Stock News Sentiment Analysis Tool: The AI-Powered Edge Every Trader Needs

The best AI stock news sentiment analysis tool just arrived — and it’s free, frighteningly accurate, and powered by Claude

--

Press enter or click to view image in full size

In today’s fast-moving stock market, staying ahead of price swings requires more than just charts and fundamentals — it demands real-time emotional intelligence from the market itself.

The Stock News Sentiment Analysis Tool delivers exactly that: an AI-powered solution that instantly analyzes thousands of news articles, headlines, and financial reports to detect positive, negative, and neutral sentiment toward any stock or sector.

Whether you’re a day trader hunting alpha, a swing trader timing entries, or a long-term investor protecting your portfolio, this intelligent sentiment tool gives you a powerful edge by turning unstructured news chaos into clear, actionable trading signals before the crowd reacts.

Discover how this cutting-edge AI sentiment analysis technology is becoming the must-have advantage for serious traders.

Let me tell you something about stock trading that nobody puts on a motivational poster: the market does not care about your feelings. Not even a little bit.

You could be having the best day of your life — your team won, your kid said something adorable, the coffee was perfect — and the market will still look you dead in the eye and take 18% off your portfolio before lunch. I know this because it happened to me. On a Tuesday. With perfect coffee.

I had been trading for nearly a decade when I made a catastrophically preventable mistake. I held a significant position in a mid-cap tech company going into earnings week. I thought the sentiment was positive. My gut said “buy.” My gut, it turns out, had been spending too much time eating nachos and not enough time reading Reuters.

Three negative analyst downgrades, a supply chain article buried in the Financial Times, and one very ominous SEC filing later — and my position was down 22% in 48 hours. That’s the kind of loss that makes you sit very still and stare at the wall for a long time. Then it makes you build something better.

That experience is exactly why I want to talk to you today about stock news sentiment analysis, why it matters more than ever, and specifically why the StockPulse AI Sentiment Analysis Tool represents one of the most genuinely useful free trading aids I’ve seen in years of testing platforms, bots, and dashboards. We’re going to get into the science, the case studies, the real academic research — and I’m going to make you laugh a few times, because the stock market is objectively hilarious if you’re not currently bleeding out on a position.

What Is Stock News Sentiment Analysis — And Why Should You Care?

Stock news sentiment analysis is the process of systematically evaluating news articles, press releases, earnings reports, analyst notes, and financial media to determine whether the overall narrative around a given company is positive, negative, or neutral — and then translating that reading into an actionable trading signal.

Sounds simple. It is not simple. Reading every relevant news item about a stock, sourcing it correctly, weighting it by importance, factoring in recency, and then translating all of that into a coherent directional view — that’s hours of work per ticker. Hours most retail traders and even many professionals simply do not have.

And here’s the thing that makes it even more important: the market moves on perception before it moves on reality. An earnings report doesn’t have to be bad for a stock to drop. It just has to be worse than what the market felt was coming. Sentiment is the shadow of reality that the market actually trades.

The academic world has been confirming this for years. A landmark study published in the Journal of Finance by Tetlock (2007) showed that media pessimism about the market reliably predicts downward pressure on stock prices and elevated trading volume.

More recently, research published in the Journal of Risk and Financial Management (2024) found that news sentiment can predict short-term stock market fluctuations, with “forward-looking implied sentiment” capturing approximately 45–50% of the variation in stock returns.

That’s not a rounding error. That’s nearly half the movement you see in a stock price, potentially explained by what the news was saying before the market opened.

If you’re not reading the news — systematically, objectively, and at scale — you’re basically flying the plane with one eye closed.

The Problem With How Most Traders Currently Handle News

Here’s how most retail traders “analyse sentiment”: they open Twitter, see three bullish posts about NVDA, feel good about it, and buy. That’s not sentiment analysis. That’s vibes.

I say this with love, because I used to do it too. I’d spend forty-five minutes reading headlines, get increasingly confused by contradictory information, and then just go with whatever felt right.

The problem with “whatever feels right” is that it is vulnerable to the last thing you read. If the last article you read was bearish, you go bearish. If it was written by someone who was clearly compensated to be bullish, you go bullish. You’re not analysing sentiment — you’re absorbing it, uncritically, and calling it research.

The correct approach is what professional trading desks have always done: structured, systematic, source-weighted news analysis. But that requires either an entire research team or a seriously good AI tool. Until recently, retail traders had neither.

