We are excited to announce our newest Automatic Extraction API. The Blog Comments API is now publicly available as a BETA release.
If you want to skip the introductions and just get stuck in, here are the links you need:
Zyte Automatic Extraction Comments API sets out to bring the power of our automatic data extraction capabilities currently used for applications such as media monitoring, job postings, and more into the arena of blog comment analysis.
The underlying data model for the API was released to production as part of 20.6.0 release of AutoExtract.
Customer support management presents many challenges due to the sheer number of requests, varied topics, and diverse departments within a company that might have a say in resolving the matter.
Sourcing structured data from blog comments as provided by our API can be used in tandem with natural language understanding (NLU) solutions to quickly and effectively identify, track and act upon particular conversation strings ‘hidden’ amongst the noise of thousands of comments. You are effectively highlighting warning signs that your CX team should become involved before an incident takes place.
Another particular powerful insight that can be derived from comments revolving around the sphere of Voice of Customer (VoC) and product analysis. By tapping into blog comments, you can search keywords for a particular product or feature or use the parsed data to train sentiment analysis model to find only the information you need.
Without our Comments API (Beta), you would need to write custom code for each blog post to extract and parse the data. Let alone the time and overhead spend required to maintain the necessary infrastructure to deliver the data in real-time, in a scalable fashion and reliably.
We are continuously improving the underlying AI technology, so you can be assured you get the highest quality comment data possible.
Using the API is simple:
Here’s an example response:
[ { "comments": { "url": "https://example.com/article-with-comments", "comments": [ { "text": "A comment on article", "datePublished": "2020-01-30T00:00:00", "datePublishedRaw": "Jan 30, 2020", "upvoteCount": 12, "downvoteCount": 1, "probability": 0.95 }, { "text": "Another comment", "probability": 0.95 } ] }, "webPage": { "inLanguages": [ {"code": "en"}, {"code": "es"} ] }, "query": { "id": "1564747029122-9e02a1868d70b7a3", "domain": "example.com", "userQuery": { "pageType": "comments", "url": "https://example.com/article-with-comments" } ]
Learn more about the API in the documentation.
Our Comments data API is ideal for
Data fields that Comments API can extract for you:
Read the documentation for more information about the fields.
Here’s what you need to do if you want to get access to Comments API:
If you want to check out any of our other Zyte Automatic Extraction APIs, check them out here for free!
PS: I said it before, no better place to gain product feedback than in a blog post comment. Please share your thoughts in the comments below