Text analysis tools, often known as text mining solutions, have been around for many years. But recent advances in artificial intelligence, machine learning and data analytics have led to a dramatic improvement in the ability of computer systems to extract meaning from structured and unstructured data in documents. And this has led to an increase in demand.

Today, most text analysis tools make use of AI-powered natural language processing (NLP) to interpret human language. Many also include ML capabilities, using models to improve their abilities over time. Common features of these platforms include the following:

• Topic extraction — Tagging text based on its subjects and themes.
• Entity extraction — Identifying the important nouns (including addresses, phone numbers and email addresses) in a piece of text.
• Keyword extraction — Highlighting the words used most often.
• Sentiment analysis — Classifying text as positive, negative or neutral.
• Emotion analysis — Identifying how the writer was likely feeling.
• Language detection — Identifying language the writer was using.

Some text analysis tools also have additional features beyond these core capabilities. To find the right tool for your enterprise’s needs, take a look at the list of leading text mining solutions below. Read More

Recent progress in pretraining language models on large textual corpora led to a surge of improvements for downstream NLP tasks.Whilst learning linguistic knowledge, these models may also be storing relational knowledge present in the training data, and maybe able to answer queries structured as “fill-in-the-blank” cloze statements. Language models have many advantages over structured knowledge bases: they require no schema engineering, allow practitioners to query about an open class of relations, are easy to extend to more data, and require no human supervision to train. We present an in-depth analysis of the relational knowledge already present (without fine-tuning) in a wide range of state-of-the-art pretrained language models. We find that(i) without fine-tuning, BERT contains relational knowledge competitive with traditional NLP methods that have some access to oracle knowledge, (ii) BERT also does remark-ably well on open-domain question answering against a supervised baseline, and (iii) certain types of factual knowledge are learned much more readily than others by standard language model pretraining approaches. The surprisingly strong ability of these models to re-call factual knowledge without any fine-tuning demonstrates their potential as unsupervised open-domain QA systems. The code to re-produce our analysis is available athttps://github.com/facebookresearch/LAMA. Read More

#nlp