Make sense and make use of your data

Dcipher Lab provides consulting services in analytics, data science, and machine learning. We champion a lean, lab-based approach to data mining and machine learning, using rapid experimentation as an efficient ways of finding and realizing value in data. Run small-scale experiments to establish proofs of concept before you make big investments in IT infrastructure and development.

Dcipher Lab’s data scientists and machine learning engineers help organizations visualize and understand the value in the treasure troves of internal and external information. Based on these insights, machine learning models and automated analytics workflows are developed to achieve monetization and cost savings.

Does your organizations have an internal file system containing in the range of thousands to millions of files in a variety of formats – documents, PDFs, spreadsheets, PPTs, images, and so on? Dcipher Lab can help you structure this data into a knowledge graph that can easily be browsed and queried to provide the information your teams need, when they need it. Complement this with a cognitive enterprise search solution, enabling more efficient search of structured and unstructured information even without knowing the exact keywords.

How organizations are leveraging Dcipher Lab

The Swedish Public Dental Organizations wanted to provide AI-powered decision support to their dentists to increase the consistency and quality of their diagnoses. The Dcipher Lab team used millions of historic x-ray images and logs to train a deep learning model to automatically detect caries in x-ray images. The model reached superhuman accuracy.

The image on the right shows the part of the image that the neural network responded strongly to (in white), corresponding to the location with caries, as seen in the dental x-ray on the left.


How is COVID-19 affecting consumers?

To find out, we will use the latest AI-powered technology to conduct social media mining and netnographic analysis, analyzing posts shared publicly online by people affected.