: Is Grammarly's "Engaging" metric trustworthy? Grammarly has an "engaging" metric on its web page when you submit a document for autocorrection. Is this metric trustworthy? What is it based upon?
Grammarly has an "engaging" metric on its web page when you submit a document for autocorrection.
Is this metric trustworthy?
What is it based upon? Is there an objective theory behind it or is it just something they whipped up using some obscure algorithm?
Just to be clear, I'm asking if that metric has basis on any known study or theory.
More posts by @Carla500
: Preposition at the end of sentence Is it wrong to have a preposition at the end of sentence? In the context of writing a statement of purpose, I cannot of any substitute for the following
: Describing shapes and objects While I absolutely agree, that descriptions should be experienced through the narrator's / protagonist's eyes in terms of personal involvement, I oftentimes find myself
1 Comments
Sorted by latest first Latest Oldest Best
No, the metric is not what one word ('engaging') tries to describe what it potentially is. If a reader is least interested, no metric can predict engagement. If an administrator's job is to configure the server, they will be super-engaged reading dull documentation! I can only think of casual communication (email, forum, StackExchange, etc.) where this particular metric will help. And yes, in many such instances it'll make the reading more engaging. In layman's term, I'd say Grammarly's metric will help not put readers to sleep.
On the official website they claim to add "vivid words" and to vary "the rhythm". The examples provided there makes sense from casual writing point of view. However, if someone is writing fiction, they'd have their own flair and style of writing. No rules or metric required there. On the other hand, in technical writing where superfluous words are anyway not used and there's a style guide to be adhered to, there's no room for vivid words and the metric doesn't apply.
To answer the other half, the metric is based on AI/ML algorithms. The content is processed probably using NLP techniques that try to give score to abstract things like emotion, sentiment, engagement, etc. related to some text. While the techniques are super helpful and are well-established, it is still man vs machine. The AI/ML algorithms have much to learn. Use the advice where it is helpful. Ignore the rest. Demand to train the algos better. The other comment mentioned manned aircraft example, that's concrete reason to ignore and move on to a human editor.
Terms of Use Privacy policy Contact About Cancellation policy © selfpublishingguru.com2024 All Rights reserved.