ronakmshah / params-di

Params Data Insight

About:

Idea behind this application is to implement self-service data intelligence.

1. Connect to your datastore.
2. Get to know all the possible questions in a natural language that your data can answer.
3. Ask the question and get back the answer.

Setup:

1. git clone this repository and go to params-di folder.
2. pip install -r requirements.txt
3. Setup flask app environment variable export FLASK_APP=params-di.py
4. Run the server: flask run [--host=]
5. Go to http://ip:5000 on web browser and start using application

Sample Question-Answer:

Question: For last 90d return average for value where product = "bat" ( Index reference: sales ) Answer: Elasticsearch response: {"hits": {"hits": [], "total": 90, "max_score": 0.0}, "_shards": {"successful": 5, "failed": 0, "total": 5}, "took": 124, "aggregations": {"string_field_agg": {"numeric": {"value": 140}, "doc_count": 90}}, "timed_out": false}

TODO:

1. Tabify the response for better visualization
2. Handling of elasticsearch's error response
3. Optimization of the questions generated.