Samuel Pepys was a 17th century Englishman who kept a daily diary for 10 years which gives us a unique insight of life in Stuart England and includes eye-witness accounts of events such as the Great Fire of London and the Great Plague of London. I have started a project to analyse the text from the diary using common Python NLP libraries in order try and use data from the texts to find insights into Pepys’ thoughts and life. The […]

A not-too-technical guide to find the ideal text analytics solution for your application If you are leading a digital product development that transforms unstructured text into value for your internal / external users, yet have difficulties navigating through the Natural Language Processing (NLP) solutions or making the build-or-buy decision, this article helps you identify key decision criteria to make informed choice.

The recent explosion of Question Answering datasets and models is pushing the boundaries of QA systems and making them widely used by the general public in virtual assistants or chatbots (Rogers et al., 2021). However, it is well-known that is possible to “trick” machine learning systems. For instance, Goodfellow et al. (2014) shows that a clear picture of a panda can be modified with some invisible noise for humans that makes the ML model classify it as a    

CI: Continuous Integration CD: Continuous Deployment/Delivery

Is a universal translator the end goal?

Pharmacovigilance (PV) is the process of collecting, detecting, assessing, monitoring, and preventing Adverse Events (AEs) of pharmaceutical products that ensure product safety. PV is a critical area in which language technology can significantly add value. Due to the increasing occurrences of mentioning drug side effects on social media, the data has become a crucial source of public information for evaluating the consequence and efficiency of the medicine.

Everyday models get heavier and heavier (in terms of learnable parameters). For example, LEMON_large has 200M parameters and GPT-3 has over 175 billion parameters!

Determining sentiment in the text using neural network models has become almost a school task for beginner data scientists. However, despite the high quality of models and the ease of working with them, it is not so easy to characterize real data from reviews of restaurants, hotels, or museums. Even Google’s algorithms can’t do it properly. What are