Service Provider Cloud

Salesforce Using AI, Natural Language to Search Databases

Salesforce is proposing using artificial intelligence and natural language capabilities to allow almost anyone to search relational databases, such as SQL, making it easier for all workers, not only developers, to find the data they need.

On Tuesday, Salesforce introduced Seq2SQL, which allows anyone to search a database by typing a question into a search field or speaking a question into a device, without having to know much about relational database development and programming languages.

In addition to a blog post, Salesforce Research released a paper entitled "Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning," which details how researchers are using neural networks to translate questions into SQL queries.

In the August 29 blog post, Richard Socher, chief scientist at Salesforce, notes that by 2020, IDC found that the world of connected devices and software will produce some 44 zettabytes of data, and it'll be more important than ever to be able to search that data to find the answers that businesses need.

(Source: Salesforce)
(Source: Salesforce)

The problem is that not everyone has a background in database management or development.

This is where Seq2SQL, when applied to a SQL database, comes in.

"The ability to quickly and easily communicate with relational databases allows business users to achieve new levels of productivity and a better understanding of their customers," Socher writes. "For example, with Seq2SQL, we take a first step towards a world in which service leaders can quickly access their most important key performance indicators (KPIs) in plain English versus manually selecting columns or inputting conditions."

In the example within the blog, researchers ask: "What place did Phil Mickelson finish with a total of 282?" within a standard search bar. The Seq2SQL then matches up the query with the appropriate database to provide the answer. (The answer, by the way, is tied for 16th place in 2005.)

Part of the ability to find accurate answers comes from the second part of the announcement.

Salesforce also detailed its work on WikiSQL, which is an open source dataset that researchers used to help train the model to work and to take advantage of the various machine learning capabilities. The wiki is a "dataset of 87,726 hand-annotated examples of questions and SQL queries distributed across 26,375 tables from Wikipedia. This dataset is required to train our model and is an order of magnitude larger than comparable datasets," according to the research paper.

Keep up with the latest enterprise cloud news and insights. Sign up for the weekly Enterprise Cloud News newsletter.

Taken together, Seq2SQL and WikiSQL look to pare down the amount of answers that a query could return and "learn" to return the most accurate answers possible.

In its blog post, Salesforce did not announce a specific commercial application for this new AI technology but Socher wrote that the "impact of this research does not stop at service, and can also benefit sales and marketing by unlocking valuable insights about incoming leads, opportunities and overall pipeline health."

That likely means that Salesforce is planning to incorporate this technology into its Einstein AI offering at some point, or create a whole new commercial offering. (See Salesforce Brings Einstein AI to Field Service Lightning.)

Related posts:

— Scott Ferguson, Editor, Enterprise Cloud News. Follow him on Twitter @sferguson_LR.

Page 1 / 2   >   >>
kq4ym 9/15/2017 | 9:28:18 AM
Re: 44 zettabytes of data Searching data with natural language may well be a significant tech change. Allowing non-database folks to easily and quickly search and find the right data to come up with the answers will be a huge change, that as noted may not be welcomed by the current data base administrators and those folks whose job now is to do that work.
hugefan 9/1/2017 | 11:55:40 AM
Re: 44 zettabytes of data useful info I am new here so.. but I really agree with this.
Susan Fourtané 9/1/2017 | 8:57:24 AM
Re: 44 zettabytes of data Yes, Maryam. And there are many of those cases now being affected by several technologies. There are new jobs that will be created. Some that don’t even exist now. The thing is that people should be prepared to upgrade their skills, or learn new ones.
[email protected] 8/31/2017 | 4:51:45 PM
Re: 44 zettabytes of data Very true especially when it impacts someone's job growth directly. Those technologies tend to face the greatest resistance.
Susan Fourtané 8/31/2017 | 5:25:11 AM
Re: 44 zettabytes of data Not everyone is always happy with the new technologies. It has happened for certuries. It won’t change.
[email protected] 8/30/2017 | 4:58:00 PM
Re: 44 zettabytes of data Many may be happy but the database admins may not be!
Susan Fourtané 8/30/2017 | 4:39:54 PM
Re: 44 zettabytes of data I think that’s exactly the point. To create more business agility, and simplify your work and the time wasted waiting when you could do your own search. Many people will love this.
[email protected] 8/30/2017 | 4:34:02 PM
Re: 44 zettabytes of data I think it's an awesome idea, making data searches easier and more intuitive is a win for everyone. I have spent endless hours in my career waiting for database queries to be scheduled because of limited resources to perform the queries. If I could do them myself it would move business initiatives forward much more quickly.
Susan Fourtané 8/30/2017 | 2:00:25 PM
Re: 44 zettabytes of data Joe, I personally believe it’s impossible to know the real amount of data that will actually exist in 2020, or next year for that matter. The same goes for IoT, and so on. No one’s crystal ball will ever be precise, unless they can also predict or forecast the unexpected, which I doubt anyone can. However, as per a figure for illustrating that both data and connected things (some include devices in this cathegory, and some don’t) are rapidly growing, what the research companies estimate works for me. In fact, estimations is all what it is. I could predict I will drink X amount of tea by the year 2020 based on my daily number of cups of tea consumed. However, I have no way to anticipate the unexpected, which is what, in the end, will determine the real figure. Therefore, any prediction about my cups of tea (I promise, it’s a large number) will prove to be innacurate in 2020. It may sound silly to you, but that is exactly how it is. No matter matter is it’s data, IoT, or cups of tea. The result will be the same.
Joe Stanganelli 8/30/2017 | 12:23:10 PM
Re: Sign me up @John: For my own part, my "Google-Fu" is pretty good...but sometimes I have difficulty searching data within my own company systems.
Page 1 / 2   >   >>
Sign In