Smart Reach for Square Demo

Finance

Customer Personalized Marketing AI Model

Paper

Smart Reach is an AI-powered marketing solution that utilizes the Square Customer API and the Kenoobi Decision Engine. It enables Square merchants to analyze customer behaviors and attributes, create targeted marketing messages, and automate their delivery. By integrating with the Customers API, merchants can easily manage customer profiles, search for customers based on specific criteria, and gain valuable insights for their marketing campaigns. The Kenoobi Decision Engine powers the AI algorithms, allowing merchants to make data-driven decisions and optimize their marketing strategies.

Your IP: 216.73.216.210
Test Neurons: 20

Workflows (Fuel)

Operations that generate data

These workflows demonstrate how the Smart Reach can leverage various data points and machine learning techniques to deliver highly targeted and personalized marketing messages.

Data(Air)

Used to generate analytics.

By providing your own data, you'll be able to witness the model in action and see how it recommends marketing suggestions. If you want to use actual data, check API section.

This attribute indicates the customer's age, which can be used to target specific age groups or personalize offers based on different age demographics. For example, different marketing messages can be created for younger customers versus older customers.

The gender attribute represents the customer's gender, which can be utilized to create gender-specific marketing messages. This allows for more targeted and relevant communication, ensuring that the message resonates with the customer.

The preferred category attribute captures the customer's preferred product category or area of interest. This information is valuable in tailoring marketing messages to match the customer's specific interests and promoting products or services that align with their preferences.

Analytics(Spark)

Used to generate decisions.

A detailed overview of the analyzed data and the decisions generated by our AI model.

API Documentation


Endpoint: https://console.kenoobi.com/restapi

Description: This API endpoint allows you to send customer attributes data to Smart Reach for analysis and marketing recommendation.



Request Headers:

Header Value
apikey KDE_API_TOKEN
squaretoken SQUARE_API_TOKEN
Content-Type application/json


Request Body:

{
   "name":"Square Test Customer",
   "custom_attributes":{
      "age":20,
      "gender":"Male",
      "preferred_categories":"Snacks"
   }
}


Request Parameters:

Parameter Description
age This attribute indicates the customer's age, which can be used to target specific age groups or personalize offers based on different age demographics. For example, different marketing messages can be created for younger customers versus older customers.
gender The gender attribute represents the customer's gender, which can be utilized to create gender-specific marketing messages. This allows for more targeted and relevant communication, ensuring that the message resonates with the customer.
preferred_categories The preferred category attribute captures the customer's preferred product category or area of interest. This information is valuable in tailoring marketing messages to match the customer's specific interests and promoting products or services that align with their preferences.


Response:

The response will provide analysis and recommended targeted marketing messages,

Example Response:

{
   "custom_attributes_analysis":[
      {
         "category":"Snacks",
         "suggestion":"Target customers between 18 - 24 years of age, who prefer Snacks"
      }
   ],
   "recommended_messages_to_send":[
      {
         "type":"text",
         "message":"Check out our new Snacks, now available for our younger customers aged 18-24!"
      }
   ]
}

Note: Replace KDE_API_TOKEN and SQUARE_API_TOKEN with your actual API token provided by the Smart Reach and Square Developer portal.

Please ensure that you include the necessary request headers, provide accurate data in the request body, and handle the response accordingly.

Code Snippet

$curl = curl_init();

                              curl_setopt_array($curl, array(
                              CURLOPT_URL => 'https://console.ai.kenoobi.com/restapi',
                              CURLOPT_RETURNTRANSFER => true,
                              CURLOPT_ENCODING => '',
                              CURLOPT_MAXREDIRS => 10,
                              CURLOPT_TIMEOUT => 0,
                              CURLOPT_FOLLOWLOCATION => true,
                              CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
                              CURLOPT_CUSTOMREQUEST => 'POST',
                              CURLOPT_POSTFIELDS =>' {
                                          "name":"Square Test Customer",
                                          "custom_attributes":{
                                             "age":20,
                                             "gender":"Male",
                                             "preferred_categories":"Snacks"
                                          }
                                       }',
                              CURLOPT_HTTPHEADER => array(
                                 'apikey: KDE_API_TOKEN',
                                 'model: fraudai',
                                 'Content-Type: application/json'
                              ),
                              ));

                              $response = curl_exec($curl);

