MediBot
An Chatbot based expert system to assist doctors improve screening of patients and diagnosis .
Runnable environment
Database connection
- Install the redis-py package with
pip install redis. - Install redis with:
wget http://download.redis.io/redis-stable.tar.gz
tar xvzf redis-stable.tar.gz
cd redis-stable
make
sudo cp src/redis-server /usr/local/bin/
sudo cp src/redis-cli /usr/local/bin/
- Start with
redis-server. This will start redis on localhost on port 6379.
HTML
Check out the zip file in lib/ for instructions to install HTML support.
#Date : 2 JAN 2016
Conventions
- Tabs = Spaces*2
- Filenames = abc_efg.py
- Verbose logging needed
- ownership = Funtctional level
##Softwares
- Anaconda - latest
- Sublime - defaults
- Python : 3.x / 2.7 TBD after delebrations , for now 3.4
##Folders
- Consumer = Sameer -To poll Telegram API and GET JSON dict .
- Core = Jaideep Kekre -To play with JSON dict and display tokenized contents in log / CLI / Return to user as an echo (DEBUG)
#Deliverables on 10 JAN
- Jaideep : Tokenizers working , JSON accept , LOG (PRINT TOKENS / USERNAME)
- Sameer : Create Medibot , 1_Medibot in telegram , Create listener , pass JSON to Core
##Functions affected
- Core : core() , make_token , get_user() , read_response()
- Consumer : TBD
#Deliverables on 24 JAN
- Jaideep : convert obo to json , service sample knowledge request , parse knowledge json db , make string input to core pretty
- Sameer : convert / implement non blocking queue based bot producer / consumer , create basic 4 QA db independent EXPERT_SYSTEM
##Functions affected
- Core : get_struct()
- Consumer : producer() , consumer()
- Expert_system: (Add functions here sameer)
- knowledge_store:obo_to_json() , service_knowledge_request() , parse_knowledge()