A multi-objective portfolio optimization model incorporating sentiment analysis of quarterly reports and LSTM-based price prediction
Esmaeil Taheripour, Seyed Jafar Sadjadi, Babak Amiri · 2026
Read A multi-objective portfolio optimization model incorporating sentiment analysis of quarterly reports and LSTM-based price prediction by Esmaeil Taheripour, Seyed Jafar Sadjadi, Babak Amiri (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0335036.txt.
Description
Taheripour, Sadjadi, and Amiri built a stock portfolio method that reads company reports in a 2026 study. Quarterly reports carry mood about a firm's health that prices may not yet reflect. Their three-stage method scores report mood with a finance-tuned language model, forecasts prices with a memory network, then builds a portfolio balancing several goals at once.
Mood scores feed expected return estimates alongside forecast-based risk measures. No earlier work had joined report sentiment to portfolio building in this way, the authors note. The approach suits investors drowning in filings: machines read the tone, then optimizers weigh return, risk, and mood together.
Catalog record
| CREATOR | Esmaeil Taheripour, Seyed Jafar Sadjadi, Babak Amiri |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 107271 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |