Engineering the AI applications (Part 2): "Rules First" approach.

In this short post, I would like to share a few lessons that we have learnt while developing the AI applications.?

Rules first. Back to the basics!

In order to provide IoT and Data solutions, we approach local businesses. We were asked if we could assist them with online sales after hearing our proposal for our IoT/Data solution.

Online sales are a hurdle, as we have discovered from our research with local business owners (who may utilize our software solution).

It is undoubtedly a new opportunity. A major concern for us is how to move forward.

The challenges (constraints) that apply to sales vary depending on the type of business. It necessitates creating unique data solutions.

After identifying the root cause of their sales difficulty, we set out to develop an AI program to help the sales.

Smart Catalog - AI enabled product database with rules.

We took an approach to develop Rules for product (or family of products) and create a Smart Catalog to address the challenges of online sales.

After a few proof-of-concept implementations, we started gaining the understanding of the process to develop rules, facts database etc., for every product (or family of products).

It is still a very beginning to streamline the automation of Rules, but PROLOG and CLIPS seem to be very good fit for our need.

Conclusion

• Rules based AI systems are quite suitable for many business cases.
• Prolog and CLIPS provide backward chaining and forward chaining implementations; these are mature languages to implement AI systems.

Affiliations

I write posts for Vidcentum Technologies , where I am founder director. Some of the articles are published in my blogs / articles / posts as well as on the company timeline / posts.

Vidcentum Technologies is a data and software company. It provides AI, IT/OT related services.