How to turn large amounts of text into insights with machine learning
Lots of important information in companies are only available in text form. Methods of automated text analysis methods help to leverage this treasure trove of knowledge. Whether user-generated content from the web, customer inquiries by mail, or contracts and specifications: A lot of information is available in form of unstructured texts. As this volume of data grows ever faster, it is no longer possible to simply "read along". The solution is to automate the processing and analysis of such documents. NLP, clustering, topic modeling, classification, universal embeddings – there are currently many different AI technologies for the automated processing of text volumes. Some of these technologies solve very different problems and can achieve their greatest performance in combination. The speaker will use a concrete example to explain the stages of a text analytics project and show which techniques are suitable for which areas of application and which new use cases are thus possible. The session is aimed at developers, business users, managers and data science beginners who want to better assess the potential of automated text analysis for their own applications.