给受影响的客户，这些客户将从这些通知中受益。该项目将使用CRISP-DM（数据挖掘的跨行业标准过程）方法进行，即首先开发业务理解，然后了解可用数据及其局限性和有用性。然后，这些数据将准备用于建模和测试，并进行所有必要的转换。将使用各种验证技术开发和测试几个不同的模型，并选择最合适的模型进行部署。该项目的结果将是一个原型仪表板应用程序，它将作为概念验证。这个原型最终将为客户提供一个商业的生产级web应用程序。

In this research, the aim is to analyse the data related to a bus transit system and to extract patterns from these data, as well as anomalies. These patterns and anomalies will then be notified to the affected customers who would benefit from these notifications. The project will be conducted using CRISP-DM (cross-industry standard process for data mining) methodology, which is, developing a business understanding as the first step and then understanding the available data along with their limitations and usefulness. These data will be then prepared for modelling and testing, with all necessary transformations. Several different models will be developed and tested using a variety of validation techniques and the most appropriate model will be selected for deployment. The outcome of this project will be a prototype dashboard application which will act as a proof-of-concept. This prototype will ultimately lead to a commercial production-grade web application for customers.