Techniques for The Optimal Usage of Already Existing Knowledge

• Client: Leading international manufacturer
• Sector: Automotive

As one of the largest suppliers of premium passenger cars and commercial vehicle manufacturers, the company sells its vehicles and services in almost all countries of the world.

• The challenge:

  The client has an extensive archive of research documents, which differ significantly in terms of both quality and format. Due to the enormous size of the archive, the manual viewing of all documents can no longer be granted. Furthermore, the documents should be classified thematically and further explorable through this classification. Since the documents were all considered as confidential, the client could not provide direct access to the data. Instead, software was to be provided to help the client to execute this task himself. Therefore, the client not only wanted to present an analysis software, but also to add an interactive user interface.
  

• Our solution:

  KÖNIGSWEG consultants relied on robust and proven methods of computer linguistics to meet the challenges. The aim was to find an application that would produce helpful results for a wide range of data situations and that would also be reliable without direct access to the data.
  For each document an informative content summary was automatically generated, keywords and relevant items were highlighted. By using a cluster analysis, KÖNIGSWEG put the documents in relation to each other and thereby revealed thematic connections. All results of the data analysis were made accessible through an intuitive user interface - also a detailed exploration of the data through thematic clustering, keyword search and automatic summaries was made possible. Throughout the entire process, KÖNIGSWEG was in direct communication with the client. This meant that ideas and any concerns regarding natural language processing could be taken into account at any point and be considered by using algorithmic implementation.
  

• Why this case could be relevant to you:

  A large collection of texts, often the case in large and established companies, can very quickly reach a critical size. A manual review is no longer sufficient or, in the worst case, becomes impossible. KÖNIGSWEG's expertise, both in statistical and linguistic methods, cannot just evaluate the hidden knowledge for you. In close cooperation with our clients, we provide suitable tools for you to explore the sources of knowledge yourself and to gain new insights. Through robust algorithms and an appealing visual presentation of the data, you will be able to view, classify and use new and old files long time after the project is completed.