Introducing StockPulse: AI-Powered Stock News Sentiment Analysis

The StockPulse AI Sentiment Analysis Tool is a free, browser-based application that takes two inputs — a stock ticker and your Anthropic API key — and returns a professional-grade, structured sentiment report in about 60 seconds.

Screenshot 1: The StockPulse interface showing a live GOOG analysis — POSITIVE verdict with timestamp, powered by real-time web search

Press enter or click to view image in full size

What makes it different from just asking an AI chatbot “is the news good for Apple right now” is the structure. StockPulse doesn’t return a paragraph of mush. It returns a professional analyst report with five distinct sections: a summary of key news items, a sentiment breakdown for each, an overall sentiment score with confidence level, a short and medium-term price impact assessment, and specific trading implications including support/resistance levels, risk factors, contrarian views, and upcoming catalysts.

Let me show you exactly what that looks like in practice.

A Live Case Study: GOOG Analysis on 30 April 2026

I ran the tool on Alphabet (GOOG) on April 30, 2026, the day after Q1 2026 earnings. Here’s what it returned:

Section A: Summary of Key News Items

The tool identified seven major news items from the preceding 7–10 days:

1. Q1 2026 Earnings Beat (April 29, 2026) — Alphabet reported Q1 EPS of $5.11 versus consensus estimates of $2.67, a 91% beat, on revenue of $109.9B versus $107B expected. Source: CNBC/Multiple outlets.
2. Google Cloud Revenue Surge (April 29, 2026) — Cloud revenues grew 63% to $20.03B with a backlog nearly doubling to over $460B. Source: SEC filing/CNBC.
3. Dividend Increase (April 27, 2026) — Alphabet’s Board declared a 5% dividend increase to $0.22 per share from $0.21. Source: SEC filing/Benzinga.
4. Anthropic Investment Expansion (April 24, 2026) — Google plans to invest up to $40B in Anthropic with an initial $10B investment. Source: Bloomberg/CNBC.
5. Waymo Milestone Achievement (Q1 2026) — Waymo surpassed 500,000 fully autonomous rides per week at a $126B valuation. Source: SEC filing/CNBC.
6. Increased CapEx Guidance (April 29, 2026) — 2026 capex guidance updated to $180–190B (up from $175–185B). Source: CNBC/Motley Fool.
7. Analyst Price Target Upgrades (April 2026) — Multiple analysts upgraded, including Pivotal Research raising target to $470 and BMO to $410. Source: Multiple outlets.

That’s seven pieces of material news, properly dated and sourced, in about 60 seconds. That alone would have taken me two hours with a coffee and multiple browser tabs. And I still would have missed the Waymo SEC filing because, honestly, I would have given up after article four.

Section B: Sentiment Breakdown

Screenshot 2: The Sentiment Breakdown section — each news item receives a colour-coded sentiment badge with qualitative reasoning

Press enter or click to view image in full size

This is where the tool starts to genuinely earn its keep. Rather than just listing the news, it classifies each item:

• Q1 Earnings Beat — Strongly Positive. A 91% EPS beat demonstrates exceptional operational performance.
• Google Cloud Growth — Strongly Positive. 63% revenue growth with doubled backlog validates AI investment.
• Dividend Increase — Mildly Positive. Signals management confidence in cash generation.
• Anthropic Investment — Positive. Strategic AI positioning, though increases investment risk.
• Waymo Progress — Mildly Positive. Strong operational metrics, but minimal revenue contribution at $411M vs $450M prior year.
• CapEx Increase — Neutral to Mildly Negative. Higher AI spending raises near-term margin pressure concerns, especially with 2027 “significant increase” language.
• Analyst Upgrades — Positive. Multiple price target increases to $470+ reflect renewed Wall Street confidence.

Notice something important here: the tool isn’t just applauding everything. It’s calling the CapEx guidance mixed. That matters. A tool that says everything is “Strongly Positive” is useless — it’s the equivalent of a yes-man broker who tells you everything you want to hear right up until the margin call.

Section C: Overall Sentiment Score

Overall Sentiment: POSITIVE | Confidence Level: High | Strength of Sentiment: Strong

The tool’s justification: the earnings beat was exceptional with AI investments clearly driving revenue acceleration. The combination of strong execution and strategic positioning outweighs capex concerns.

This is a clean, decisive read. High confidence, strong signal. No hand-wringing, no hedge-everything-into-nothing output that leaves you no wiser than before you asked.

