Make internal knowledge base digital

Organisations build their knowledge throughout their business history and it is the most valuable asset to them. I have witnessed a national library has digitalized their diet collections and offers public users to search information by keywords or images (https://dl.ndl.go.jp/). Some organisations can monetize professional information by providing powerful search capability to users (https://www.rakennustietokauppa.fi/). Some industrial technology companies support their engineers or technicians to access technical reports while doing maintenance or repairing services for their clients.

I have supported couples of my clients to turn a huge amount of documents into digital format and to make them accessible by users through semantic search. Typically, all the document in multiple formats is turned into text using an OCR engine and indexed into an information retrieval system. Users can send a query and receive the needed documents, together with recommended content.

For open-source technologies lovers, we can use Tesseract as an OCR engine to extract text content from documents, and ElasticSearch as an information retrieval engine with semantic search capabilities.

For ones who want to use enterprise tools, we can use e.g. AWS or Azure services. Azure Cognitive Service or AWS CloudSearch can be used to ingest and manage a huge amount of documents in multiple formats and from multiple sources. It provides semantic search capabilities to the users.  These AWS and Azure services are high performance, scalable and reliable.