Architecture review for Custom Anomaly AI

0

I am trying to build a architecture for custom anomaly ai on AWS for my startup. Please let me know if my way of thinking is correct or not

  1. Data Ingestion: Ingesting the data into AWS S3 in JSON format
  2. Data Preprocessing: Converting JSON format to Excel format using AWS Glue and storing it back into S3
  3. ML : Using AWS Sagemaker to learn from the Excel format created
  4. Quicksight to show the data for data visualization

is my way of thinking correct here?

  • May I ask why converting json to excel?

  • it is just my way of thinking. I am open ears to learn in case you have any other idea. But I want to implement Custom Anomaly AI detection.

1回答
0

I don't see the point of converting to Excel if you are not going to use a spreadsheet editor to open but instead using Quicksight. Any other format like CSV will allow you better support and parallelism.

profile pictureAWS
エキスパート
回答済み 10ヶ月前

ログインしていません。 ログイン 回答を投稿する。

優れた回答とは、質問に明確に答え、建設的なフィードバックを提供し、質問者の専門分野におけるスキルの向上を促すものです。

質問に答えるためのガイドライン

関連するコンテンツ