“Although a cliché, delivering the right information to the right person in the right context still has not come to fruition with search engines. Unproductive workers cost US businesses almost $600 billion per year. Without meaningful metadata and categorization, the cost is going to continue to rise.”
The conceptClassifier platform and conceptClassifier for SharePoint enabled the university to auto-classify all content, regardless of where it resides, to generate semantic metadata, and to normalize complex medical terms to make it easier for site visitors to find what they are searching for. Because the solution runs natively in the SharePoint environment, all automatically generated metadata could be managed through the SharePoint administration interface.
- Normalization and disambiguation of complex medical terms
- Ability to automatically assign or change a content type, based on the meaning within the content
- Automatic suggestion of appropriate terms
- Ability to search multiple repositories concurrently
- Elimination of end user tagging
This medical university’s mission is to improve the health of the public by advancing medical knowledge, providing outstanding primary and specialty care to the people of the region, and preparing tomorrow’s physicians, scientists, and health professionals. It is nationally recognized and ranks among the top medical centers in the country.
The existing intranet site provided a free text search to allow users to find information. As with most free text search solutions, end users would get too much irrelevant information with little ability to filter out unwanted results, leading to a longer search for information and a less than satisfactory user experience. As part of a medical center, end users had many diverse objectives when searching for information, depending on their role. For example, a researcher may be looking for information that is quite different information that needed by an administrator, nurse or doctor, yet the environment provided no way to assist users in finding exactly what they needed.
End users needed to identify content in the context of what they were seeking. The fundamental problem with most search solutions is that they are based on an index of single words. Yet most searches are expressed in short patterns of words and single words in isolation, which creates ambiguity.
What were the problems?
- Keyword search was insufficient
- End users did not always know exactly what they are looking for
- Most users were not adept at performing complex keyword searches
- Ambiguity of words – one word can have many meanings, two or more words can share the same meaning
The university selected the conceptClassifier platform and conceptClassifier for SharePoint due to the transparent integration with SharePoint and the ability to automatically classify and generate semantic metadata. The technology also enabled organizational taxonomies to be easily developed, deployed, and managed through the taxonomy component, conceptTaxonomyManager.
A key driver for the solution was the underlying compound term technology. Compound term processing is a new approach to an old problem. Instead of identifying single keywords, compound term processing identifies multi-word terms that form a complex entity – for example, colorectal cancer is also called colon cancer or large bowel cancer – and identifies them as a concept. By forming these compound terms and placing them in the search engine’s index, the search can be performed with a higher degree of accuracy because the ambiguity inherent in single words is no longer a problem. As a result, a search for ‘triple heart bypass’ will locate documents about this topic, even if this precise phrase is not contained in any document. A concept search using compound term processing can extract the key concepts, and use these concepts to select the most relevant documents. Content that shares the same concept will be retrieved, even if the search terms do not match.
conceptClassifier for SharePoint enabled the university to auto-classify diverse repositories, both internal and external, to generate semantic metadata, and to normalize complex medical terms to make it easier for site visitors to find what they are searching for. Also, all automatically generated metadata could be managed through the SharePoint administration interface, significantly reducing the complexity of the solution. conceptClassifier for SharePoint is fully integrated with content types, and can automatically change or assign a content type based on the meaning found within a document.
Additional findability features included in the solution were the ability to automatically suggest the appropriate terms for end users, eliminating the need for complex medical terms and associated syntax to be entered. The result is the ability to deliver relevant and precise information.
Staff at the university, regardless of their role, can now find the information they are seeking on the intranet, improving information retrieval and maximizing the investment in SharePoint. The solution improves productivity and delivers the right information to the right person at the right time.
Within a healthcare setting, accurate and relevant information retrieval cannot be on a wish list, it must be a requirement. After all, peoples’ lives may depend upon it.
- Ability to identify relevant content to assist end users, regardless of their organizational role, to find relevant and accurate information
- Improves productivity through enhanced ‘findability’ in search
- Enables existing content to be discoverable for reuse and repurposing
- Ability to normalize complex medical terms, enabling end users to more easily find what they need
- Normalize the organizational vocabulary, and automatically suggest appropriate terms to end users
- Native integration with SharePoint, reducing complexity and training, and simplifying ongoing management