Machine Learning with Amazon Comprehend for News Article Insights
Question
You work as a machine learning specialist for a news organization with a very active online community that contributes comments on your organization's news articles very frequently.
Your news editors wish to use the comments from their users to gain insight into what interests them the most.
Instead of just relying on the raw count of comments per article, the editors would like to use machine learning to find the underlying intent of the comments.
This will allow them to understand their readers better to provide more tailored articles for the most popular subjects. You have decided to use Amazon Comprehend as your machine learning platform for this task.
Which of the listed Comprehend APIs would give you the information your editors have requested? (Select THREE)
Answers
Explanations
Click on the arrows to vote for the correct answer
A. B. C. D. E. F.Answers: B, D, E.
Option A is incorrect.
The CreateDocumentClassifier Comprehend API creates a document classifier that you use to categorize documents.
Your editors want you to find the underlying intent of the comments.
Option B is correct.
The DetectSentiment Comprehend API gives you the underlying sentiment (positive, neutral, mixed, or negative) of a string, such as a comment.
Option C is incorrect.
The DetectSyntax Comprehend API gives you the part of speech of each word in a string.
This would not help you understand the underlying intent of a comment.
Option D is correct.
The DetectEntities Comprehend API finds named entities in text.
This would help you find entities such as a news organization, politicians, celebrities, companies, etc.
This information will help you identify the subject matter of the comments.
Option E is correct.
The DetectKeyPhrases Comprehend API finds key noun phrases in the text.
This will also help you identify the subject matter of a comment.
Option F is incorrect.
The DetectDominantLanguage Comprehend API finds the language (English, French, Spanish, etc.) used most frequently in the comments.
This would not offer you much insight into the intent of a comment.
Reference:
Please see the Amazon Comprehend developer guide titled Amazon Comprehend.
The task at hand is to use machine learning to understand the underlying intent of comments posted on news articles, in order to tailor articles to the most popular subjects. To achieve this goal, Amazon Comprehend, a natural language processing (NLP) service offered by Amazon Web Services (AWS), can be used.
There are several APIs offered by Amazon Comprehend, and the ones that can provide the information requested by the editors are:
DetectSentiment: This API can be used to determine the sentiment (positive, negative, or neutral) expressed in a given comment. This can help the editors understand the readers' emotional response to an article, and identify the topics that evoke the strongest reactions.
DetectEntities: This API can identify and extract entities such as people, organizations, and locations mentioned in a comment. This information can help the editors understand the specific topics or entities that are being discussed by the readers.
DetectKeyPhrases: This API can extract key phrases that summarize the main topics or ideas expressed in a comment. This can help the editors identify the most important topics discussed by the readers and tailor their articles accordingly.
The other APIs listed in the answer options are not relevant to the task at hand:
CreateDocumentClassifier: This API is used to train a custom document classifier to categorize documents into one or more categories. While this API can be useful for other tasks, it is not relevant to the current task of understanding the intent of comments.
DetectSyntax: This API can parse a given comment and provide information on the grammatical structure of the sentence. While this can be useful for other NLP tasks, it is not directly relevant to the current task of understanding the underlying intent of comments.
DetectDominantLanguage: This API can detect the language of a given comment. While this can be useful for other tasks, it is not directly relevant to the current task of understanding the underlying intent of comments.
Therefore, the three relevant Amazon Comprehend APIs for the given task are DetectSentiment, DetectEntities, and DetectKeyPhrases.
