AI-Powered Earnings Sentiment Analysis & Investment Research Platform

Linking Earnings Call Sentiment to Market Performance

NLP GPT-4 Python Investment Research

Executive Summary

Earnings calls carry information that doesn't show up neatly in the numbers — tone, hedging, confidence, and how executives handle tough questions all shape how the market reacts afterward. This project builds an NLP pipeline that converts unstructured earnings call transcripts into a structured sentiment signal, using GPT-4 alongside custom ML processing, and studies how that signal relates to subsequent stock performance. A conversational query interface sits on top, letting a user ask questions and pull insights across many filings at once instead of reading each transcript individually.

Core Contribution

Combined a large language model with a custom scoring pipeline to turn unstructured transcript text into a consistent, comparable sentiment signal — then paired it with a retrieval-based conversational interface so a user can query insights across many filings at once, instead of reading each transcript individually.

Core Model

GPT-4

Sentiment & tone extraction

Data Source

Earnings Calls

Transcript-level analysis

Interface

Conversational

Query insights across filings

What It Does

From Transcript to Signal

The pipeline processes raw earnings call transcripts and extracts a sentiment/tone score for each one, capturing language cues that go beyond the reported financial figures. Those scores are then structured so they can be compared across companies, sectors, and time — and lined up against subsequent stock price movement to study the relationship between what executives say and how the market responds.

Ask Questions Across Filings

Rather than searching transcript by transcript, the conversational interface lets a user ask a question in plain language — about a company, a sector, or a theme — and get an answer synthesized from across the relevant filings, cutting out the manual read-through that investment research usually requires.

Methodology Overview

The platform was built as an end-to-end pipeline, from raw transcript to queryable insight:

  1. Data Collection — earnings call transcripts and corresponding stock price history
  2. Sentiment Extraction — GPT-4 processes each transcript to score tone, confidence, and language cues
  3. Feature Construction — sentiment scores structured for comparison across companies and time periods
  4. Performance Linkage — sentiment features analyzed alongside subsequent stock price movement
  5. Conversational Layer — retrieval-based Q&A interface for querying insights across filings

Pipeline Components

  • Ingestion — transcript collection and cleaning
  • LLM Scoring — GPT-4 used to extract structured sentiment/tone features from unstructured text
  • Retrieval Layer — filings indexed so the conversational interface can pull relevant context across multiple transcripts to answer a query
  • Analysis — Python-based comparison of sentiment features against subsequent price action

Business Applications

The combination of a sentiment signal and a cross-filing query interface has a few natural uses in an investment research workflow:

Faster Research

Ask questions across many filings at once instead of reading each transcript in full.

Sentiment Screening

Surface companies or sectors with notably positive or negative call tone for further review.

Signal, Not Just Numbers

Add a language-based signal alongside traditional financial metrics for a fuller picture.

Strategic Insight

How something is said in an earnings call carries information the reported numbers alone don't — a systematic way to capture that tone at scale is a useful complement to traditional financial analysis.

Conclusion & Future Work

This project shows how a large language model can turn unstructured, high-volume text like earnings calls into a structured research asset, paired with a natural-language interface for exploring it.