What’s the quickest way to get answers? Simply ask a question.
Give your users (with any type of technological background) a natural language interface to ask questions over data.
They will thank you.
Kueri is a modern interactive natural language interface. It’s friendly. It’s intuitive. Let us convince you why Kueri is such a good idea:
- It is great in helping users construct their query with as-you-type suggestions
- It is forgiving when it comes to spelling mistakes
- It is capable of getting it right even with just parts of a sentence
- It can intelligently handle data ambiguity
- It gets better with time by learning
HOW DOES IT WORK?
Kueri is based on as-you-type query construction/completion. With each keystroke, users are constantly supplied with valid choices for query construction based on the database to which Kueri is connected. Once a valid question is formulated by the user, Kueri generates the appropriate properly formatted database query (depending on the type of database, e.g. Relational or NoSQL) and sends it to the host application. The query is then effectively run on the database and it is up to the host application to take care of any visualisation for the results (the Kueri SDK also comes equipped with a free visualisation tool, just in case).
Kueri is easy to integrate to other applications.
Kueri SDK is equipped with a complete backend configuration and management tool which allows for a smooth and quick setup.
Kueri supports relational databases including: MySQL, Oracle, MSSQL, PostgreSQL and Amazon Redshift. NoSQL databases are also supported via a simple to use JSON connector. Additional connectors such as Google BigQuery and HP Vertica are in the pipeline. Contact us for special requests.
THE SEARCH BOX
The search box is built with 3 blocks: the user input search-box; a computer sentence which displays the current query the machine understands; and lastly, valid suggestions from which the user can choose.
Kueri supports many language capabilities including filtering, aggregation, comparisons, sorting, mathematical functions and adding additional columns.
In many cases, a user might type text which is relevant to several columns, creating an ambiguous situation. As a default, Kueri will present users with various options to choose from, leaving it up to them to fine tune their query. In addition, Kueri can be configured by system administrators to favour a specific interpretation over others.
Kueri understands verbs out of the box. It does not impose a lengthy pre-configuration process. In the case shown below, the word “paid” entered by the user is automatically picked up by Kueri to be related to the “pay date” column and the proper query is constructed, as can be seen in the computer sentence section – “Show invoices with pay date in the last year” (in grey, directly under the input zone).
MIDDLE SENTENCE SUGGESTIONS
One of the most compelling features in Kueri is the ability to go back and see relevant suggestions that fit with the entire sentence. By moving the cursor back and forth along the already typed query in the search-box Kueri will display suitable possible recommendations to replace the selected word. This is highly useful for insightful data exploration.
AUTO-COMPLETE FOR VALUES
Kueri can suggest possible values for field values while constructing the query.
In the example show below, once the user writes the words “with supplier name” in the search-box, Kueri quickly rushes to the rescue with possible option of values from the related datasets. The indexing process is predetermined by the system administrator.
Kueri is capable of accepting basic math functions such as addition, subtraction, multiplication and division. Users can use the words plus/minus/divide/multiply or the signs themselves (+, -, / and *). Watch:
ADDITIONAL FEATURES AND CAPABILITIES
Kueri is still in its early stages as a product and undergoes changes all the time. Be sure to become a part of our development community to contribute and learn more about the potential of the Kueri SDK. SimpleQL, as a company, welcomes and feedback from its developer community and is eager to hear what any customer, developer or end user has in mind. We welcome your feedback as much as your criticism.