We talk about the cognitive movement and the future of Artificial Intelligence, but lets never forget that data is the critical competent for successful AI. Without the right data, you won’t have a cognitive system. Big data and AI work in synergy with each other - and it is this relationship that can drive business to the future. I want to break this down for you, so I’ll answer the following questions: What is big data? What is the difference between structured and unstructured data? Why is data so important?
Big data involves managing massive amounts of info, at speed and in time, to allow an analysis which will enable you to resolve an issue. This includes complex data from a variety of sources - the volume of data, the variety, the velocity (speed of the data movement), and the accuracy of the data (extremely important). From a business perspective you want to be able to analyze massive amounts of information to give you an insight, that will ultimately add value to the business.
Structured data typically refers to data that has a defined length and format. This could include dates, or clusters of words - and these are usually kept in a data warehouse for analysis. Twenty percent of the world’s data is structured. Examples include GPS, financial data and point-of-sale.
Unstructured data is data that has no specific format.This is eighty percent of the world’s data. Machine generated unstructured data would include photographs and video footage (surveillance & security), and scientific data. Human generated unstructured data would include emails, texts, and social media data.
A lot of attention in the AI field has been focused on analyzing complex unstructured data, but it is important to remember that these results must be integrated with your traditional databases and business applications. For a system to be cognitive it needs to have a bigger, more complete, amounts of the data, so that the context is structured. A learning machine requires masses amounts of data that must be analyzed. This big data needs to be managed and integrated, in order for the cognitive system to be effective.
An effective machine learning system needs ALL data. This includes structured and unstructured data. The data will be different in size, and type, but the important thing is that the data is accurate and being used in thesis right context. So both unstructured and structured data are important, and should be included. Data and AI play symbiotic role in making applications smarter. The future of AI lies in the data so make sure it’s selected and managed properly, and important to understand. Just remember - the better the data, the better the AI. With the amount of data out there, the possibilities are endless.