Hello everyone, welcome to another Redapt webinar. Thanks for joining us. Today we're covering how to transform unstructured data into actionable insights and user value with AI.

Joining us on the call is Michael Kimbrough, Senior Enterprise Architect with Redapt and data scientist Matt Honecker, also with Redapt.

In the business world of today, almost every company collects a huge amount of data about their operations and their customers. One of the really big problems with this big data revolution is transforming all of that data into something that actually can mean something and provide actionable insights.

There are two general categories when working with data and AI. One has taken over the world in these past few years - that's generative AI, think ChatGPT. Generative AI is open-ended generation of outputs, typically natural language outputs, but it can also be things like code in response to an input.

Structured data can be thought of like rows and columns, kind of like a matrix. Almost anything stored in a SQL database is almost always going to be structured data. If you use spreadsheets, that's also an example of structured data - anything that can be easily categorized in this matrix, row, column format.

Unstructured data, which works really well with generative AI, can be thought of as text or transcripts. So if you have a document or book, any kind of transcripts of conversations or videos, images are also considered to be unstructured data because generative AI is typically built around natural language.

Generative AI is built on something called a large language model. They're sophisticated and highly complex machine learning models designed to output the most probable correct response to a natural language input.

One of the most common ways that we see generative AI being used is in the Retrieval Augmented Generation, or RAG pattern. Here we're providing a corpus of knowledge to the generative AI which it can efficiently search based on natural language and semantic meaning.

Intelligent applications can be thought of as any sort of application where we're taking this AI technology and the data we have from your business and external data sources, learning and improving interactions, with agentic autonomous responses over time and transforming experiences for users of all types.

You can break it into about 3 different general areas: adaptive experiences, intelligent decision making, and process enhancement. We're looking at bringing the connecting analytics, contextual connected data and informing those decisions for you.

The self-help chatbot is a canonical example. You've ingested a corpus of knowledge, documentation, content, and you use RAG and generative AI to answer user questions, guide them through a process using natural language and chat interface.

Form recognition is another great area where we have the capability to enhance the user and business experience. In this example, we worked with a loan originator looking to reduce the number of applications missing required information.

With generative AI and machine learning growing by almost unfathomable leaps and bounds every day, incorporating this into everyday business is just starting and we're going to see it more and more. You don't want to get left behind, but you do have to be careful that you have good vision and you're doing it in an appropriate manner.

Thank you so much for your time and great insights. Hope to see you again on another Redapt webinar. In the meantime, make it a great day.