Microsoft Text Analytics API
It is a cloud-based service that provides advanced natural language processing over unstructured text and includes four main features: Sentiment Analysis, Key Phrase Extraction, Entities Recognition, Language Detection.
It is also possible to automatically generate text summaries by extracting the sentences that represent the most important or relevant information in the original content. This function reduces articles or documents to key sentences.
The API is part of Azure Cognitive Services, a set of machine learning and artificial intelligence algorithms.
Clients: KPMG, WiLSON ALLEN, IHC, LaLiga, TIBC, Kotak.
Use cases
- Unstructured data analysis (Wilson Allen)
Solution: the powerful AI solution that can helps legal and other professional services firms around the world find unprecedented levels of insight in previously siloed and unstructured data.
Result: the work for employees has become much easier. Data that were previously useless are now structured and will help in the further development of projects.
- Optimisation of fraud analytics (KPMG)
Solution: a system for customer risk analysis that identifies specific patterns and keywords in text that indicate compliance risks.
Result: Previously, it would have taken 12 to 14 weeks to get even limited results with the help of the local KPMG solution. Today, the company can deliver useful, insightful results within two to four hours.
Benefits
AI methods are used for natural language processing. Texts in English, German, Spanish and Japanese are supported. Content linking feature eliminates ambiguity by linking text to additional information on the Internet.
Classifier input features include N-grams, features created using lexico-grammatical analysis.
