Extracting Targets and Assesments for Opinion Mining

Davis, Hunter 1 Reputation point
2022-03-28T21:55:10.723+00:00

Hello,

I am using the REST API (found at the following link: https://learn.microsoft.com/en-us/azure/cognitive-services/language-service/sentiment-opinion-mining/quickstart?pivots=rest-api ) to do sentiment analysis and opinion mining for comments data. When using the v3.2 API in Power BI (with opinionMining set to true), I am able to easily extract the confidence scores, sentiment, and sentences into columns. Ultimately, I would like to also build a list of the targets and assessments and their respective sentiment within those sentences to identify common topics.

The issue I am running into is that for some sentences, the assessment and targets are pulled from the sentence and I can clearly view them, while for the majority of the sentences, the assessments and targets are blank. In these situations, I still get a sentiment label and confidence scores for the overall sentence, but when I click on the target or assessment, it is blank.

I cannot seem to find any pattern as to why some sentences work while others do not and a majority of them do not return anything. There are even cases where within an overall comment, I get the targets and assessments within one sentence, but not another. For example, the sentence "These virtual time studies are ridiculous" gives me the targets and assessments, but the sentence "Measure time correctly so we can staff properly" returns blank results for the target and assessment. Both sentences are part of the same comment (document).

Thank you

Azure AI services
Azure AI services
A group of Azure services, SDKs, and APIs designed to make apps more intelligent, engaging, and discoverable.
2,645 questions
{count} votes