Introduction

Introduction

Classification and Language Service is used to provide classification facilities for applications. The application analyses the submitted text and returns categories from a particular “model” (“classification scheme”) for which the document is appropriate. This information can be used, for example, to ensure government eGMS compliance of the web pages found on a council web site by being used to generate appropriate “META” tags for inclusion on web pages such as “DC.subject”.

Classification and Language Service runs as a service on a server (computer) that continuously listens to a configured TCP/IP port for incoming requests. These requests take the form of XML submitted via the HTTP protocol to the server via the configured port.

This document is intended to give the reader an understanding of this interface and how it can be utilised for their specific requirements.

Audience

This document is intended for those individuals who are implementing interfaces to the Semaphore Classification and Language Service software (typically system developers or system integrators). It is assumed that readers of this document have a good understanding of the following:

• Web technologies including HTML, CGI, and XML

Implementing Classification and Language Service in a CMS environment

A sample implementation is where Classification and Language Service is called directly from a Content Management System (CMS), web site management facility, or other data repository as follows:

1. The CMS retrieves the document from the document store. Note that Classification Server is able to process (natively) files in many different file formats (for a list of supported file formats please see Appendix D).
2. The CMS sends a correctly formatted request (containing appropriate document information) to Classification Server for classification.
3. The Classification Server returns the classification information in XML format.
4. The CMS updates the metadata information in the document store for the document.

It is recommended that this process is performed “off-line” and not as pages are displayed from the web site (for performance reasons). A strategy would be to perform a one-off update of the CMS when the Classification Server is first implemented then submit documents to Classification Server only when they are modified (or added) by the CMS.

Pre-Requisites

The Classification and Language Service must be installed and running before it can be used (as specified in this document). It should be noted that the computer on which it is installed must be accessible from the machine from which it is to be accessed including any relevant ports open on firewalls, etc. Additionally, any request submitted will need to know the machine name and port to specify when sending the request (specified as part of Classification Server configuration which is not discussed in this document).

Additional documentation available

The following documentation provides additional information that may be helpful to the reader:

• Semaphore Classification Server Rulebase Reference