In the pharmaceutical industry, progress is enormous. Every week, tons of new studies and scientific articles are published. How can one keep up with this flood of knowledge? Chemists, pharmacists, biologists and other scientists mark the articles in their literature databases that are relevant to them and those that are not. If there are 15 of them, that’s no problem – but what if there are 250? And this is where Averbis literature classification with AI comes into play.
Text classification is a process of assigning labels or categories to documents depending on the document content. The wide field of use cases for text classification in the healthcare and life science industry includes literature reviews, competitor monitoring, adverse events detection, regulatory intelligence, technology scouting and many more.
Automated classification: reduce effort up to 85% and save time
Automated classification can drastically reduce manual effort up to 85% and save time – for example, by filtering irrelevant documents in an automated process. Averbis offers a machine-learning based text classification, which can be easily trained on your documents and use cases. It leverages information types that are particularly relevant for health care and life science industries, such as diagnoses, drugs, substances, chemicals… With our life science knowledge we combine these information types to create relevant concepts e.g. adverse events from our “competitor drugs”.
No AI or machine learning knowledge is required. Our text classification can be easily integrated into your own workflows.
Christian Gaege, Head of Platform, Averbis
No AI or machine learning knowledge is required to use Averbis text classification – all you need is to provide sample data for the respective target categories into which new documents will be classified. Existing categorised documents can be used as training data for automatic text classification.
Averbis Text Classification API
Averbis text classification can be easily integrated into your own workflows, and run on-premise or in a private cloud when dealing with sensitive data. The entire workflow from creating and training a classifier to the automatic classification of new documents can be integrated and automated via REST API.
The text classification API includes:
- Classifier configuration: Single label and multi label classification
- Training classifiers by providing categories for imported documents
- Evaluation by providing performance metrics for trained classifiers
- Automatic classification of new documents. A confidence level is provided for each automatically assigned category.
Using the comprehensive API you easily integrate Averbis text classification into your applications and workflows. Save time and resources with automated text classification without a steep learning curve and with easy integration into your existing application landscape.