New Delhi, November 5
Microsoft on Thursday mentioned it has added Hindi to its Textual content Analytics service to assist companies strengthen buyer help via an entire evaluation of consumer notion and suggestions in probably the most extensively spoken language in India.
Textual content Analytics is a part of the Microsoft Azure Cognitive Companies.
The performance supplied by Textual content Analytics contains sentiment evaluation, opinion mining, key phrase extraction, language detection, named entity recognition, and Personally Identifiable Data (PII) detection.
Utilizing this service, organisations can discover out what folks consider their model or subject as this allows analysing Hindi textual content for clues about constructive, impartial, or detrimental sentiment.
The Textual content Analytics service can be utilized for any textual / audio enter or suggestions together with Azure Speech-to-Textual content service, Microsoft mentioned.
“Underlining our dedication to serving to empower each enterprise to realize extra, Microsoft has added Hindi to the already sturdy set of worldwide languages supported by Textual content Analytics service,” Sundar Srinivasan — Basic Supervisor — AI & Search — Microsoft India, mentioned in an announcement.
“We’re serving to manufacturers break language boundaries and attain out to Hindi-speaking prospects to grasp the client’s sentiment about their merchandise, companies, and broaden their consumer suggestions attain.”
Microsoft’s Textual content Analytics service makes use of AI fashions to analyse content material in Hindi, utilizing Pure Language Processing (NLP) for textual content mining and textual content evaluation.
Its Sentiment Evaluation characteristic evaluates the textual content and returns confidence scores between zero and 1 for constructive, impartial, and detrimental sentiment for every doc and sentences inside a doc.
The service additionally offers sentiment labels (akin to “detrimental”, “impartial” and “constructive”) based mostly on the best confidence rating at a sentence and document-level.
“This helps manufacturers in detecting constructive and detrimental tonality in buyer opinions, social media and name centre conversations, and discussion board discussions, amongst different channels regardless of the place their information resides,” mentioned the corporate.