Enterprise Search Requirements

•  Support federated search of multiple sources, including file servers, web, e-mail, enterprise application data and reports, records within databases and the content management system.

•   Have the ability to control access to items of content from search and/or site map indexing based on AOC business rules or security roles (i.e. users shall only see results to which they have access).

•   have the ability to integrate simple and advanced search screens and result screens into the user interface.

•   control the discoverability of content items by external search engines.

•   have the ability to generate a plain English URL for key web pages for easy referencing by non-technical end users and search engines.

•   have the ability to interface with thesauri or allow import of thesauri data into search engine thesaurus for browsing, selection and searching.

•   provide a soundex (sounds like) capability in the thesaurus function so that misspelled terms can be related to their correct spelling.

•   provide comprehensive search facilities across the entire website, multiple sites or sub sites to support content publishing.

•   retrieve results in a timely fashion: goal is to ensure that tool adoption will not be negatively impacted by the speed at which it returns results.

•   support “pre-processing” of all search terms against the thesauri and automatically include related terms in the search.

•   have the ability to query and retrieve standard metadata.

•   support automatic indexing, keyword generation, and full-text indexing.

•   support comprehensive external search facilities across the entire website, multiple sites or sub sites for end users.

•   support key word and metadata search.

•   support multilingual search.

•   support separate indexing of content items for each supported website.

•   support the automatic modification of search results based on site usage and search patterns (e.g. weighting the most popular sites/search results selected by the user to appear high in the search results list).

•   support the measurement of the use/effectiveness of each keyword search query entered by a user in searching the website. This capability may be provided by a third party web metrics facility.

•   support natural language search interaction

•   support case independent search

• Systems shall conduct present partial matches (2 words out of more) after exact matches and occurs-in-page matches

• Support the use of Boolean operators (and; or; not) and proximity operators (near; with) to refine searches and accept phrases for full-text searches.

•   provide the ability for the CMS to control display of and access to items of content from search and/or site map indexing based on AOC business rules or security roles (i.e. users only see results to which they have access)

•   provide the ability to search by learning object or learning object type

•   support indexing of text and common attachments (including Microsoft Office documents, Adobe PDF, etc.) within the CMS and across a number of repository types such as file systems and databases.   Detail any formats not supported by system indexing.

•   allow end user to create and save searches.

•  provide a user interface to allow users to customize the sources included in a given search scope.

•   support highlighting of search terms within a result set or document.

•  provide control parameters that allow the administrator to control connection performance to content so that, for instance, a live Web server indexing can be throttled back so that indexing does not crash the server (but a file server may have a different control setting which allows for potentially faster retrieval during indexing, depending on network connectivity.)

•  provide configurable search indexing schedules to allow administrators to set indexing at various times so they do not conflict with scheduled downtime of servers (for nightly backups).

•   provide reporting features to enhance search engine results sets and design.