**6. Results discussion**

This chapter reviewed the literature on BDI tools and applications in diverse industries and presented the highlights from each domain. All most all organizations are gathering a huge quantity of big data in real-time, Online and offline modes. Managing, real-time processing this big data to extract useful business information for intelligent decision making is the real challenge. The big data processing systems are empowered by big data integration and analytics platforms. BDI systems are facing the challenges in integrating and synchronization of heterogeneous big data from multiple distributed sources. The lack of comprehension and management of uncertainty in big data is another challenge faced in the big data processing. BDI processing should ensure context-sensitivity and extracting the semantics in the distributed data processing. Research on designing effective machine and deep learning algorithms is going on in the BDI domain. BDI Processing uses Hadoop environment with HDFS,

Spark computation model with a Hive database as a distributed data warehouse. The use of Apache Spark enhances short time frames for quality and availability reporting. BDI processing involves data acquisition, semantic integration, statistical data analysis, data visualization, data query language, geospatial data techniques. Big data analytics framework enables us to create business value.

The economic model of the cloud promotes BDI processing by providing online services for decision-makers and the business community. Usage of various tools, techniques and applications of BDI leverages the ability to handle speed, variety and uncertainty. Knowledge-driven framework for BDI describes a knowledge graph. Human-machine interface for BDI integrates data from heterogeneous resources in a secure and scalable way. Ontologies and unified schema as a knowledge graph for describing integrated data. Multi-Source BDI is a framework for integrating data in the distributed storage environment. Authors have discussed BDI research issues and challenge data accommodation, data irregularity, query optimization, extensibility, ETL processing. Remote sensing data. Real-time big data analytics processes stream of big text, image, and videos generated from IoMT, IoT and CCTVs systems. Implementing real-time, complex big data analytics in the cloud using BDI process reduces the cost of operations. This paper discussed five case studies of BDI applications implemeneted in the in world-class organizations.