Section D: Potential Stock Price Impact

Screenshot 3: Price impact section — short-term Bullish (stock already up 7.6%), medium-term Bullish, with specific key levels

Press enter or click to view image in full size

• Short-term (1–7 days): Bullish — The stock already surged 7.6% post-earnings. Earnings momentum typically sustains for several trading sessions, and options activity suggests continued upward pressure.
• Medium-term (8–30 days): Bullish — Strong analyst support with average price targets of $387+ and median of $397.50. AI narrative acceleration provides sustained institutional buying interest.

Key Catalysts: Google I/O conference (May 19, 2026), Anthropic revenue integration visibility, cloud backlog conversion metrics, TPU hardware delivery milestones.

Section E: Trading Implications

• Recommended Bias: Bullish
• Support: $350 (recent earnings base)
• Resistance: $380–390 (analyst target concentration)
• Breakout level: $400+ (highest analyst targets)

Risk Factors:

• Iran war impact on AI infrastructure costs and broader tech sector concerns
• CapEx spending acceleration without proportional margin expansion
• Competitive AI model pressure from OpenAI and others

Contrarian Views:

• YouTube advertising missed expectations at $9.88B vs $9.99B estimated
• Free cash flow margin declined to 9.2% from 21% prior year
• Execution risk on massive 2027 capex increases

Upcoming Events:

• Google I/O (May 19) — AI product announcements
• Q2 earnings (July) — first full quarter of Anthropic integration
• Waymo commercial launch with Lyft (later 2026)

The Verdict

Screenshot 4: The single-line final verdict — POSITIVE | Confidence: High | Trading Bias: Bullish

Press enter or click to view image in full size

VERDICT: POSITIVE | Confidence: High | Trading Bias: Bullish

Clean. Decisive. Actionable. One line. That’s what fifteen years of trading experience looks like when it’s been codified into an AI prompt.

The Science Behind Why This Works

Now here’s where I put my serious face on, because the academic literature on this topic is genuinely fascinating — and it’s what convinced me that AI-powered sentiment analysis isn’t just a shiny toy.

Large Language Models and Financial Prediction Accuracy

A study by researchers analysing 965,375 U.S. financial news articles from 2010 to 2023 found that GPT-based large language models predicted stock market returns with an accuracy of 74.4%.

More strikingly, a long-short strategy based on these predictions, accounting for transaction costs, yielded a Sharpe ratio of 3.05 — and produced a 355% gain over two years from August 2021 to July 2023. To put that in context: a Sharpe ratio above 1.0 is generally considered good. Above 2.0 is excellent. 3.05 is the kind of number that makes quantitative hedge funds extremely interested.

The same paper notes that LLM-based sentiment analysis “significantly outperforms” traditional dictionary-based approaches (like the Loughran-McDonald financial word dictionary), underlining why modern AI tools like StockPulse — built on Claude, one of the most capable language models available — represent a meaningful step forward from older sentiment scoring methods.

Sentiment Predicts Half the Movement

Research published in the Journal of Risk and Financial Management (2024) examined nearly 1.86 million news headlines and found that forward-looking implied sentiment accounts for approximately 45–50% of the variation in stock returns.

The study also found a structural market bias toward bullish states — meaning the market tends to price in optimism, which makes it more important, not less, to identify when the news cycle is genuinely negative before the broad market notices.

Chari et al. (2023), also cited in the same paper, demonstrated that aggregate news sentiment can predict short-term fluctuations in major index returns, with the effect particularly pronounced during periods of economic or political uncertainty — exactly the kind of environment we’ve been operating in throughout 2025 and 2026.

The Hybrid Model Advantage

A January 2026 study published in the Journal of Modelling in Management evaluated AI-driven sentiment analysis using a hybrid LSTM–Random Forest framework on data from 2019 to 2024, drawing on sources including Bloomberg, Reuters, Reddit, and Twitter. The model achieved 68.5% directional accuracy and a 22% reduction in prediction error compared to traditional ARIMA benchmarks — specifically in technology and finance sectors, which are precisely the sectors most heavily covered by tools like StockPulse.

The study also noted something relevant to how StockPulse is designed: sectors like healthcare and energy showed “minimal sensitivity to sentiment,” underscoring the importance of domain-specific analysis. StockPulse’s prompt is calibrated to weight news by magnitude and type — an FDA approval matters differently than a minor analyst note, and the tool is explicitly instructed to treat them differently.

