Since its launch, AIE has provided a patent-pending SQL JOIN operator. The 2.1 release adds support for many more SQL commands, which offers profound capabilities:

• Developers can use standard SQL queries to deliver information from the index to a user interface or to another application

• SQL queries allow precision and complexity far beyond the capabilities of search engine queries and use standard operators to apply AIE's analytical capabilities across disparate sources

• Unlike search queries, SQL retains cardinality between related data tables: one-to-one, one-to-many and many-to-many relationships can be used and combined with full text search; this expands the options for querying dramatically

• Attivio's AI-SQL extensions go beyond standard SQL - combining the capabilities of relational and search queries and creating tables on the fly without first defining a schema

To support the inclusion of SQL, AIE 2.1 also includes a JDBC driver, which allows standards-based integration of AIE with many of off-the-shelf BI, reporting and analytical tools and enables database applications to query information directly from the AIE index.

Why SQL and JDBC Are Important Enhancements

The addition of SQL and JDBC to AIE's platform allows greater diversity and accuracy in query results and in BI dashboards that leverage AIE as a data source. This means that applications and interfaces that rely on AIE can provide the insight and range of information business users need to optimize the way they make decisions, detect and intervene quickly on issues, and find opportunities for growth. For example, a BI dashboard that provides information about a company's product lines can now include customer opinion about those products - including insight on specific features of the product so the dashboard can reveal which features may be dragging down the overall product reception.

Adding content and content analytics is a hot topic for BI applications, and a variety of approaches are currently being promoted. AIE's addition of SQL and the JDBC driver offers a way to add this rich source of insight to BI applications that is both simpler to implement and more comprehensive than other methods. Using SQL in AIE takes less design time than using SQL in a database environment because AIE does not require ETL steps, is schema-neutral and allows tables to be added at any time without developers having to change the data model or re-ingest information.

Since more than 95% of developers already know how to write SQL queries, and the JDBC driver is a standard component enabling any Java application to interact with a database, these enhancements to AIE can be easily and quickly adopted for a rapid ROI. Most important, they allow the user interface strengths of BI dashboards and the information richness of AIE to deliver critical insight for both strategic and tactical business decisions and actions.

What Else is New in AIE 2.1?

Key Phrases. AIE 2.1 includes additions to its content analytics capabilities with the introduction of Key Phrases, which automatically extracts and indexes statistically improbably phrases from text. These phrases identify the uncommon and interesting elements in the text and can be used to enrich information discovery and navigation.

Entity-level Sentiment Analysis. AIE 2.1 extends its sentiment analysis capabilities to measure not only the overall quality of opinion expressed in a document but also to detect and assess diverse opinions about specific entities within the document. This allows a generally positive review to be further examined for any negative comments about details. For example, a positive product review may include a negative comment about a specific feature. The ability to automatically detect the specific negative factors allows product managers to expedite off-setting improvements; customer support to better prepare for response; sales and marketing to adapt and so on.

Content Spotlighting. Gives business managers powerful editorial control over the user search experience, allowing management of result ordering for sets of keywords or delivery of promotions that pull in external content (such as advertisements). You can also control when, how and for which sets of content a spotlight operates. Content spotlighting also supports editorial review of changes to ensure management oversight.

Enhancements to Active Security. Release 2.1 includes integrated connector support to enhance AIE's ability to manage up-to-date permissions for documents from a variety of sources. The ALLOW and DENY JOIN clauses bring together documents that match the user's query and instantly verify which documents the user is authorized to view without any need for pre-filtering.

With the new capabilities introduced in 2.1, the Active Intelligence Engine (AIE) becomes the only UIA platform that improves productivity and competitiveness by delivering complete, analyzed information with real-time, BI-style query and exploration capabilities and with the ability for the platform to serve as a database that can add new sources of data to database applications.

TrackBack URI for this entry

Comments (0)