Hey everyone, welcome back to the Data and AI with Mukundan show. If you notice recently I've undergone a rebranding. So I used to have a different name for this podcast. 

I just wanted to focus specifically on data and AI. So that's why this show has a different name now. So if this still sounds interesting to you, please stay tuned.

Alright, so back to the episode. Now imagine that you're having a tool that helps you understand the strengths, the weaknesses, the opportunities and the threats of anything like a business or a project. Just like you'd like to make a list of what you're good at versus what's hard for you and what exciting things are possible and also what you need to watch out for.

So when I had done this initially, I want to call this like maybe a v1 of this tool. I had used GPT 3.5 and this is basically I'm talking about SWOT analysis. I had written a very successful blog. 

So I mean it was my most successful blog at least so far and that's why I wanted to just focus on this particular blog because it gave me a lot of success. So I wanted to use a newer approach, a newer AI method to do SWOT analysis instead of just using GPT 3.5 which is AI by itself but still I would call it a more basic AI. And I'll explain what traditional basic AI is versus the newer version of AI which I'm talking about. 

So the first version which is the basic version we can call it was using GPT 3.5 which is that's how GPT became famous in the first place. So GPT 3.5 is nothing but a large language model and it works like a search engine. So you tell it a company name and it would quickly come up with a list. 

Now for example, you can think like you've asked about it, asked about a company and I mean it's the same thing. I'm sure you've had familiarity with GPT by now but just wanted to give you some more background about what we're going to be covering today. So for our example today, we can just consider a fake toy company right. 

Let's just say this company is good at making fun toys and consider that as the strength or that it has a big factory or so could be a strength. But sometimes I mean things like this could be too simple. So chat GPT 3.5 or I mean GPT 3.5 in this case was giving you a very simple answer. 

It could tell you that the company, the toy company had a big factory but not explain why that is important right. So now I want to talk about Neuro Symbolic AI which is the newer AI model that I'm proposing to do SWOT analysis. So in this project just to give you more insight, I used Neuro Symbolic AI versus traditional GPT 3.5 to do SWOT analysis and I'll link the blog post in the show notes and also the episode where I talked about this in the show notes as well. 

So now coming back to Neuro Symbolic AI, it is a new smarter version and it is a mix of two ways of thinking right. So one is pattern recognizing. So it helps you find out what's common in all the information just like GPT 3.5 did. 

So GPT 3.5 which is a language model which powers you know the LLMs like chat GPT. It helps you find out what's common in the information. So that is something which is useful and also it does rule following. 

So it's basically applying smart rules to understand why something is important. So for example instead of just saying that the company has a big factory, the Neuro Symbolic AI might say this company has a big factory which helps it make toys faster and keep up with the big orders from stores. It's like having a tool that doesn't give you a list but also explains why each thing matters and what could happen in the future right and that's more useful to have than just generic information.

So you're already seeing the advantages here. But I just wanted to maybe explicitly call that out. So why Neuro Symbolic AI is better is that with this newer approach you get deeper insight.

It doesn't just say good at making toys. It explains how and why that helps the company. Right so the deeper insight is there. 

So you're getting more details as well. It can notice things which others might miss like future problems or special opportunities. And it is also better for planning. 

So it can help the company plan because it knows what might come next. So if you're running a toy company, a fake one for that matter, the first tool is like getting a list of facts. The first tool which is using GPT 3.5 is just getting a list of facts. 

But Neuro Symbolic AI is like having a guide who understands the facts and can give you advice for making the company even better. Right so just want to talk about like why I even revisited SWOT analysis. And I gave you a brief context about this in the beginning. 

But the GPT 3.5 based SWOT analysis it provided a quick accessible way to generate SWOT analysis. But it often resulted in general general and generic insights basically. So there was not enough context. 

The predictive power was lacking. So that's why I wanted to just try this out. And Neuro Symbolic AI like I just mentioned it goes beyond the simple pattern recognition and it combines structural reasoning for more detailed and contextually rich analysis. 

Right and so this it has a lot of real world benefits with using Neuro Symbolic AI. So something like SWOT analysis which I tried in my case it helped transformed it into strategic tool by delivering transparent predictive insights and also helped rank these priorities in order. So it's very valuable for companies needing to anticipate and strategize around evolving market conditions as it turns these more general observations into a more focused roadmap. 

So I wanted to just emphasize that Neuro Symbolic AI has and will continue to set a new standard for strategic analysis making you know problems like SWOT analysis predictive contextual and actionable. So I just want to encourage you to use Neuro Symbolic AI for anyone who's seeking a more sophisticated and a deeper approach to business insights. That's all for me today. 

Thank you and I look forward to seeing you in the next one.