A biotechnology company leveraged the Semaphore Semantic AI platform to improve knowledge worker productivity, enhance customer experience and satisfaction, and optimize their SharePoint repository to efficiently manage and display enterprise information using an auto-classification approach.
Information associated with the drug development pipeline was spread across multiple file shares, sites, and libraries throughout the organization. It was stored in a variety of formats, and used diverse vocabularies. Sites and libraries contained multiple folders, which often contained duplicate content, resulted in wasted time and effort locating information and a less than valuable user experience.
"Thanks so much for the hard work re: the POC. It’s great to see something move from a conversation in a conference to demonstrated value."
Using Semaphore Knowledge Model Management (KMM), they built a knowledge model that reflected the products, development activity, stage of development, and other relevant topics associated with drug development. To ensure the model reflected their domain and use case, they leveraged Semaphore text mining capabilities to examine their content, identify its context and meaning, and use it to enrich the model and enhance classification outcomes.
Classification results were iteratively validated using a representative subset of the model (3 – 5 facets) and a predefined corpus of information (approximately 100 documents). As each iteration was processed, the model and classification outcomes were reviewed and modified to further enrich the model and refine classification results.
Part of the initiative was to validate their SharePoint on-premises implementation to ensure their future migration to O365 would be successful. Semaphore was configured to classify a single SharePoint knowledge library, which had a set of subsites. A broad model, fed from Semaphore, exists in the SharePoint term store to support tagging and autocategorization.
Today, required information can be quickly located saving knowledge workers time and effort. The validation of the SharePoint implementation was successful and classification results achieved or exceeded initial target results in the auto-tagging of content.
Life Sciences organizations can improve knowledge transfer across all stages of product development and manufacture with Semaphore. Taking a product through research and development, prototype manufacturing to industrial scale manufacturing is a process that can take anywhere from three to 10 years and can involve millions of documents. Semaphore’s model management, auto-classification, and fact extraction technologies transform enterprise knowledge, accumulated throughout the R & D process, into actionable intelligence that organizations leverage to gain insight, drive profits and improve efficiency.