I already have financial news that I got from financial news sites. Now I want to apply sentiment analysis to classify news as positive, neutral, or negative. I do not know what to do. I know some sentiment analysis models like VEDER and TextBlob, but I know they are not trained in corpus finance and they are not trained in my language. So does it make sense for me to use it to do sentiment analysis on my financial news after I've translated my data to English?
Asked
Active
Viewed 27 times
1
Eko Putra
- 11
- 2
-
it's like asking "how to build a social network app?" you're asking how to do something that takes multiple man-years and a ton of money to build. for instance, you need someone to label the news for you, which costs money. you probably could go unsupervised route but that would require linking news to a lot more features, like stock price movements intraday which also costs a lot of money. you need to narrow down your question. – Aksakal Jan 07 '22 at 13:12
-
No, I'm asking for my undergraduate thesis. I want to do a sentiment analysis on financial data, but in my language there is no corpus. So, what should I do? – Eko Putra Jan 08 '22 at 09:48