RAG has been producing hard coal by underground mining in Germany until the end of 2018 and continues to be responsible for the resulting consequences in North Rhine-Westphalia and Saarland. Since coal production stopped, RAGs headcount was sharply reduced and many long-term employees are close to retirement. Effective ways to preserve expert knowledge and efficient ways to access it are therefore in high demand.
RAGs company wide search engine is called DSA (“Digitale Service-Akte”). It collects tags from multiple other systems (e.g. GIS, SAP, ELO) and thousands of documents (e.g. Sharepoint, shared folders, filesystems). Some of the documents are of classic office document types, but also different kinds of GIS and map documents are common. A user can search for tags and will be provided with list of search results. Results can be filtered in multiple ways, e.g. by source system or time period.
But still, not all search requests can be answered correctly. Especially with more complex searches it is often faster and more efficient to ask an experienced colleague. With this use case we want to make the results of DSA as smart, complete and precise as the answers of your colleague sitting next to you.
Some ideas to achieve this goal are:
- use of natural language to define a search query
- optimizing the list of results automatic evaluation (eg. by a synonym finder, pogo sticking, ...)
- optimizing the list of results by manual evaluation (eg. by selecting “5-star” results or by manual “pinning” of results)
- enriching results by external data sources (eg. Wikipedia, research catalogs, …)