The PhenomAInon platform provides text analysis by first ingesting documents, images, videos, and audio files. Some documents we receive come in the format range of 100% digital to handwritten napkins and pieces of paper that are photographed and then saved as a PDF. We are sometimes require to pull text from images and videos as well and even capture audio whereby we provide transcription and translation (where necessary) to complete the extraction process and ensure we have clean raw data to work with. From there, we apply additional AI services including sentiment analysis to fully comprehend the meaning of the text collected. Text data is critical to add to our curated data sets to ensure we are capturing correlations in data across all incidents reported regardless of media type.
The following is a demo of one example of how we process text extraction by automating a service hosted by Amazon Web Services called Textract.
-NATURAL LANGUAGE PROCESSING (NLP)-
The following is a demo of an example of how once we have extracted text and documents across multiple data sets, we can enable for Natural Language Search (NLS) to retrieve relevant information quickly. In this example, we are showing a demonstration of Amazon Kendra.
Subscribe to see our published data sets.