Technology Blogs by SAP
Learn how to extend and personalize SAP applications. Follow the SAP technology blog for insights into SAP BTP, ABAP, SAP Analytics Cloud, SAP HANA, and more.
cancel
Showing results for 
Search instead for 
Did you mean: 
wejoyb
Product and Topic Expert
Product and Topic Expert
485

Objective: I aim to demystify what represents an AI business scenario by presenting it in layman's terms.

Let's commence with a straightforward task: folding an A4 paper four times and cutting it along the folds. This yields 16 uniformly sized pieces. Reassembling them back into an A4 shape is a simple feat, requiring minimal cognitive effort and time. In this scenario, a basic program or algorithm suffices to guide a computer through the task.

Now, envision a different scenario: tearing the A4 paper randomly into 16 pieces. Here, we encounter a diverse array of shapes and sizes, each fragment boasting its own distinct pattern of torn edges. Reconstructing the A4 sheet from these irregular pieces becomes a formidable challenge, demanding both intelligence and time. In this context, the need for AI becomes evident. Teaching a computer to navigate through this labyrinth of irregular shapes is quintessential for success.

These scenarios parallel the world of data. In the structured scenario, data mimics the uniformity of the paper pieces, easily interpretable and analysable. Conversely, in the unstructured scenario, data assumes a chaotic form, akin to the torn paper pieces. Here, human intervention is necessary to train the computer, enabling it to decipher shapes, measure dimensions, and ultimately, assemble the pieces. This exemplifies the role of AI tools in tackling unstructured data.

In practice, infusing intelligence into computers necessitates vast amounts of data in various formats—text, images, audio, color, and more. Consider the staggering volume of text and images required for a computer to identify a cat or a dog. The magnitude of data required is immense, underscoring the monumental task of imbuing machines with cognitive capabilities.

In systems like SAP, data stored in databases is typically considered structured data. Leveraging this data to extract meaningful insights often requires sophisticated algorithms. Particularly, when algorithms need to analyse data in various forms such as text, images, or natural language, the complexity escalates. In such cases, training the computer to generate insights from this diverse data landscape becomes imperative.

Key Takeaway : 

Consequently, scenarios demanding multi-modal data analysis represent prime candidates for Artificial Intelligence applications.

Thank you for your patient reading. Your feedback and insights are invaluable in fostering a collaborative learning environment. Together, we advance our understanding and harness the transformative power of AI.