NLP Text Analysis
Easily extract important topics and entities from any text or webpage with our NLP Text Analysis tool. Using popular APIs like Google NLP, TextRazor, and Dandelion, it supports around 20 languages.
Last updated
Easily extract important topics and entities from any text or webpage with our NLP Text Analysis tool. Using popular APIs like Google NLP, TextRazor, and Dandelion, it supports around 20 languages.
Last updated
Whether optimizing content for SEO, performing competitive research, uncovering hidden trends, or finding gaps in topics and entities, NLP Text Analysis provides the clarity and precision your workflow demands. Boost your efficiency and elevate your strategy with deep semantic insights at your fingertips.
Register for an account at TextRazor.
TextRazor offers a free plan with up to 500 requests per day. The free tier includes full access to all features, allowing you to extract entities and topics effectively.
If you need more than 500 requests per day, please consider a paid plan at TextRazor Plans.
After registering an account, you can obtain your API key at your TextRazor Console.
To enter the API key to SEO Utils, please visit the Services > Natural Language Processing page.
To use Google NLP, you must first enable the "Cloud Natural Language API" in your project.
Google NLP offers a free tier providing 5,000 units per month, where each unit represents 1,000 characters processed. Additional usage beyond the free tier incurs minimal charges based on character count. For complete pricing details and calculations, visit Google NLP Pricing.
Visit Google Cloud Console and search for "Cloud Natural Language API" in the top search bar.
After enabling the API, go to the Credentials tab from the left sidebar. Click on the "Create Credentials" button and select "API Key."
If you already have an existing API key, you can use that instead.
Copy the API key from the pop-up window.
You can edit the API key to add a name and restrict access. It is recommended to restrict the key to only allow access to the Cloud Natural Language API.
To enter the API key to SEO Utils, please visit the Services > Natural Language Processing page.
Register for an account at Dandelion.
Dandelion offers a free tier allowing up to 1,000 requests per day, approximately 30,000 requests per month.
Obtain your API key by visiting your Dandelion Dashboard.
For additional requests beyond the free tier, consider upgrading to a paid plan at Dandelion Pricing.
To enter the API key to SEO Utils, please visit the Services > Natural Language Processing page.
To access the NLP Text Analysis tool, navigate to NLP > Text Analysis in the left sidebar. Hit the "Run" button to perform a new analysis.
You can enter raw text, HTML, or multiple URLs into the provided input area.
Next, select your preferred API Driver and the language for analysis from the dropdown menus.
After the analysis process is finished, SEO Utils will provide three report pages: Text Analysis, Entities, Topics.
The Text Analysis tab breaks your content down sentence by sentence and highlights all detected entities. Simply hover over an entity to view its details!
Instead of scanning through large blocks of text, you get a clear, digestible view of the key topics, terms, and important concepts in every sentence. This makes it incredibly easy to refine your content, optimize for SEO, and ensure that your messaging is precise and impactful.
An entity is a meaningful word or phrase that represents a real-world concept, such as a person, place, object, event, or idea. Unlike simple keywords, entities have deeper meanings and are linked to other related concepts. This helps you understand not just what words appear in your content, but what they actually represent.
For example, if your text mentions “New York”, SEO Utils doesn’t just recognize it as a word but understands it as a city, which belongs to a larger category of locations. This deeper understanding helps you structure content better, and ensure search engines correctly interpret your topics. By detecting and categorizing entities, you can fine-tune your content, focus on the most important subjects, and make it more relevant to your audience.
Relevance Score: Measures how strongly an entity relates to your content. A high score means the entity is central to the text.
Confidence Score: Measures how confident the API Driver is that the detected entity is correct. A higher score means the detection is more reliable.
Mention Count: Tracks how many times an entity appears in your content. A high count means the entity is frequently referenced.
Types: Categorizes entities based on their meaning, such as Software, Person, or Location. This helps you quickly understand what kind of concepts appear in your content.
Key difference between Relevance Score and Confidence Score:
Definition
Measures how relevant an entity is to the input text.
Measures how confident the API Driver is that the detected entity is correct.
What it Evaluates
Contextual similarity between the entity and the source text.
Probability that the detected entity is correct, based on multiple signals.
Use Case
Helps rank entities by importance in the text.
Helps filter out incorrect or uncertain entity detections.
Supported API Driver
All API Drivers.
Using Mention Count Toggle Button
You can see how many times each entity appears in your content by clicking the toggle button on each entity row in the table.
Using Filters for Entity List
Filter by Relevance Score – Focus on the most important topics in your content.
Use Case: If you want to analyze core topics for SEO or research, filtering by high Relevance Score ensures you see only the most relevant entities. For example, if your article is about “coffee brewing”, terms like “espresso” and “French press” will have high relevance, while broader terms like “beverages” might be excluded.
Filter by Confidence Score – Remove incorrect or uncertain entity detections.
Use Case: If you are analyzing a large text and need high-accuracy results, filtering by high Confidence Score ensures that only correctly detected entities are included. For example, if “Apple” appears in a tech article, a high confidence score ensures it refers to the company rather than the fruit.
Filter by Mention Count – Identify the most frequently used entities.
Use Case: If you want to track dominant topics in your content, filtering by mentionCount helps you find terms that appear most often. This is useful for content structuring and topic clustering. For example, if “climate change” is mentioned 20 times in an article, but “global warming” only appears twice, you might focus on emphasizing “climate change” in your SEO strategy.
By adjusting these filters, you can fine-tune your entity analysis, whether you need precise topic identification (high relevance), accurate entity detection (high confidence), or frequently mentioned topics (high mention count). 🚀
Only TextRazor supports topic extraction, so you can only see the Topics tab when using that API driver.
A topic is a broad subject or theme that your content discusses. Topics provide a high-level understanding of what your text is about, grouping related concepts under general categories like Business, Science, or Technology. Unlike entities, which represent specific names or objects, topics capture the overall theme of your content, helping you see its main focus at a glance.
Score: Measures how relevant a topic is to the input text. The higher the score, the more relevant the topic is. Score ranges from 0 to 1.
Coarse Topics: General categories.
Key Differences Between Topics and Entities
Definition
General themes or subject categories
Specific real-world objects, names, or concepts
Purpose
Helps categorize content and understand its broad focus
Identifies key terms and their semantic meaning
Examples
Business, Science, Mathematics
New York, Tesla, Python (programming language)
Granularity
High-level classification
Detailed and specific references
While entities tell you exactly what appears in the text, topics help you see the bigger picture, making it easier to categorize, summarize, and optimize your content for relevance and clarity.
Key Differences Between Topics and Coarse Topics
Granularity
More specific topics
General categories
Example
Google Search
Technology
Use Case
Detailed topic extraction
Broad text classification
Use Cases for the Topics Tab
Content Categorization & Relevance
Use Case: Ensure your content aligns with your target niche.
Example: If you’re writing about “eCommerce trends,” but the detected topics are Business and Marketing, it confirms your content is on track. However, if unrelated topics like Mathematics or Arts appear, you might need to adjust your content focus.
Topic Gap Analysis
Use Case: Compare your content topics against competitor content to find missing areas.
Example: If competitors covering “email marketing” are categorized under Digital Marketing, but your article only shows Business, you may need to add more relevant details to strengthen topic alignment.
Improving Search Intent Matching
Use Case: Make sure your content aligns with what search engines expect for a given keyword.
Example: If you’re targeting “AI in healthcare,” but your topics only show Technology, adding more medical references can help trigger Health & Medicine topics for better search relevance.
Optimizing Internal Linking & Topic Clusters
Use Case: Identify content clusters for better internal linking.
Example: If multiple articles on your site are categorized under Web Development, you can interlink them to create a stronger topical authority on that subject.
Structuring Content for Featured Snippets
Use Case: Improve content structure by ensuring clear topic coverage.
Example: If your article about “Best SEO Practices” is categorized under Business but lacks SEO or Digital Marketing, adding a structured section with on-page, off-page, and technical SEO could improve your chances of ranking in Google’s featured snippets.
By leveraging topic detection, you can ensure better content alignment, improve topical authority, and optimize your content for higher rankings in search results. 🚀
only.
