
Industrials
synvert implemented an NLP solution for a German car manufacturer, using technologies like Apache Solr and Spark to automate report processing, improve problem identification, and enhance failure prevention through efficient data ingestion and topic modeling.
Initial situation
At the customer, a German car manufacturer, mechanics were tasked with collecting repair information by manually writing reports.
Going over the reports and combining the information to create insight was tedious and inefficient.
Architecture
For the solution a Banana Framework and technologies such as Apache Solr and Spark are used.
Generated benefits
It is much easier to understand which problems are occuring, and group them together in order to gain actionable clusters.
This allows much quicker reaction and more robust failure prevention.
Services accomplished by synvert
synvert created an ingestion mechanism including OCR digitization, artefact removal and text cleansing. This mechanism has since been fed by well over a million reports.
As a final step in the Natural Language Processing pipeline, topic modelling was used to tag the documents according to their content.
Furthermore, synvert built a web UI for efficient retrieval and exploration of the parsed documents.