StockWatch

StockWatch is an open source stock market analysis software that uses Elasticsearch to store Twitter data and Tesla's stock price data. StockWatch performs sentiment analysis on the text to determine the general sentiment picture and displays the results in Kibana.

Download / Clone

git clone https://github.com/GigliOneiric/StockWatch.git
cd StockWatch

Install - Docker

1. Download / Clone StockWatch repo
2. Set up logstash, elasticsearch and kibana containers using Docker compose

cd StockWatch/Docker
docker-compose build && docker-compose up

Requirements

pip install -r requirements.txt

Setup

API-Keys in Config-Folder

1. Rename twitter_api_keys (Demo).py to twitter_api_keys.py
2. Create a twitter application at https://developer.twitter.com/en/portal/dashboard
3. Insert Twitter API Keys to twitter_api_keys.py

consumer_key = "yourKEY"
consumer_secret = "yourKEY"
access_token = "yourKEY"
access_token_secret = "yourKEY"
bearer_token = "yourKEY"

Change Twitter query in Twitter-Folder

• See: https://developer.twitter.com/en/docs/twitter-api/tweets/filtered-stream/integrate/build-a-rule
• Note: Rules are stored until they are deleted

stream.add_rules(tweepy.StreamRule("(tesla AND #tesla) lang:en -is:retweet -is:reply"))

Start

• Run main.py

Kibana

Kibana can be accessed via the following link: http://localhost:5601/app/dashboards

The login details are:

Username: elastic
Password: changeme

Optional: Setup Kubeflow Pipelines (KFP) with Sentiment-KFP

Sentiment KFP is an open source analysis software used to perform sentiment analysis using Kubeflow pipelines with custom models.

Setup

Follow the example at

https://github.com/GigliOneiric/Sentiment-KFP

KFP-Configuration in Config-Folder

1. Rename kfp (Demo).py to kfp.py
2. Insert data to kfp.py

username = "yourUsername"
password = "YourPassword"
host = "YourHost"

kfp_analyzer = 'true'