The Technology

CXAIR provides users with a search-based interface and API that can be used to query and analyse data contained in structured and unstructured data sources, allowing the analysis of content as diverse as documents and websites alongside traditional databases.

Unlike traditional relational and OLAP reporting tools, CXAIR reports from a set of physical Lucene indexes that are logically grouped together to form a search engine. This gives access to underlying data, unlike OLAP Cubes or equivalents that take time to build and hold an aggregated view.

For more detailed information, please refer to the CXAIR Technical Architecture.

Performance and Scalability

CXAIR is able to search millions of rows of data in a fraction of the time it would take for equivalent searches using traditional solutions.

Unlike traditional OLTP databases that have been written to support fast concurrent transaction-based data entry, search engines are written specifically around information access and retrieval, resulting in a fundamentally different design principle.

CXAIR has been developed to specifically address the issue of querying and returning large result sets with sub-second response times, making the solution ideal for fast, ad hoc data retrieval and analysis. The underpinning technology negates the requirement for pre-aggregation, instead performing calculations at runtime. This means there is no need to build and maintain aggregated cubes or specify in advance the data that requires aggregation.

Key Features

Ad hoc Searching and Intelligent Data Access

CXAIR provides extremely simple and fast access to data that could reside in many separate systems and databases through a single browser-based interface. The data is assembled and held in a highly-optimised search engine away from the original source data and can be accessed through natural language search terms.

Users can access information by loading and running pre-defined searches or reports, or by finding the data themselves by using wild card searches, range filters and data driven selectors to rapidly narrow down the data to the records that they are interested in. Once filtered, users can then choose how to display the information on the screen by using one of the many visualisation tools to intelligently transform the data into meaningful information.

Regardless of how the user chooses to view the data, they will always be able to drill down to the underlying records, providing a quick and easy way of validating the information presented on the screen.

For system integrators who want to leverage the powerful search and analysis capabilities of the product, CXAIR includes a comprehensive set of APIs that can be called from any application that supports web services.

Data Visualisation

CXAIR provides a multitude of display options for users to gain insight from data.

Crosstabs

Crosstabs are an extremely powerful and easy way to turn filtered results into meaningful information.

Users can analyse results by selecting one or more fields to use on the X and Y axis of a grid, selecting the measures or calculations to perform on the data and whether to see the data rendered as a table, chart or both.

Pages

Pages allows users to present data without the visual restrictions of traditional reporting packages, providing the power to position reporting data anywhere on-screen. 

Users can structure and group related data together to incorporate logos, pictures, images and text as if typing data into a publishing package.

Venn Diagrams

Venn diagrams are an excellent way of identifying and understanding relationships in data that are not immediately apparent when using traditional reporting methods and languages.

CXAIR is able to assemble the information in seconds, allowing users to quickly add and move Venn sets around the screen to create clusters of related and unrelated data.

Statistical Modelling

With statistical modelling, CXAIR is able to approximate reality from a loaded index and display the results using a range of visualisations as a fully interactive Bayesian network or decision tree.

Harnessing a wealth of configurable algorithms and display options, accurate and insightful statistical models provide a level of insight far beyond the reach of traditional business intelligence tools.

Key Performance Indicators

By transforming key data items into a visual experience, users are able to view a multitude of key performance indicators at a glance.

Each key performance indicator (KPI) can be allocated any number of bandings, which are then colour coded and labelled based on how the score is recorded. These bandings can then be compared on the same screen to quickly and effectively monitor performance across multiple areas at a glance.

Report Viewer

Report Viewer allows the communication of key information to clients and colleagues without the need for full access to the expansive solution by allowing registered users restricted access to pre-canned reports from within a browser window.

Publishing and Sharing Reports

By default, users can only see their own reports. However, if the user has the relevant feature enabled for their user account, they can copy their local reports to one or more shared report folders which are visible to other users.

Users can be configured to receive alerts in the form of either an email, informing the user when a new report is available for them to run or that the underlying data has been updated.

While reports and queries return with sub-second response times, often there is a requirement for a report to be pushed to a client at pre-defined time intervals. To support this, CXAIR allows users to schedule single or batches of reports to be ran in the background.

Calculations

Users and administrators are able to create custom calculations using the point and click Calculation Builder.

The calculation menu allows a user to create ‘pseudo fields’ and re-usable variables by using one or many of the preconfigured functions, CASE statements, IF statements or mathematical calculations.

Calculations can be based on formulas derived from event-level data or from one or more distinct counts.

Data Integration

CXAIR is able to assemble data from multiple, disparate data sources into individual indexes, where data originating from different systems can be grouped together and queried as if they were from a single data source.

In addition to combining indexes together, CXAIR also provides a number of features for transforming the data as it is loaded into a CXAIR index to address any data quality issues.