Funnelback is a feature-rich search solution with high quality out of the box rankings, customisable results presentation, Contextual Navigation, Faceted Navigation, flexible reporting and a highly usable administration interface. For more detail on these, and additional, features of Funnelback search, click on the feature categories below.
Key Features of Funnelback Search
High quality and tunable ranking algorithm
The algorithm determines search results by applying various ranking criteria. Ranking can be automatically and manually tuned to increase the value of search results to suit your business and user requirements.
Intuitive administrative interface for configuration and reporting
Funnelback is managed via a highly usable, browser-based administration interface. The interface allows the search administrators to have complete control over ranking, features, styling and reporting.
Flexible Results Presentation
Funnelback search results are presented in a completely customisable, user friendly format and can use in-built features such as Faceted Navigation and Contextual Navigation to improve the usability of the results page.
Speed and Scalability
The Funnelback search engine can index millions of documents with fast query response times.
Funnelback can be used as a platform for developing complex search applications such as multimedia search, (including video, audio and images), related content (e.g. related news stories), e-commerce, geospatial search, email based media alerts and plagiarism detection systems.
Contextual Navigation allows users to navigate search results via dynamic sub-topics, which are derived from the content without the need for metadata.
Funnelback uses metadata fields or other structured information to provide users with related options for refining their query, by file type, for example.
Funnelback’s Pattern Analyser
Funnelback's Pattern Analyser senses changes in market demand and enables performance based management. It can measure the effectiveness of a marketing campaign by reporting the magnitude and duration of a spike in search traffic.
Combined full-text & metadata search
Allows search over unstructured data with a sophisticated ranking algorithm returning high quality results. Use metadata to up weight results or deliver faceted and geospatial searching.
Scalability & speed
Search many millions of documents with fast query response times.
Search many repositories enterprise-wide with a single query via a simple web interface. Separate indexes for each repository are created and users are able to choose which repositories to search over.
Integrates with Content Management Systems and Portals.
Search results with spatial data presented on a map.
Search audio and video metadata.
RSS or subscription search
Allows users to keep up to date.
Key word searching
Available across document content and metadata.
Provides alternatives from the Funnelback dictionary.
Provides a list of search suggestions based on a custom thesaurus.
Specified pages are promoted when triggered by specific keywords.
Documents are ranked by their relevance score.
Boost the rankings of important sites or pages. Up-weight or down-weight metadata, search a query term in a URL or use the ‘same site suppression’ feature to create a valuable diversity of results from different content.
Sorting of search results
Results can alternatively be sorted by date, title, size, URL as well as any other specified metadata class, in ascending or descending order.
Provides users with a set of sub topics related to their search query without the need for metadata. The sub topics provide users with query refinement suggestions derived from the content.
User click feedback
Provides a mechanism for search users to influence the result relevance over a period of time. By monitoring the search patterns of users, and the documents which they go on to access, Funnelback can up-weight these documents for the particular query.
Provides the ability to expand query terms or create synonyms to match up with an organisation’s terminology.
Intuitive web administration interface
Create, configure and monitor the status of managed collections.
Crawler and indexer configuration
Provides the ability to create new collections, specify spider rules and start/stop the crawling process, all from a single administrative web interface.
An offline view is maintained for indexing purposes. This ensures that search operations aren’t affected whilst the collection is being updated. Once it is updated it’s switched to the live view.
Full/Incremental index update
Updates to the collections can be triggered through the administrative web interface including; Full update (a complete crawl from scratch); Restart crawl from the latest checkpoint; Filter: extracts textual information from specific file formats; Index: indexes the existing offline crawl data and then swaps the live and offline views; and Swap views: swaps the live and offline views.
Scheduled updates of indexes
Provides the ability to schedule the updating of collections on an hourly, daily, weekly and monthly basis.
Inclusion and exclusion patterns
Restrictions to the crawl are governed by a set of inclusion and exclusion patterns. External metadata indexing. Assign metadata to index records using a central configuration. Assign additional metadata to other pages.
Crawls can be set to update only parts of the index which have changed since the last crawl and create a balance between freshness and infrastructure load.
