search

michael-spengler / WWI18SEAC-Semester-6

8 stars 5 forks source link

Trading Bot #1

Open michael-spengler opened 3 years ago

michael-spengler commented 3 years ago

Sentiment Analyse schon da:
https://www.youtube.com/watch?v=qCB8MZ-W1Ig&t=105s

Technische Analyse schon da:
https://www.tradingview.com/ siehe auch Strategy Analyzer mentioned by Anton

Quality: hummingbot connectoren zu exchanges - analog dazu: seactrader connectoren zu text sources (reddit, twitter....) --> klare sinnvolle interfaces

Goals: Recommendations in Telegram Gruppe + evtl. an die klopapier.exchange UX Improvement Gruppe

Connection zu Exchanges schon da:
https://hummingbot.io/ siehe dessen "Connectoren" z.B. zu binance.com --> I tested it successfully

aaronschweig commented 3 years ago

Troy, Jan, Matthias, Patrick, Michael und ich würden das gerne mal näher bearbeiten 😉

michael-spengler commented 3 years ago

Challenge: der trading markt scheint keinerlei feste regeln zu beinhalten - daher scheint die RL / DL Variante wenig aussichtsreich.

Eure Learnings: NLP Neuronale Netze DeFi

Loose Coupling

hummingbot connectoren zu exchanges seactrader connectoren zu text sources (reddit, twitter....) --> klare sinnvolle interfaces

michael-spengler commented 3 years ago

for the long term outlook after the semester... https://www.youtube.com/watch?v=c0gpgCyjTM8

michael-spengler commented 3 years ago

arbeiten aktuell an twitter sentiment analyse

michael-spengler commented 3 years ago

Daten da Modelle trainiert loading sentiment model remotely challenge - later TA da API will be provided to deliver price predictions based on SA + TA Disclaimer: This is not financial advice. Predictions are meant for educational purposes.

michael-spengler commented 3 years ago

https://github.com/michael-spengler/ml-server/commits/master