Semantic information is a living, evolving part of your content.

Key RSuite components that can be used to semantically enrich content:

• API-driven architecture

Anything possible in the RSuite application through user activity can also be driven through the API. This makes it possible to integrate semantic enrichment into your solution in the manner best suited to your requirements.

• Robust, configurable workflow engine

The included workflow engine can combine automated and manual tasks, allowing workflows to be configured according to your current needs. Additionally, the engine can be used to organize and track future reprocessing of semantic information if necessary. Information from these workflows is made available for reporting so that you can better understand not only the current state of your content but also past personnel, tools, and processes that helped develop it.

• Customizable event model

RSuite’s event model lets you set conditions for triggering semantic processing. This could be in response to changes in the content itself, changes in an external source of semantic information at a predetermined future time, or virtually any other condition accessible to RSuite directly or through an API.

• Flexible metadata forms

RSuite does not assume that everyone will interact in the same way with all of the metadata and semantic information relating to an asset. Instead, you can configure any number of specialized forms that can be used to present users with a subset of information relating to an item. In this way, you can avoid presenting users with a bewildering array of fields that are not relevant to their particular activities. Using targeted forms will ease the burden on your users and can raise the quality of the content being managed.

• Taxonomy controls

RSuite has optional taxonomy controls available to allow users to assign values for a formal taxonomy to your content as well as controls to allow users to search and navigate content through the assigned hierarchy. Our tools also allow easier assignment of free-form “tags” for cases where a formal taxonomy isn’t practical. Of course, these tools work in conjunction with RSuite’s flexible metadata model to address simple as well as complex requirements when needed.

• Semantic information about the keywords/concepts
• Answers to the searcher’s questions
• Lists
• Information based on relations among the keywords/concepts
• Different kinds of — but related — information such as events, persons, locations, or organizations

A user can search and discover content by keyword, user, date, document identifier, XPath, customer-defined and system metadata, and more. RSuite search services provide industry-leading search capabilities covering full-text search, relevancy ranking, integration with non-search systems currently used, metadata-based faceted search, search logs, custom search, proximity search, and targeted search. Wildcards such as * and ? are available. System metadata filters include creation and modified dates, file type, XML element type (if relevant), folder type, etc.

We support full-text search of XML, Microsoft Office, Adobe PDF, and HTML artifacts. We include metadata values when performing full-text searches, enhancing findability and content reuse.

• Initiate and track enrichment activities

The workflow engine in RSuite can be configured to ensure consistent handling of content. Based on requirements, customers can initiate and track manual semantic enrichment assignments to appropriate personnel using the workflow’s task assignment capabilities. Workflow steps can also call on external sources of semantic information, such as a third-party database or automated enrichment engine. The workflow can be configured to mix and match automated and manual enrichment tasks according to your business requirements.

• Adapt or replace a semantic information model

Semantic information rarely can be treated as a monolithic, static set of information. Most organizations need to plan for updating either the metadata and semantic models, the actual data itself, or both periodically. Tools that may not be practical today may be commonplace in the future. RSuite’s configuration can be adapted over time to assist in adapting or changing the semantic models using the workflow engine, custom forms, and reporting. RSuite’s flexibility enables it to support multiple approaches simultaneously. For example, you might decide to maintain legacy content using its original semantic model but apply a new model to new content.

• Apply “expensive” processing selectively

RSuite’s event system can help manage processes that may be too expensive to be practical for your entire content repository. For example, you may decide only to apply certain resource-intensive processing to content when it has been selected for inclusion in a publication or added to a distribution package. RSuite’s event handlers can be used to add conditional logic enabling the use of high value-add processing that might be impractical to apply to the entire repository.

• Automate quality assurance checks

Most organizations use labor-intensive manual quality checks on content that can be error-prone and is difficult to scale. It is straightforward to organize these manual efforts into a workflow that can be tracked and recorded. Over time, however, automated steps can be added to take advantage of RSuite’s processing capabilities to catch and even correct many problems in content automatically. RSuite provides a platform to pursue new avenues of efficiency while increasing the overall quality of the content you produce.

• Review your semantic strategy

Organic content collections over multiple silos often prevent organizations from evaluating their use of semantic information for effectiveness and applicability. With RSuite as your content platform, it is possible to develop reports and queries that can assist in identifying strengths and weaknesses in your semantic practices. RSuite’s tools can be used to ensure these changes are applied consistently and efficiently.

Once loaded into RSuite:

• The taxonomy itself becomes available to populate the taxonomy control in search forms and metadata forms.

Thus, your taxonomy becomes content just like anything else in your system. Furthermore, it is maintainable by typical users rather than requiring help from an engineer every time an update is needed. This is important because good data governance requires ongoing monitoring of user search queries and results, then tuning of vocabulary and content classifications.

• The taxonomy model can also be used to validate metadata field values when data is loaded or created and expand metadata values from taxonomy IDs to their text equivalent.

RSuite also has the ability to support the bulk classification of assets using the bulk actions menu or via importing metadata assignment worksheets.

• Custom taxonomy-based searches can be configured within RSuite to allow the search and retrieval of assets and other objects meeting taxonomy subject search criteria.

Our versatile browser control supports browsing in terms of the taxonomy tree as well as search and multi-select for users responsible for manual classification. It can even be repurposed to display other forms of hierarchical information such as content outlines and lists. This is especially useful for developing UIs needed for custom link management applications.

• Automatic Tagging

Enable an enormous increase in productivity for developing artifact correlation and search/retrieval solutions by removing the manual process of tagging content. Automatically correlated content minimizes irrelevant artifacts from hindering artifact discovery. Manual tagging can require more than 100 hours for a single document; our semantic data services can perform the same task in seconds! Over a large document cluster (50-150), RSuite can save you thousands of hours in employee productivity.

• Semantic Navigation

Eliminate hours of searching through irrelevant artifacts, which can be particularly challenging when the artifact corpus is large — say, 100M+ text documents, videos, audio files, and/or images from 1,000+ sources.

• Targeted Search

Classify objects in accordance with client models which can preserve organizational knowledge within existing IT systems. Enable implicit relationship extractions, which improve search accuracy.