Office of Acquisition and Grants, 1710 GWYNN OAK AVENUE, BALTIMORE, MD, 21207-5279

70 -- INTERNET/INTRANET SEARCH ENGINE SOFTWARE/HARDWARE DUE 030399 POC JERRY BURTON, 410-965-9487 WEB: Download Solicitation when/if available, http://www.ssa.gov/oag. E-MAIL: Contracting Officer, SSA/OAG, JERRY.BURTON@ssa.gov. The Social Security Administration (SSA) is interested in obtaining information from vendors who can provide and support an Internet/Intranet Search Engine that has high- performance and "portal" functionality. Intranet portal to include search, taxonomy, links to internal sites, and personalization. The requirements are itemized below within five categories. This hardware/software will be used on equipment located at SSA's National Computer Center, Baltimore, Maryland. At a minimum, the portal/search engine must provide the following functionality: RESULTS -- obtain precise results in a timely manner; a search across gigabytes of data should produce results within seconds even when searching across disparate sources and disparate formats -- allow search navigation facilities including clustering, relevancy ranking, automated document summarization and query by example -- allow user to customize the display of search results by using a variety of ranking options such as date or relevancy—allow user to customize the results as to sort order, number of records returned per page, etc. -- automatically monitor documents and alert users of changes in or updates to predefined profiles -- provide automatic delivery of content for user notification when new information from a specific source becomes available -- provide knowledge application tools that enable users to identify most frequently sought references for a given source -- results should be returned in a format that maximizes the ability to cut and paste into MS Word and Word Perfect documents so that quotation marks, section symbols, footnotes, and formatting such as underlining, bold, and centering are retained. -- provide clear, meaningful end-user "help", error messages, and an on-line tutorial -- provide automatic highlighting of search terms and linguistic and thesaurus equivalents -- results should contain the document title or file name, date and byte size, an extract of the document, a link to the document, and where that document resides SEARCH -- provide various search and query capabilities including key word, query by example, natural language query, and spelling error override -- use a combination of natural language in various tense and form, Boolean commands or proximity operators within a single query -- search against multiple formats across many servers, including CD-ROM's simultaneously (ability to operate in a distributed, multi-host environment) and merge the results into a single list that provides the best matches first -- allow the user to selectively search data sources to search either all or specific repositories of information -- provide field search capabilities to navigate quickly through pre-defined document attributes -- allow advanced query expansion and thesaurus expansion -- allow custom thesaurus creation -- must have a search interface that is fully customizable to allow various groups the searching options that best fit their needs -- provide Alpha/Numeric type search of instructional sets -- span multiple servers, both NT and Unix, with one key word search -- search a collection of complex research material -- consolidate information repositories (both multiple repositories and/or the repositories themselves) to enable searches across a broad range of enterprise data and maintain its own index stored apart from the documents -- search products index user-definable fields (i.e., author, title, and date) -- support the Bayesian inferencing network that combines multiple source calculations involving the frequency of the appearance of specific terms within a specific context to determine a relevancy score -- ability to capture queries and search results to create an additional body of information to be searched -- allow fuzzy search capability -- allow browsing on any platform-specific documents without the original application INDEX -- update indexes automatically with an option for manual indexing -- allow document indexing, classification and retrieval to make simultaneous use of full text and metadata search -- understand and correctly index a large variety of formats including but not limited to html, sgml, gif, tiff, bmp, cgm, jpeg, pdf, Lotus notes, Lotus 1-2-3, e-mail formats, Microsoft Office, news feeds, ODBC databases, Word Perfect, as well as other industry standard formats -- establish "personal briefbanks" [personal user indexing] and/or regional briefbanks so user can search his or her hard drive or local server and index their own work product -- automatically index any website into a searchable collection or index to provide SSA's public users with an information domain of related information located at other websites across the Internet and Intranet -- provide an enterprise-wide search and indexing server -- allow indexing on multiple domains, run multiple indexing jobs against different collections, schedule automatic indexing tasks on a regular basis, and provide user authentication during indexing—provide online incremental indexing so only new, updated or deleted documents require index changes -- provide knowledge application tools and enhancement navigation -- provide spidering capability to at least 20 servers SECURITY -- provide assurance that documents can be kept secure, not accessible by other components -- limit access to certain documents to certain employees ADMINISTRATION/CONFIGURATION -- allow easy configuration with Netscape Enterprise Server (Internet Web Server software) and Microsoft Internet Information Server (Intranet Web Server software) -- have plug-and-play designed to allow additional functions and components to be added at any time. -- administration must be relatively simple and allow organization and integration of documents contained within the Agency's Internet and Intranet web servers -- run centrally and not need a client workstation piece -- be Year 2000 compliant -- include full documentation and technical support which includes documentation for specifying results and document presentation -- provide a User friendly interface that is both flexible and powerful -- be fully scalable -- allow administration from any browser -- provide plain English interface and support query-by- example -- fully compatible with the NT 4.0 platform, all popular UNIX platforms, and Microsoft and Netscape browsers -- provide support for Microsoft's ODBC and Solaris' Intersolv DataDirect Connect -- Web servers support must include Netscape, Lotus Domino and Microsoft IIS -- allow installation to be wizard driven to ensure quick and easy installation -- adhere to Accessibility Compliance with American Disabilities Act -- ability to complete all functions in Spanish -- provide a platform to easily manage large and complex information environments on the Intranet/Internet Any vendor possessing the requisite expertise and resources is invited to submit information concerning their capabilities, experience (including current customer contact names and phone numbers), and other pertinent data (including GSA contract information) within 15 calendar days from the date of this publication. Any literature, or information submitted as a response to this notice, must contain enough information attesting to the capability of a product offered to meet SSA's requirements. This is not a request for proposal and the Government does not intend to pay for any information provided. While receipt of responses will not be acknowledged, information submitted will be used to broaden SSA's awareness of available products/solutions. When responding to this announcement, respondents should refer to Search Engine/Portal Software/hardware. Posted 02/11/99 (D-SN297804). (0042)

Loren Data Corp. http://www.ld.com (SYN# 0435 19990219\70-0011.SOL)