- AI For All
- Posts
- Natural Language Processing in AI
Hello readers!
On this week’s podcast, Ryan and Neil speak with Deep Dhillon, the Founder and Chief Data Scientist at Xyonix, about the fascinating evolution of natural language processing (NLP) and the valuable insights that can be extracted from business data using NLP and AI.
As always, thanks for being a subscriber! For the full conversation, watch the episode on YouTube or see our podcast links below.
Now, let’s take a closer look at what Deep had to share with us!
“It's just not cost effective for one startup company to build and train their own LLM just because their investors are obsessed with them owning their own destiny. Are you going to give me that two million bucks to train up? No? Okay. Then we're not doing that.”
The Evolution of NLP (and the search engine)
Deep takes listeners on a fascinating historical journey, which begins approximately three decades ago. In the early years, NLP activities were rudimentary, focusing on isolated word analysis and their frequency of occurrence. These efforts culminated in the primitive search engines of the 1990s, which relied on term frequency matrices to understand and categorize online content.
Deep elaborates further on the progression to search engines which not only became robust in their search capabilities but also began to harness a nuanced understanding of web content. The NLP realm witnessed a transformative breakthrough between 2010 and 2013 with the advent of deep neural networks, making training on vast scales feasible.
Talking to Machines
Neil introduces prompt engineering, where he refers to artists who craft 50 unique songs in just two days using generative AI. A crucial ingredient in achieving such feats is the art of crafting "prompts". Unlike the common perception of just inputting a few words, effective prompts are often intricate, encompassing an array of parameters.
Deep extends this discussion, illustrating that today's prompts are far more dynamic and multifaceted. At Xyonix, for example, systems are built to comprehend and generate based on thousands of prompts.
Know Your Data
For businesses to harness AI, Deep suggests a two-fold approach: First, companies should introspect on areas where they experience bottlenecks or inefficiencies. These pain points often present the most immediate opportunities for AI intervention. The next step involves taking stock of your business data. By assessing the volume and type of data – be it text, images, videos, or specialized data – it will become clear the potential insights that could be extracted.
For instance, an insurance company with hours of recorded conversations between claimants and agents can employ AI to extract patterns. This could help in discerning characteristics of valid claims versus fraudulent ones based on speech patterns, call frequencies, and more.
Teaching Machines to Understand Humans
Meet Deep Dhillon
Deep Dhillon has a wealth of expertise in AI, ML, and NLP - all of which he brings to bear as Founder and Chief Data Scientist at Xyonix. Deep helps clients accelerate AI-driven innovation in their products by teaching machines how to read, see, listen, and extract valuable insights from their business data.
About Xyonix
Xyonix teaches machines how to read your content, see what is in your imagery, watch what is in your videos, and understand what is in your data. With this enhanced ability, your custom-built AI and machine learning models can generate insights and make high-value predictions.