Google’s future search engine

Richard Waters in Financial Times:

According to the search giant, the new technology — a large-scale AI model known as MUM — could one day turn internet search into a far more sophisticated service, acting like a virtual research assistant as it sifts the web for solutions to complex questions. MUM — short for multitask unified model — is the latest in a series of behind-the-scenes upgrades to the Google search engine which the company claims have brought step-changes in the quality of its results.
These include the introduction, a decade ago, of a “knowledge graph” that defined the relationship between different concepts, bringing a degree of semantic understanding to search. More recently, Google sought to apply the latest deep learning technology to improve search relevance with a tool called RankBrain.

Ah.

Knowledge graph has been the Achilles’ heel. If you remember, I had written about semantic web, where links (and back-ported links) provided the system to understand how the content was “linked” together. It was difficult to apply the computer-language approach. However, much has changed since then. There was an uptick of PhD’s totting with their credentials in semantic web without anything much to show for. It has now morphed to AI.

I won’t go into details around “private search” but this concept can be extended to healthcare. An approach like “personalised medicine” (which surprisingly is missing from the marketing linguistics). It was definitely an idea much ahead of its time, and I assume insurance companies never paid for costly investigations without weaving in the cost-benefit analysis. It is difficult, if not impossible, to understand the interplay of genomics at cellular level, and requires sophisticated real-time modelling. Of course, you can integrate it with the clinical workflows (EMR, for instance), but it will become problematic to separate signal from noise.