Mining encyclopedic websites

Searching through pages and pages for an answer...

The internet has everything… except the exact answer you need on the first try.

Let’s look at 3 examples.

Who are the top competitors for X product in Europe?

The problem: The answer from ChatGPT and general Gen AI tools might list global players but miss niche competitors operating in targeted regions, even though they’re on the Internet!

The solution: Leverage AI to parse all available web search results. The AI can extract, classify, and rank competitors, including niche players.

What do Gen Z consumers think about X product?

The problem: Results may rely on outdated surveys or high-level summaries, missing real, specific consumer sentiments in reviews on the Internet!

The solution: Use AI to gather discussions on platforms popular among Gen Z (e.g., Reddit, TikTok, Instagram). The AI can then conduct sentiment analysis and surface insights.

Has this idea been planted before?

The problem: Results may miss obscure patents, international filings, or pending applications in databases on the internet.

The solution: Retrieve all relevant patents, including those from international databases. Have AI cross-reference findings to determine if the method has already been patented.

In the work that we’ve been doing, we’re making sure to apply the necessary solutions to cover these scenarios.