stock recommendation engine based on all my stonk data.
fauna db is the file that stores all the ai data coming from my cron jobs (thanks github actions)
/home/codespace/.deno/bin/deno run --allow-env --allow-net --allow-read fauna.ts
- source (dli_invest)
- country (what country the company is in)
- exchange
- description
- author
- fields
- subjectivity
- senitment
- company (boolean)
- category (0-10)
- useless
- unrelated
- maybe
- useful
- excellent
- expeditious
Rename to classify engine to classify news items as they come in.
Purge documents that are not interesting to me, logic.
Typical Tasks is 5, anything exciting is rated from 6 to 10
Dividends - 3 Earnings - 5 Rest - 0 Motley Fool, Zacks, Simply Wall Street - 2 Bloomberg - 3 Extension to File Form - 4 Earnings Call Transcript - 5 initial public offer - 2 or 3 private placement - 2 made an offer, acquisiton - 5 Beat Analyst Forecasts - 3 termination, bid together rate hikes & fed 7 S&P 500 Any tickers gives 1 point automatically People like elon musk and nhtsa Supply Chain Federal Trade Commission CNW Group
Extract first few words with - number with deal.
Any numbers above 0.9 in subjectivity, or senitment
Sample graphql structure
type Field {
name: String!
value: String!
article: [Article!] @relation(name: "article")
}
type Article {
source: String!
country: String!
exchange: String!
description: String!
author: String!
url: String!
company: String!
fields: [Field!] @relation(name: "fields")
}
type Query {
allArticles: [Article!]
}
The server code has been moved to https://github.com/dli-invest/dli-fauna-gql/blob/main/fauna.ts
to ensure that the files there are easier to update.