Web Search Enables Real-Time Edge

A 2025 UCLA study integrating real-time sentiment from financial news with technical indicators for S&P 500 trading found that combining sentiment-driven insights with traditional models “improves trading performance” and “offers a more dynamic approach that adapts to market changes in volatile environments.”

The key word here is real-time. Static training data has a cutoff. StockPulse uses Claude with live web search enabled — meaning every analysis queries current news, not news from months ago. That’s not a trivial distinction. In the hours following an earnings release, new analyst notes, media reactions, and institutional commentary emerge rapidly. A tool that’s searching in real time catches those signals. A static model doesn’t.

Case Study 2: How Sentiment Analysis Would Have Saved My Portfolio in 2023

Let me tell you about the time I didn’t have a tool like this. It cost me significantly, and the story is instructive.

In March 2023, I held a position in a regional U.S. bank. This was during the Silicon Valley Bank and Signature Bank collapses. Now, I was following the news — but not systematically. I was reading headlines, not sources. I was getting summary vibes, not structured sentiment.

If I had run a tool like StockPulse on my position during that period, here is what a structured analysis would have flagged:

1. Strongly Negative — Federal Reserve rate environment creating unrealised losses in held-to-maturity bond portfolios across regional banks (WSJ, March 8, 2023)
2. Strongly Negative — SVB Financial Group announced a $1.75B capital raise after a $1.8B after-tax loss on bond sales (Bloomberg, March 8, 2023)
3. Negative — Bank depositor confidence declining rapidly as social media accelerated bank run dynamics (Financial Times, March 10, 2023)
4. Strongly Negative — FDIC takeover of SVB, triggering systemic contagion fears across regional banking sector (Reuters, March 10, 2023)
5. Negative — Analyst downgrades and sector-wide re-rating of regional bank risk (multiple outlets)

Any tool delivering that Sentiment Breakdown would have produced a crystal-clear: VERDICT: NEGATIVE | Confidence: High | Trading Bias: Bearish.

Instead, I held. Because I was reading the news wrong. I was letting the noise crowd out the signal.

I closed the position five days later at a loss that I will charitably describe as “motivational.” Motivation to never again go into a high-risk period without structured sentiment analysis.

Case Study 3: NVIDIA — Riding the AI Wave With Sentiment Clarity

Now let me give you the opposite case — a situation where having clear sentiment data would have helped a trader hold rather than sell too early.

Throughout 2023 and into 2024, NVDA was one of the most news-rich stocks in the market. Every week brought a new wave of coverage: data centre expansion, new GPU announcements, enterprise AI adoption, supply constraints, geopolitical chip restrictions, analyst upgrades. For a retail trader trying to manage a position, the volume of information was overwhelming. Many traders I know sold NVDA positions early in that run because individual negative news items — export restrictions, one slightly disappointing margin figure — spooked them into exiting what was, in aggregate, a massively positive news environment.

A structured sentiment tool running on that news flow would have consistently returned: Strongly Positive on AI adoption, Positive on enterprise demand, Mildly Negative on export restrictions (real risk, but contained), Positive on analyst upgrades — and an overall POSITIVE | High Confidence | Bullish verdict, week after week.

Research from arXiv (2023) on daily news sentiment and stock price forecasting confirms this pattern: sentiment scores derived from financial news articles provide meaningful predictive lift for short-term price movements, particularly when the sentiment is consistently directional rather than mixed. NVDA in 2023–2024 was, in sentiment terms, almost relentlessly bullish — and the traders who stayed with the data rather than reacting to individual scary headlines made extraordinary returns.

How to Use the StockPulse Tool: A Step-by-Step Guide

Using the tool is genuinely straightforward. Here’s how to get your first analysis in under two minutes:

Step 1: Get your Anthropic API Key Visit console.anthropic.com, create a free account, and generate an API key. It will look like sk-ant-.... The key is used to query Claude directly — it's sent only to Anthropic's API and never stored by the tool.

Step 2: Navigate to the Tool Go to https://marketinvestigation.com/wp-content/uploads/stocknewssentimentanalysis.html

Step 3: Enter Your Ticker Type the stock ticker in the TARGET STOCK field. The field auto-capitalises. Examples: AAPL, TSLA, NVDA, MSFT, AMZN, GOOG, META.

Step 4: Paste Your API Key Enter your Anthropic API key in the API KEY field.

Step 5: Click “Analyse Sentiment” The tool will query Claude with web search enabled, retrieve recent news from high-quality sources, run the structured analysis, and return your report. This typically takes 30–90 seconds depending on how much recent news exists for the stock.

