Experiment #102 Architectural Diagram 📍
The Experiment #102 researches about the possibilities of applying Cognitive Services and Deep Learning to classify and analyze texts and images from hotel customer reviews.
You can check out the Experiment Technical Sheet and the whole project here.
We tried to simulate hotel customer inputs from different sources (01) like Social Media, this hotel reviews (02) will be send to an Artificial Intelligent for its analysis.
We used Azure Notebooks with Jupyter to create, train and deploy the services we need. Ones of them are Containerized Services (03) we trained with Deep Learning like the Summarize Service (05) and the Classification Service (06). Others are Cognitive Services (04) we used directly from Microsoft like Computer Vision (07) and Text Analytics (08).
The Step By Step process of the Summarize Service creation can be found here.
The Step By Step process of the Classification Service creation can be found here.
The Step By Step process of the Cognitive Services deployment can be found here.
Using these endpoints and APIs to analyze the reviews we can get information from the text like the summary, classification, sentiment or key phrases and from the image like description and tags (09). After the analysis, the processed review is sent to a SQL Database (10).
We also created a Microsoft Flow template to automatize a part of the process. This flow starts with the SQL record creation as a trigger (11). Then, the flow looks at the sentiment field of the processed review (12). If the sentiment is Negative the flow launches a task to send an email with Outlook 365 (13) to the appropriate hotel department (14) (rooms, diner or pool).
You can check out the process of how to create a Microsoft Flow template here.