Welcome to the Semantic World!
This week a little light came on for me. At first I thought it was a mania of mine to point out everything, and want to know the reason for certain things that usually turn out to be inconsequential, but the fact is that I can’t get it out of my head and I decided to write about it.
Have you noticed that now everything has the adjective “Semantic”? I don’t like to think of coincidences as the first option, so I started to think if this “fashion” had something in common, because in theory they are solutions that are not very related to each other. Shall we do a quick list? Semantic Kernel, Semantic Link, Semantic Search, Semantic Models,… do I have any left? Why all Semantics? (Below is a definition and a link for each of the more information if you need it.
Basically, this trend is given by the evolution of technological platforms after the emergence of Generative AI. They are semantic, because they focus on understanding the meaning and context of the data, rather than simply processing the data literally, providing a richer and more meaningful interaction with our data. You don’t have to be an eagle to guess that if we want to develop in this area, we must understand and explore in depth these processes and how they work in order to take real advantage of these solutions and create solutions that adequately capture the intentions of users, improving productivity and agility in the development of tasks.
I confess that the light came on when my teenage daughter, who is studying 4th year of secondary education, asked me what was the use of learning to do a semantic analysis of text. It’s been easy to motivate her 😊
Semantic Kernel. It is an open-source SDK, created by Microsoft, to orchestrate Artificial Intelligence solutions, combining different services (OpenAI, Azure OpenAI, Hugging Face,…) with functions that we can define ourselves, to expand the knowledge of these models, using different programming languages https://learn.microsoft.com/en-us/semantic-kernel/overview/
Semantic Link. It is a Microsoft Fabric tool that allows you to connect the world of Semantic Models with the Data Science projects that are developed within the platform. https://learn.microsoft.com/en-us/fabric/data-science/semantic-link-overview
Semantic Search. Semantic search captures the semantic meaning of the query so that the semantically similar text has similar numeric values to return the best matches, based on other “semantic” concepts (Semantic Index, Semantic Ranking,…) https://techcommunity.microsoft.com/t5/microsoft-mechanics-blog/build-semantic-search-into-your-apps-latest-in-azure-cognitive/ba-p/2220690
Semantic Models. What used to be Datasets in Power BI are now called semantic models. https://powerbi.microsoft.com/en-us/blog/datasets-renamed-to-semantic-models/
