Some of the projects we have worked on in the past include:
- An unsupervised learning model to identify recurring topics and errors in the manufacturing and supply chain processes of a multinational company. The errors were written in plain English or the local language of each facility.
- A neural network based model to extract structured data and statistics from clinical trial protocols.
- A predictive model using neural networks to deduce attributes of jobseekers’ CVs, deployed on a recruitment company’s website.
- A predictive model for customers’ online purchases, for a UK retailer.
- A recommender system to recommend jobs to candidates.
- A model to predict the unloading time of vehicles, used to improve accuracy of logistics planning for grocery deliveries.
- A convolutional neural network based face recognition system, built for Android, iOS and desktop apps and used for biometric security.
- A voice controlled smart home application.