BERT relevance score API (beta)

Price: 1 credits per request

The BERT relevance score API calculates relevance between your keyword / phrase input and text using contextual word embeddings and bidirectional attention. Instead of just matching keywords, BERT understands the meaning and relationships between words by analyzing the entire sentence before and after the keyword. It then computes semantic similarity between the keyword and the surrounding text, helping Google assess how well the content aligns with search intent.

Query string parameters

Name Parameter Description
Keyword &keyword=

The keyword or phrase is used by the the algorithm to analyze content relevance

example: Core web vitals
Key &key= Your private API key

Response elements

Name Description
keyword Your input keyword
content Your input content
BERT score The relevance score calculated by the BERT algorithm. The closer the score is to 1.0 the higher the relevance, if this number is below 0.5 your content probably isn’t really relevant to your keyword. 
label The label, functions as a quick indicator to show you how to interpret the BERT score. 

Labels

BERT score Interpretation
1.0 Perfect match
0.8 – 0.99 Very strong relevance
0.6 – 0.79 Good relevance
0.4 – 0.59 Somewhat related
0.2 – 0.39 Weak relevance (might share a few common words but different meaning)
0.0 – 0.19 Very weak or no similarity
< 0.0 Opposite meaning

PHP code example

<?php     

/*
    API documentation for SEO Review Tools.
    API info: https://www.seoreviewtools.com/seo-api/
    API documentation: https://api.seoreviewtools.com/documentation/
    LinkedIn: https://www.linkedin.com/company/seo-review-tools
*/ 

function curlFunction ($toolRequestUrl, $data){

    $ch = curl_init();
    curl_setopt($ch, CURLOPT_URL, $toolRequestUrl);
    curl_setopt($ch, CURLOPT_HEADER, 0);
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
    curl_setopt($ch, CURLOPT_POST, 1);
    curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false); 
    curl_setopt($ch, CURLOPT_SSL_VERIFYHOST,false); 
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
    curl_setopt($ch,CURLOPT_POSTFIELDS, $data);
    $html = curl_exec($ch);
    $curlInfo = curl_getinfo($ch);

    return($html);
}
        

$data = "Web Vitals is a Google initiative that defines key quality signals essential for a great user experience on the web. While Google has offered various performance measurement tools over the years, some developers have mastered them, while others find the vast array of tools and metrics overwhelming."; 

$data = stripcslashes(trim($data));

// Keyword to check
$keywordInput = "Core Web Vitals"; 

// Encode data to JSON
$data = json_encode($data);

// API key 
$apiKey = "YOUR API KEY"; 


$toolRequestUrl =  "https://api.seoreviewtools.com/bert-score/?content=1&keyword=".urlencode($keywordInput)."&key=".$apiKey;

$seoDataJson = curlFunction($toolRequestUrl, $data);

header('Content-type: application/json');
echo($seoDataJson);

                
?>  

Python code example

import requests
import json
import urllib.parse

# API documentation for SEO Review Tools:
# API info: https://www.seoreviewtools.com/seo-api/
# API documentation: https://api.seoreviewtools.com/documentation/
# LinkedIn: https://www.linkedin.com/company/seo-review-tools

# Content data to be analyzed
data_text = (
    "Web Vitals is a Google initiative that defines key quality signals "
    "essential for a great user experience on the web. While Google has offered "
    "various performance measurement tools over the years, some developers have "
    "mastered them, while others find the vast array of tools and metrics overwhelming."
)

# Clean the data (strip whitespace)
data_text = data_text.strip()

# Keyword to check
keyword_input = "Core Web Vitals"

# API key
api_key = "YOUR API KEY"

# Encode the data to JSON
json_data = json.dumps(data_text)

# Build the request URL
tool_request_url = (
    "https://api.seoreviewtools.com/bert-score/"
    "?content=1"
    "&keyword=" + urllib.parse.quote(keyword_input) +
    "&key=" + api_key
)

# Set headers to indicate JSON content
headers = {
    'Content-Type': 'application/json'
}

# Make the POST request
response = requests.post(tool_request_url, data=json_data, headers=headers, verify=False)

# Print the JSON response
print(response.text)

JSON response example






Trusted by →

clients