Step 6: Read the Report Work through sections A–E. Pay particular attention to:

• The Sentiment Breakdown (Section B) — individual items can reveal risks the headline number hides
• The Risk Factors and Contrarian Views in Section E — these are what most bullish sentiment tools deliberately omit
• The Upcoming Events — these are often the most actionable

Step 7: Verify and Act This is not a trading signal service and does not constitute financial advice. Use it as one input in your overall analysis framework. Cross-reference key claims. But use it — because a structured second opinion costs you 60 seconds and could save you a position.

What Makes This Tool Different From Other Sentiment Platforms

There are paid sentiment platforms out there. Some are quite good. But they share common limitations that StockPulse addresses differently:

1. Source Breadth vs. Source Depth Many sentiment platforms aggregate social media mentions and apply basic NLP scoring. They count the volume of positive vs. negative words. StockPulse’s underlying prompt explicitly prioritises quality: Reuters, Bloomberg, CNBC, WSJ, FT, SEC filings, and company press releases are weighted more heavily than forums, blogs, or social chatter. This matters enormously. A thousand Reddit posts about a stock do not carry the same predictive weight as one WSJ investigation or a single SEC filing.

2. Reasoning, Not Just Scoring Most automated sentiment tools give you a number between -1 and 1. That’s it. You get a score with no explanation of what drove it, which news items were most significant, or why the model is uncertain. StockPulse returns a full qualitative explanation for each rating. You can see why the Anthropic investment is rated Positive rather than Strongly Positive, and you can disagree with that reasoning if you have additional context the tool doesn’t.

3. Real-Time Web Access Because the tool uses Claude with live web search, it is not limited by a training data cutoff. When you query GOOG the day after earnings, it reads today’s news. This is a significant advantage over static LLM-based tools.

4. Cost Zero. Beyond the minimal API costs of running a Claude query (a few cents per analysis), the tool itself is free. Compare that to institutional sentiment platforms that can cost thousands per month.

The Limitations — Because I’m Not Going to Sell You Nonsense

Any tool worth using has limitations, and StockPulse is no exception. Let me be direct about what this tool is not:

It is not a crystal ball. Sentiment analysis improves your odds. It does not guarantee outcomes. The stock market can and will surprise you regardless of what every news article says. Black swan events — sudden geopolitical shocks, unexpected regulatory actions, executive scandals — can override any sentiment signal within minutes.

It relies on available news. For smaller-cap or more obscure stocks, there may be limited high-quality news coverage. The tool is designed to flag this with a “Low information” warning when news is sparse, but you should always sanity-check coverage quality for less-covered names.

It is not financial advice. I am a trader writing an article. StockPulse is a tool that analyses publicly available information. Neither of us is your financial adviser. Use this as one input. Not the only input.

It reflects the news as of the moment of query. Run it again tomorrow and the report may be different. That’s a feature, not a bug — markets change, and so should your sentiment read.

The Broader Landscape: AI in Trading Is No Longer Optional

Let me step back for a moment and make a broader point about where we are in 2026. Five years ago, AI trading tools were primarily the domain of institutional quantitative hedge funds with research budgets larger than the GDP of some small nations. Today, the tools are available to any retail trader with a browser and an API key.

Research published in Big Data and Cognitive Computing (2024) evaluated three AI models — FinBERT, GPT-4, and Logistic Regression — for stock sentiment analysis and prediction. The study found accuracy rates ranging from 54% to 82% depending on approach and data quality. Even at the lower end of that range, 54% directional accuracy consistently outperforms random chance — and in trading, consistent edge over time is everything.

The democratisation of AI means that retail traders who refuse to engage with these tools are now genuinely at a disadvantage — not just relative to institutional players, but relative to other retail traders who are using them. The information asymmetry is no longer “professionals vs amateurs.” It’s becoming “AI-augmented traders vs unaugmented traders.”

StockPulse represents an accessible entry point into that augmented tier. It doesn’t require coding knowledge, quantitative finance expertise, or a large capital commitment. It requires a stock ticker and 60 seconds.

Frequently Asked Questions

Is this tool really free? The tool itself has no subscription fee. You do need an Anthropic API key, and Claude API calls have a small per-query cost — typically a few cents for a full StockPulse analysis. For context, if you run three analyses per day, your monthly API cost would likely be under $5.

How current is the news analysis? Because Claude uses live web search, the analysis reflects news available at the moment of the query. For most major stocks, this means coverage updated within hours.

