A few years ago, I proposed an Iterative Search architecture that when used in conjunction with a Virtual Assistant would help feed the conversational dialogs when the Virtual Assistance was at a loss for an answer. See white paper. In a different post on this blog, I commented on how the newest releases of Solr and ElasticSearch can supplement a Virtual Assistant in these scenarios. In fact, Search has now been adopted by Virtual Assistants for that same reason. Digital Assistant, like Siri and Google Assistant, usually provide a specific answer to a question, e.g., “What is the oldest city in the world,” it will give you the answer, or ask to do something for you, e.g., “Turn the lights on in the living room.” The digital assistant, when confident of the answer, reads back aloud over the home speaker or phone the answer or confirms the action taken. What they do, if they have no knowledge about the question or do not understand the intent? They do a web search to find the best possible matches to the query. Siri and Google Assistant have improved this feature recently. You may have noticed that if they do not have a specific answer to your question they show a list of web links containing the information you're looking for. Both have a filtering algorithm that ranks them and shows only the top three or four. Siri used to use Bing for these search suggestions; it recently switched to Google Search. As expected, Google Assistant uses Google Search as well. Why are Siri and Google Assistant not iterating further the search step to find among the search suggestions the most likely answer to the question and read it back aloud? These search suggestions, however, can hardly be read back to the user, for good reason because the list would be too long, for one thing, but also because these are not answers but links to pages that may contain the answer. So, they both display the list of links on your screen. Having the option, though, to have the list of suggestions read aloud would be my preference. Imagine being in the car asking Siri something, the search suggestions are not helpful at all, if you want to keep the eyes on the road.
Also, these search suggestion are even less useful with a home speaker like Google Home or the upcoming Apple HomePod, given they do not have a screen to display the results. Hopefully, we'll have an answer in a future post. So, my point is this: why aren't Siri and Google Assistant iterating further the search step to find among the search suggestions the most likely one, extracting and reformulating the significant sentence from the page? Essentially, doing the step 3 in my Iterative Search. The penalty may be just a few hundred of milliseconds to compute the similarity and re-rank the top handful of results. From a user experience standpoint, it would definitively be better, in my opinion, even waiting for that extra half a second. The only caveat is that Siri and Google Assistant should qualify these answers differently. Just like they do today prefixing the search suggestions with, "This came from a search..." or "OK, I found this on the web..." they should append that with "...If unsure of the answer, ask me the rest of the results." Comments are closed.
|
Categories
All
Archives
January 2019
|