Funnelback Pattern Analyser and Reporting
Automatic reports on changes in market demands. Funnelback’s Pattern Analyser can measure the effectiveness of a marketing campaign by reporting the magnitude and duration of a spike in search traffic. It can also provide intelligence on the geographic regions where the change in search pattern is occurring and can provide context by linking each change to online news articles on a related topic.
Extensive query reports
Monitor search behaviour and usage over a short or extended period of time. These include: monthly summary of queries, most commonly searched for keywords, top scopes used in queries, query terms which return zero results, queries per hour and documents per site. Date based reporting is also available.
Provide statistics and reports on data gathered during the update process.
Customisable User Interface
User interface templates
Use templates as a starting point to customise the search interfaces or developers can tailor the pages to suit individual organisations.
Results as HTML, XML or RSS
Search result tiers. Results are grouped into tiers based on how many search constraints the results have satisfied. Tiers are sorted in descending order with regard to the number of constraints satisfied (e.g. fully matching, matching 1/ 2, etc.).
Results summaries relate to the current query with optional query highlighting.
Summaries can include metadata fields.
Cached document word highlighting
Provides a cached view of the selected result with search term highlighting.
Cached document view
Indexed documents are cached as part of the indexing process and are stored as plain text documents so binary formats such as PDF, Word etc. can be viewed as html through a standard browser. Search terms can also be highlighted with this cached document view.
Funnelback conforms to AA Web content accessibility guidelines.
Design results pages
Search results can be customised to match your required look and feel.
Search metadata fields sourced from the indexed documents. DC (Dublin Core), AGLS (Australian Government Locator Service) and Netscape metadata are supported by default or define your own which correspond to a URL prefix.
Numeric field searching
Range searching of numeric fields.
A customised search enabling results to be restricted based on a set of inclusion and exclusion patterns.
Boolean operators provide the user with a simple means to combine words, phrases and sub queries with Boolean operators, to either widen or narrow a result set.
Use the phrase operator which requires component words to appear consecutively and in the order specified.
Allows user to constrain the results to documents which were last modified within a given date range. Word stemming. Stemming (grammatical variations) allow the users to search for common variants of the specified word(s). Wildcard search. A wildcard search uses truncation operators (‘*’) to provide results containing variations of the search terms. Proximity search. Proximity search shows results where query words appear near each other.
Intranet & internet
Crawls HTTP and HTTPS sites with support for cookies.
Different repository types
Also indexes Shared Drives (Microsoft and Novell), Oracle, MySQL, MS SQL Server, Lotus Notes/Domino, SharePoint 2003/2007/2010, TRIM, Microsoft Exchange, Alfresco, Documentum, MatrixOne, FTP, Quickplace, IBM Content Manager, DB2, FileNet, Websphere, Interwoven, IRIS Archea, JCR, Open Text, Oracle Stellent, Symantec Enterprise Vault, WebDAV and Xerox Docushare. Support for additional document sources can be developed as required.
Windows & NFS.
Multiple file types supported
Including HTML, XML, Text, PDF, Word, PowerPoint, Excel and RTF. Ability to plug in custom filters.
Multiple languages supported
Collection level security
Security controls that restrict access to sensitive information in repositories and business applications to authorised users.
TRIM, file share and intranet document level security
Funnelback supports NTLM authentications for searching TRIM and NTFS based shared drives.
Enables access to individual collections to be restricted on an IP address/hostname basis.
Search users who are unauthorised to access a given collection can be redirected to a ‘public’ collection, which will contain publicly accessible content.
Secure sites can be crawled by providing user credentials required by the web servers.
Secure searching of information protected by NTLM.
Access all content via a single web interface.
Speed / Highly efficient
Fast indexing and query processing.
Every step of the Funnelback workflow can be expanded and customised.
Seamlessly integrates into your business interface.
Leading edge technology
The technology was developed by CSIRO, Australia’s premier scientific and industrial research organisation. Heavy investment in research and development continues.
Tailored to your needs
Funnelback resolves complex search problems which are not solved by a one size fits all system, no matter how large or small your organisation.
Quality customer service and support
A dedicated support team with access to research and development resources on site. 24/7 support packages available.