Can I use it for international stocks? Yes — enter the ticker symbol for any stock that receives English-language financial news coverage. Major U.S., UK, European, and Asian stocks with significant global media coverage will produce strong results. Very small-cap or locally-listed stocks in non-English markets may have limited coverage.

What if the verdict is NEUTRAL? Neutral isn’t a non-answer. A neutral reading on a stock you’re considering entering is important information — it tells you the news environment doesn’t currently provide a tailwind. Combined with your technical analysis, a neutral sentiment reading might support a wait-and-see approach, which is often the most profitable trade you’ll ever make.

Is the API key stored? No. The tool is a single HTML file. Your API key is passed directly to Anthropic’s API in the browser and is not logged, stored, or transmitted to any other server.

Final Thoughts: The Trader Who Reads the Room Wins

Here’s what fifteen years in the markets has taught me that I didn’t learn in any textbook: the traders who consistently outperform are not necessarily smarter, faster, or better connected. They are more systematic. They don’t rely on gut feelings. They don’t trade on vibes. They have a process, they execute the process, and they let the probabilities play out over time.

Sentiment analysis is part of that process. Not all of it — but a meaningful, academically validated, reproducible part of it. When the news is consistently, verifiably, multi-source positive about a company you own or are considering owning, that matters. When it’s consistently negative — even if each individual item seems minor — that matters more.

StockPulse gives you a structured, AI-powered way to run that assessment in under two minutes, for free, on any major stock. It won’t make every trade a winner. Nothing will. But it will make sure you never again hold a position through a clearly deteriorating news environment because you were too busy to read everything properly.

I’ve been that person. On a Tuesday. With perfect coffee.

Don’t be that person.

Try the tool at https://marketinvestigation.com/wp-content/uploads/stocknewssentimentanalysis.html and let the AI read the room for you.

References

1. Tetlock, P.C. (2007). Giving Content to Investor Sentiment: The Role of Media in the Stock Market. The Journal of Finance, 62(3), 1139–1168. https://doi.org/10.1111/j.1540-6261.2007.01232.x
2. Siala, W., Khanfir, A., & Papadakis, M. (2025). Impact of LLMs News Sentiment Analysis on Stock Price Movement Prediction. University of Luxembourg / University of Manouba. arXiv:2602.00086. https://arxiv.org/pdf/2602.00086
3. Liu, H., Lin, Z., & Rojas, R.R. (2025). Enhancing Trading Performance Through Sentiment Analysis with Large Language Models: Evidence from the S&P 500. UCLA Department of Statistics and Data Science. arXiv:2507.09739. https://arxiv.org/pdf/2507.09739
4. Shobayo, O., Adeyemi-Longe, S., Popoola, O., & Ogunleye, B. (2024). Innovative Sentiment Analysis and Prediction of Stock Price Using FinBERT, GPT-4 and Logistic Regression: A Data-Driven Approach. Big Data and Cognitive Computing, 8, 143. https://doi.org/10.3390/bdcc8110143
5. Emerald Publishing. (2026). AI-driven sentiment analysis in financial markets: using transformer base models and social media signals for stock market predictions. Journal of Modelling in Management. https://www.emerald.com/jm2/article/doi/10.1108/JM2-08-2025-0415/1336098/AI-driven-sentiment-analysis-in-financial-markets
6. MDPI. (2024). News Sentiment and Stock Market Dynamics: A Machine Learning Investigation. Journal of Risk and Financial Management, 18(8), 412. https://www.mdpi.com/1911-8074/18/8/412
7. Kirtac, K., & Germano, G. (2024). Sentiment trading with large language models. Finance Research Letters. ScienceDirect. https://www.sciencedirect.com/science/article/pii/S1544612324002575
8. Lis, S. (2024). Investor Sentiment in Asset Pricing Models: A Review of Empirical Evidence. arXiv:2411.13180. https://arxiv.org/abs/2411.13180
9. Todd, A., Bowden, J., & Moshfeghi, Y. (2024). Text-based sentiment analysis in finance: Synthesising the existing literature and exploring future directions. Intelligent Systems in Accounting, Finance and Management, 31(1). https://doi.org/10.1002/isaf.1556
10. Effects of Daily News Sentiment on Stock Price Forecasting (2023). arXiv:2308.08549. https://arxiv.org/pdf/2308.08549

Disclaimer: This article is for educational and informational purposes only. Nothing in this article constitutes financial advice. Trading stocks involves significant risk of loss. Always consult a qualified financial adviser before making investment decisions.