I find that the Sentiment Analysis in Swedish is just generally very low quality and bad. Return 0.8 for very negative text.1 vote
We need German for Image-Klassifikation!1 vote
detect more sentiments specific like : anger, happiness, sadness, fear, surprise, disgust1 vote
We need the ability to add custom entities to Text Analytics NER (Named Entity Recognition) than what is supported. For example: I want to add a skill or a quality measure to the text posted by a user.1 vote
According to the documentation at https://westus.dev.cognitive.microsoft.com/docs/services/TextAnalytics.V2.0/operations/56f30ceeeda5650db055a3c7
submitting random text like ":) :( :D" should return a result of "(Unknown)" with a score of 0, instead it returns "English" with a score of 1.
This is pretty bad.0 votes
In this experiment(https://azure.microsoft.com/en-us/services/cognitive-services/text-analytics/), it can accept text and output a JSON with sentiment, keyPhrases, detected languages and identified entities. Which is good.
But actually, there's no such single API at all. There are 4 individual APIs for sentiment, keyPhrases, detected languages and identified entities.
Can we offer an uniform API to do what exactly the experiment do?1 vote
Work with local people to provide support for local language, or exclusive partner to develop multi language text analytics API. I am looking for a hybrid approach where there is no need to translate but can detect language on real-time by having prioritisation of language used, even if the language used are mixed (max mixture of 2 languages in a sentence/dialogue)1 vote
It would be helpful if we allow user to white list the words by saying "good" or "bad" ( feedback) after key phrase extraction and feed responses back to the system to learn(ML).3 votes
Can we expect the "Custom KeyPhrase extraction" API from Azure like "Custom Vision" API in some time in near future?3 votes
Support basic level of stock market tickers. If I were to analyze, "$SPY will be up five points by end of week" the API believes it is film related. Relate "up five points" as a positive sentiment in relation to "$SPY" or any other ticker.
We have a lot of cases in Slovenia for Text Analytics.
in the following order:
Sentiment (1st), Key phrases(2nd), Entity Recognition(3rd).4 votes
Locality "here" Map coordinates if have access to location
Locality "nearby" within 0.5 miles
District "London" - Geofence for London
Affiliated address"Mum's house"
PlaceOfWork" - your workplace
Address "Barclay's Head Office"
Motorway Junction "M25 J4"
Road Vector "A256"
Road Junction "corner of 6th and 12th street"
Bridge "London Bridge"
Station "London Victoria" (Coach/Bus)2 votes
Where can I see the complete list of 120 languages as mentioned https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/overview. I can only see handful documented https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/language-support1 vote
As mentioned in the guides, Key phrase extraction works best when you give it bigger chunks of text to work on. This means that the same text results in different sets of key phrases when it is part of a larger dataset and when it is submitted as a single record.
It would be helpful, if we have the option of also including a set of key phrases (or as mentioned in other idea submissions - custom keywords or vocabulary) so that the results are better and consistent even when the API is dealing with small chunks of text.2 votes
Make "language" parameter optional or add another option as "auto" and add detected language to the response. So I won't need to make two seperate API calls one for detecting the language than for sentiment analysis.1 vote
The response includes both entities (e.g. Person, Quantity, email and etc.) and Wikipedia ID, but it is only either of them. For example, 'Donald Trump' as input will result in Wikipedia ID 'https://en.wikipedia.org/wiki/Donald_Trump'; as output, WITHOUT the entity 'person' which is supported entitiy in the API. Thus, if there any detected Wikipedia ID, could you please add entities type on the response as well? Thank you.2 votes
I got an error message from Text Analytics Sentiment
"Truncated input to first 100 tokens during analysis".
How fix it?1 vote
For information about our Italian language support, please see: https://docs.microsoft.com/en-us/azure/cognitive-services/text-analytics/language-support.
“It was great and he did help me.”
Cognitive gives 18% which to me looks incorrect.3 votes
- Don't see your idea?