How to Detect Business Sentiment Shifts with the Pulsebit API (Python)

The Problem

If you've ever tried scraping sentiment data from various news sources, you know how cumbersome and unreliable the process can be. You spend hours writing code to fetch, parse, and analyze countless articles, only to find that the data lacks the granularity or accuracy you need. This often leads to confusion and delays in decision-making.

Imagine you're trying to detect sentiment shifts in real time, but you're stuck with outdated or irrelevant information. This is a pain point that many developers face—until now.

The Solution

Enter the Pulsebit API. With a single endpoint, you can access comprehensive sentiment analysis based on real-time news data. You don’t have to worry about scraping and parsing; Pulsebit does it all for you. Specifically, the /news_semantic endpoint is where the magic happens. It provides a structured response that can help you understand the nuances of business sentiment in a matter of seconds.

The Code

Here’s a simple Python script to get you started with the Pulsebit API. Make sure to replace YOUR_API_KEY with your actual API key.

import requests

![Left: Python GET /news_semantic call for 'business'. Right: ](https://aace88ba921016d861eaeb8858550e91.r2.cloudflarestorage.com/pulsebit-marketing/figures/g3_code_output_split_1772836782897.png)
*Left: Python GET /news_semantic call for 'business'. Right: live JSON response structure. Three lines of Python. Clean JSON. No infrastructure required. Source: Pulsebit /news_semantic.*


def fetch_business_sentiment():
    url = "https://pulsebit.lojenterprise.com/api/news_semantic"
    headers = {
        "Authorization": "Bearer YOUR_API_KEY",
    }

    response = requests.get(url, headers=headers)

    if response.status_code == 200:
        return response.json()
    else:
        raise Exception(f"Error fetching data: {response.status_code} - {response.text}")

if __name__ == "__main__":
    sentiment_data = fetch_business_sentiment()
    print(sentiment_data)

Reading the Response

When you call the /news_semantic endpoint, you'll receive a JSON response with several important fields. Here’s what they mean:

• sentiment_score: This is the current sentiment score for business, which is +0.000. A score of 0 suggests neutral sentiment, but it’s crucial to consider the context.
• momentum_24h: The momentum is +0.137, indicating a slight increase in sentiment over the past 24 hours. This is where you should focus your attention; a rising momentum can signify a shift in sentiment.
• confidence: The confidence score of 0.870 suggests that the sentiment data is fairly reliable.
• semantic_clusters: With 18 clusters, there’s a rich set of topics being discussed, which can lead to more nuanced insights.
• direction: The sentiment direction is "rising," indicating that sentiment is gaining traction.
• region: This data is global, so you’re looking at a broad spectrum of sentiment.
• semantic_similarity_avg: At 0.231, this indicates how similar the discussions are, which can help in understanding the consistency of sentiment across articles.

Three Use Cases

1. Algo Alert: You could set up an algorithm that triggers alerts when the momentum exceeds a certain threshold, say +0.2. This could help you make informed decisions quickly.

2. Slack Bot: Imagine creating a Slack bot that fetches and summarizes this sentiment data every morning. You can keep your team updated on the latest trends without them having to dig through data themselves.

3. Dashboard: Integrate the sentiment data into a dashboard that visualizes sentiment shifts over time. This could be particularly useful for presentations or strategy sessions.

Get Started

If you want to dive deeper and explore more about the Pulsebit API, check out the official documentation at pulsebit.lojenterprise.com/docs. This resource will help you understand the full capabilities of the API and how to leverage it effectively in your projects.

By using the Pulsebit API, you can finally remove the headaches associated with DIY scraping and focus on what really matters: interpreting and acting on sentiment shifts that could impact your business decisions.

Arabic coverage led by 4.2 hours. English at T+4.2h. Confidence scores: Arabic 0.82, Mandarin 0.68, English 0.41 Source: Pulsebit /sentiment_by_lang.