Machine Learning and Cognitive Computing for Mobile Communications and Wireless Networks

1Machine Learning Architecture and Framework

Nilanjana Pradhan* and Ajay Shankar Singh

School of Computer Science and Engineering, Galgotias University, Greater Noida, Uttar Pradesh, India

Abstract

Machine Learning is one of the fastest developing fields in computer science with wide range of applications. The machine learning architecture involves lot of complexity. The machine learning architecture will implement learning algorithm in the application engine which will perform the predictions, perform various complex queries in database and finally use analytics tools to produce predictions based on application areas. An effective machine learning architecture helps in designing better data centers, promote human welfare, solving critical system failures. A good architecture will cover all important risks involved with data privacy and security areas. In order to set up an effective machine learning architecture the problem area must be well defined. Training data (text, images, audio, video, structured data, user generated content etc.) must be collected for machine learning development process. In most cases data are incorrect and useless. The quality of the data matters in building and effective ML system. Good data visualization, data filtering, encryption tools and analytics tools are required. The machine learning system must be tested with test data. The model must get validated. In business domain ML algorithms are implemented on business processes, business services, ...