Название: World of Business with Data and Analytics
Автор: Neha Sharma, Mandar Bhatavdekar
Издательство: Springer
Серия: Studies in Autonomic, Data-driven and Industrial Computing
Год: 2022
Страниц: 211
Язык: английский
Формат: pdf (true), epub
Размер: 63.7 MB

This book covers research work spanning the breadth of ventures, a variety of challenges and the finest of techniques used to address data and analytics, by subject matter experts from the business world. The content of this book highlights the real-life business problems that are relevant to any industry and technology environment. This book helps us become a contributor to and accelerator of Artificial Intelligence (AI), Data Science and analytics, deploy a structured life-cycle approach to data related issues, apply appropriate analytical tools & techniques to analyze data and deliver solutions with a difference. It also brings out the story-telling element in a compelling fashion using data and analytics. This prepares the readers to drive quantitative and qualitative outcomes and apply this mindset to various business actions in different domains such as energy, manufacturing, health care, BFSI, security, etc.

In a corporate network, the security resources have upmost importance to the organization. In order to maintain security, different techniques are used such as firewall and intrusion detection system. Some of the most efficient detection techniques to find network intrusion implements Artificial Intelligence, Machine Learning, soft computing techniques, and bio-inspired techniques. But very often, some intrusion attempts are able to breach these defense mechanisms. So, a new secure network model is proposed which combines both signature-based detection and data mining technique which act as an efficient detection and response system against intrusion and is well established and evolved system. The proposed model combines signature-based detection and data mining technique to improve the efficiency of the intrusion detection system.

Cloud deployments have brought promise to business services in the entire spectrum of infrastructure such as provisioning, infinite scalability, cost of utility at consumption, etc. Cloud usage data forms a critical data layer to measure usage and apply billing and is made available to cloud users that make the use of cloud services. This chapter walks through methodologies for various optimization applications aided by machine learning using cloud usage data. A framework for cloud intelligence application is recommended to be provisioned as a data analytics and intelligence gathering mechanism that will open opportunities in cost optimization techniques, Intelligent Provisioning, and capacity forecasting mechanisms. For generalization purposes, public utilization datasets from Microsoft Azure are used for Exploratory Data Analysis (EDA) and recommendations.

The chapter "Explainable AI for ML Ops" explains the significance of blending two emerging technologies in AI/ML—Explainable AI (XAI) and Machine Learning Operations (ML Ops) and demonstrates a focused use case that derives value from leveraging XAI to enhance ML Ops. The chapter starts by laying out the “growing pains” problems that enterprises are encountering to scale AI.