Ask Sendai

This article details a work we did in collaboration with the French administration (DINSIC) and a French supreme court (Cour de cassation) around 2 well-known Named Entity Recognition (NER below) libraries, Spacy and Zalando Flair. Spacy accuracy was too limited for our needs, and Flair was too slow. At the end we optimized Flair up to a point where inference time has been divided by 10, making it fast enough to anonymize a large inventory of French case law.

Natural language understanding (NLU) and language translation are key to a range of important applications, including identifying and removing harmful content at scale and connecting people across different languages worldwide. Although deep learning–based methods have accelerated progress in language processing in recent years, current systems are still limited when it comes to tasks for which large volumes of labeled training data are not readily available. Recently, Facebook AI has achieved impressive breakthroughs in NLP using semi-supervised and self-supervised learning techniques, which leverage unlabeled data to improve performance beyond purely supervised systems.

It would be easy to argue that Natural Language Toolkit (NLTK) is the most full-featured tool of the ones I surveyed. It implements pretty much any component of NLP you would need, like classification, tokenization, stemming, tagging, parsing, and semantic reasoning. And there’s often more than one implementation for each, so you can choose theexact algorithm or methodology you’d like to use. It also supports many languages. However, it represents all data in the form of strings, which is fine for simple constructsbut makes it hard to use some advanced functionality.

Machine learning systems often act on “features” extracted from input data. In a natural-language-understanding system, for instance, the features might include words’ parts of speech, as assessed by an automatic syntactic parser, or whether a sentence is in the active or passive voice. Some machine learning systems could be improved if, rather than learning from extracted features, they could learn directly from the structure of the data they’re processing.

In this article, we’ll have a look at how you can implement a content-based recommendation system using Python and the scikit-learn library. But before diving straight into this, it’s important to have some prerequisite knowledge of the different ways by which recommendation systems can recommend an item to users. Content-based: A content-based recommendation finds similar items to a given item by examining the item’s properties, such as its title or description, category, or dependencies on other items (for example, electronic toys require batteries).

Zalando research brings the latest flair to the scene. (Yes, that Zalando. The German-based fashion and beauty online shop operates in fifteen different European countries.) Flair is a simple framework for state-of-the-art natural language processing. It builds on top of Pytorch – a popular deep learning platform, which makes it easy to use. A powerful NLP library. Flair allows you to apply our state-of-the-art natural language processing (NLP) models to your text, such as named entity recognition (NER), part-of-speech tagging (PoS), sense disambiguation and classification.

Last Christmas, I went through the most frustrating experience as a consumer. I was doing some last minute holiday shopping and after standing in a long line, I finally reached the blessed register only to find out that my debit card was blocked. I could sense the old lady at the register judging me with her narrowed eyes. Feeling thoroughly embarrassed, I called my bank right away. To my horror, they told me that my savings account was hacked and thousands of dollars were already gone!