                              curl_close($curl);
                              echo $response;
Input Output 
{
  "currency": "USD",
  "average_transaction_amount": "500",
  "average_transaction_frequency": "Weekly",
  "customer_country": "United States",
  "current_transaction_amount": "700",
  "current_transaction_frequency": "Daily",
  "customer_ip": "192.168.0.1",
  "customer_ip_country": "United States",
  "current_transaction_country": "United States"
}
 
{
  "transaction_id": "123456",
  "risk_level": "Low",
  "fraud_indicators": [],
  "recommendations": []
}
 
{
  "currency": "USD",
  "average_transaction_amount": "5000",
  "average_transaction_frequency": "Weekly",
  "customer_country": "United States",
  "current_transaction_amount": "10000",
  "current_transaction_frequency": "Weekly",
  "customer_ip": "192.168.0.1",
  "customer_ip_country": "United States",
  "current_transaction_country": "United States"
}
 
{
  "transaction_id": "1234567891",
  "risk_level": "High",
  "fraud_indicators": [
    "Unusually high transaction amount"
  ],
  "recommendations": [
    "Additional verification required for this transaction"
  ]
}
 
{
  "currency": "USD",
  "average_transaction_amount": "100",
  "average_transaction_frequency": "Weekly",
  "customer_country": "United States",
  "current_transaction_amount": "50",
  "current_transaction_frequency": "Weekly",
  "customer_ip": "192.168.0.2",
  "customer_ip_country": "United States",
  "current_transaction_country": "United States"
}
 
{
  "transaction_id": "1234567892",
  "risk_level": "High",
  "fraud_indicators": [
    "Different IP address used for the transaction"
  ],
  "recommendations": [
    "Perform additional identity verification for this transaction"
  ]
}
 
{
  "currency": "USD",
  "average_transaction_amount": "50",
  "average_transaction_frequency": "Weekly",
  "customer_country": "United States",
  "current_transaction_amount": "30",
  "current_transaction_frequency": "Weekly",
  "customer_ip": "192.168.0.1",
  "customer_ip_country": "Canada",
  "current_transaction_country": "United States"
}
 
{
  "transaction_id": "1234567893",
  "risk_level": "High",
  "fraud_indicators": [
    "Mismatch between customer's IP country and transaction country"
  ],
  "recommendations": [
    "Perform additional verification for this transaction"
  ]
}

Smart Reach Security

Introduction

The Smart Reach is an AI model framework that is widely used for analyzing large sets of data. The data used in the analysis can be sensitive, containing personal or confidential information. The sensitive data needs to be handled securely to ensure that privacy is maintained. This section explains how the Smart Reach uses various security techniques to safeguard the sensitive data during analysis.

Tokenization

The Smart Reach uses tokenization, a technique that replaces sensitive data with a unique identifier or token that is meaningless without the corresponding key. This ensures that personal data is not exposed during data analysis. Tokenization is used to protect sensitive data such as names, email addresses, and phone numbers.

Data Masking

Data masking is another technique used by Smart Reach to protect sensitive data during analysis. Data masking involves replacing sensitive data with similar-looking but fictitious data to preserve the data's format while concealing its content. This technique is useful for protecting sensitive data such as social security numbers, credit card numbers, and addresses.

Limited Retention

The Smart Reach limits the retention of data to the minimum necessary period required for the analysis. Once the data is no longer needed, it is disposed of securely to prevent any unauthorized access. Limited retention ensures that sensitive data is not stored longer than necessary, reducing the risk of data breaches and unauthorized access.

Secure Data Transfer

When transferring sensitive data, the Smart Reach ensures that the data is transmitted securely using encryption protocols such as SSL/TLS. Secure data transfer prevents unauthorized access to sensitive data during transit, reducing the risk of data breaches.

Differential Privacy

The Smart Reach uses differential privacy, a technique that adds noise to data to prevent the identification of individual records while preserving the overall data patterns. This technique can be used to anonymize sensitive data before analysis. Differential privacy ensures that sensitive data is protected while maintaining the accuracy of the analysis.

Proper Configuration

The Smart Reach ensures that the data storage and processing systems used for the analysis are properly configured to maintain data security and privacy. The systems are configured to protect against unauthorized access, data breaches, and other security risks.

Conclusion

The Smart Reach is an AI model framework that is widely used for analyzing large sets of data. The sensitive nature of the data used in the analysis requires that it is handled securely and with the utmost privacy. The Smart Reach uses various security techniques, such as tokenization, data masking, limited retention, secure data transfer, and differential privacy, to protect sensitive data during the data analysis process. Additionally, proper configuration of data storage and processing systems is ensured to maintain data security and privacy.