“conceptClassifier delivers automatic conceptual metadata generation, auto-classification, and powerful taxonomy tools to improve the use of and access to content assets. The classification of content can come from multiple sources including SharePoint, file stores, Exchange Public Folders and websites.”
- Reduce time to develop and deploy
- Improve findability and reuse of information
- Improve collaboration across time zones and overall productivity
- Normalize complex terms and suggest appropriate terminology
A global professional services firm providing planning, engineering, program, and construction management services, in over 80 countries with over 9,000 staff servicing the public and private sector.
As with most organizations, increasing complexity and collaboration across times zones and organizations was a challenge as well as providing market coverage across key geographies 24/7. This organization recognized that they were faced with information overload, finding and sharing information was no longer manageable, and content could not be re-used and re-purposed.
What were the problems?
- Data replicated across various silos
- Inconsistent tagging of documents and content by end users
- Vocabulary normalization caused issues across global boundaries
- Lack of taxonomy automation and management
- No auto-classification of content to a consistent structure
The organization had already identified that even a robust enterprise search solution did not provide the framework to support a global distributed business. It had come to the conclusion that it needed a knowledge management framework to address the challenges it was facing. It selected the conceptClassifier platform due to it’s ability to deliver automatic multi-term metadata generation, auto-classification, and powerful taxonomy tools for the business professional to improve the use of and access to content assets. Classification includes diverse file types, applications, or content repositories regardless of where the information resides.
The organization had selected SharePoint search as their enterprise search engine. This search solution does have the ability to identify some concepts using linguistic analysis, but it does not provide a comprehensive framework for automatic document classification, nor can it identify all concepts that are found within the content.
Utilizing Concept Searching’s unique intellectual property, compound term processing, metadata is automatically generated and consists of multi-word terms that represent a concept or a subject. This approach establishes an enterprise standard of metadata that is consistent, reusable, and trusted. Since search is no longer restricted to keywords in isolation, the ambiguity in single words is eliminated and search results are more accurate and inclusive as documents that share similar or related topics will be retrieved even if the precise terminology is not used in the search string.
As part of the document management process, the content is auto-classified to a taxonomy where it can be maintained and managed by business professionals. From a search perspective, the rich multi-term metadata is fed to the search engine index, enabling concept based searching. The information retrieval process is significantly improved. With 9,000 employees scattered across the globe, the ability to search using phrases was entirely transparent to the end user, and required no training.
Because of the ability to capture concepts that are found within the content, during the search process similar content will also be identified. In a typical search scenario, this information would not be found as it didn’t match the query. When maintaining the taxonomy, the software suggests ‘clues’ that were found within the content, but were not identified by the taxonomy administrator. Using these clues plays an important role in the disambiguation of terms and vocabulary normalization, a requirement sought by the client.
Concept Searching’s technology enabled the creation of a global knowledge management framework platform, eliminating information overload and improving productivity.
The solution delivered the ability to:
- Create taxonomies using natural language concepts
- Reduce time to develop, deploy, and manage
- Consistently tag content with conceptual metadata
- Improve the usability of content
- Eliminate end user tagging
- Store content and metadata automatically in SharePoint
- Search across multiple data sources
- Connect data silos
- Enable enhanced knowledge management and expertise finding
- Navigate results in a hierarchy that mirrors the organizational processes
- Be a platform to leverage their SharePoint and search technologies