ProxiScan

ProxiScan is a JavaScript library designed to handle proximity-based dataset matching in 2D and 3D space. The library allows you to compare multiple master datasets and test datasets based on spatial proximity, supporting both AND/OR conditions for flexible matching criteria. It also supports merging multiple master datasets based on an acceptance percentage to generate a final master dataset.

Features

🌟 Key Features:

• 2D & 3D Support: Seamlessly handle proximity-based dataset comparisons in either 2D or 3D space.
• Multiple Master Datasets: Merge multiple master datasets into one final master dataset using proximity-based matching and acceptance percentage.'
• Multiple Test Datasets Support: Match points from a master dataset across multiple test datasets.
• Acceptance Percentage: Specify how many datasets a point must appear in (as a percentage) to be included in the final master dataset.
• Flexible Matching Conditions: Choose between AND condition (points must match in all test datasets) or OR condition (points must match in any test dataset).
• Scan Radius: Define a custom proximity radius to control how close points must be to qualify as a match.
• Similarity Scoring: Computes a similarity score based on the proportion of matched points.
• Detailed Results: Return not just the similarity score, but also detailed information on matched pairs, including which datasets the points matched in.

Installation

You can include the ProxiScan library in your project by downloading it or installing it via npm (if published):

Using npm:

npm install proxiscan

Using a script tag:

<script src="./proxiscan.js"></script>

Usage

1. Matching Points in 2D Space Across Multiple Datasets (AND/OR Condition)

To compare multiple master datasets against multiple test datasets, use the calculateMultiMasterDatasetSimilarity function.

const masterDataset1 = [
    [1, 2],
    [3, 4],
    [5, 6]
];

const masterDataset2 = [
    [1.1, 2.1],
    [3, 4],
    [5, 6]
];

const masterDataset3 = [
    [1, 2],
    [3.2, 4.2],
    [7, 8]
];

const testDataset1 = [
    [1.1, 2.1],
    [3.5, 4.2],
    [7, 8]
];

const testDataset2 = [
    [1.2, 2.2],
    [3.4, 4.3],
    [6, 7]
];

const scanRadius = 0.5;
const acceptancePercentage = 75; // Points must be present in 75% of the master datasets
const useAndCondition = true; // AND condition: points must match in all test datasets

const result = ProxiScan.calculateMultiMasterDatasetSimilarity(
    [masterDataset1, masterDataset2, masterDataset3],
    [testDataset1, testDataset2],
    scanRadius,
    2, // 2D points
    useAndCondition,
    acceptancePercentage
);

console.log(result);

2. Matching Points in 3D Space

For datasets in 3D space, use the 3D version of the function by passing 3 as the dimensions argument:

const masterDataset3D_1 = [
    [1, 2, 3],
    [4, 5, 6],
    [7, 8, 9]
];

const masterDataset3D_2 = [
    [1.1, 2.1, 3.1],
    [4.2, 5.1, 6.1],
    [7, 8, 9]
];

const testDataset3D = [
    [1.2, 2.2, 3.2],
    [4.3, 5.3, 6.3],
    [10, 11, 12]
];

const scanRadius = 0.5;
const acceptancePercentage = 100; // 100% acceptance for merging master datasets
const useAndCondition = false; // OR condition: points match in any test dataset

const result3D = ProxiScan.calculateMultiMasterDatasetSimilarity(
    [masterDataset3D_1, masterDataset3D_2],
    [testDataset3D],
    scanRadius,
    3, // 3D points
    useAndCondition,
    acceptancePercentage
);

console.log(result3D);

API Reference

ProxiScan.calculateMultiMasterDatasetSimilarity(masterDatasets, testDatasets, scanRadius, dimensions = 2, useAndCondition = true, acceptancePercentage = 100)

Compares multiple master datasets against one or more test datasets, returning the number of matched points and a similarity score based on proximity.

Parameters:

• masterDatasets: An array of arrays, where each inner array represents a master dataset to be merged into a final master dataset.
• testDatasets: An array of arrays, where each inner array represents a test dataset to be compared against the final master dataset.
• scanRadius: A number specifying the radius within which two points are considered a match.
• dimensions: Either 2 for 2D points or 3 for 3D points.
• useAndCondition: Boolean flag to determine the matching condition:
  • true: Points from the master dataset must match in all test datasets (AND condition).
  • false: Points from the master dataset can match in any test dataset (OR condition).
• acceptancePercentage: A percentage value that determines how many master datasets a point must appear in to be accepted into the final master dataset.

Returns:

An object containing:

• matchedPoints: The number of points in the final master dataset that had matches based on the specified condition (AND/OR).
• totalPoints: The total number of points in the final master dataset.
• similarityScore: The ratio of matched points to the total points in the final master dataset.
• matchedPairs: An array of matched pairs, where each entry contains:
  • masterPoint: The point from the final master dataset.
  • matches: An array of objects representing the test datasets where the match occurred, including:
    • testDatasetIndex: The index of the test dataset where the match was found.
    • testPoint: The matched point from the test dataset.

Example Response:

{
  "matchedPoints": 2,
  "totalPoints": 2,
  "similarityScore": 1,
  "matchedPairs": [
    {
      "masterPoint": [1, 2],
      "matches": [
        { "testDatasetIndex": 0, "testPoint": [1.1, 2.1] },
        { "testDatasetIndex": 1, "testPoint": [1.2, 2.2] }
      ]
    },
    {
      "masterPoint": [3, 4],
      "matches": [
        { "testDatasetIndex": 0, "testPoint": [3.5, 4.2] },
        { "testDatasetIndex": 1, "testPoint": [3.4, 4.3] }
      ]
    }
  ]
}

OR Condition Example:

With useAndCondition = false, the points in the master dataset are considered a match if they match any of the test datasets.

{
  "matchedPoints": 3,
  "totalPoints": 3,
  "similarityScore": 1,
  "matchedPairs": [
    {
      "masterPoint": [1, 2],
      "matches": [
        { "testDatasetIndex": 0, "testPoint": [1.1, 2.1] }
      ]
    },
    {
      "masterPoint": [3, 4],
      "matches": [
        { "testDatasetIndex": 0, "testPoint": [3.5, 4.2] }
      ]
    },
    {
      "masterPoint": [5, 6],
      "matches": [
        { "testDatasetIndex": 1, "testPoint": [6, 7] }
      ]
    }
  ]
}

Edge Cases

• Empty Dataset: If the master dataset or any of the test datasets is empty, no matches will be found, and the similarity score will be 0.
• Dimensions Mismatch: If the datasets have different numbers of coordinates (e.g., a 3D point in a 2D dataset), the library will throw an error.
• Radius: Ensure the scan radius is a positive number; otherwise, the function will return no matches.
• Acceptance Percentage: Ensure the acceptance percentage is a valid number between 1 and 100. Points must be present in at least that percentage of the master datasets to be included in the final master dataset.

Example HTML Test Page

You can create an interactive test page to try out the ProxiScan library directly in the browser. Here's a basic example:

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>ProxiScan Test Page</title>
    <script src="./proxiscan.js"></script>
</head>
<body>
    <h1>ProxiScan Test Page</h1>
    <p>Open the browser console to see results.</p>
</body>
</html>

License

This project is licensed under the MIT License - see the LICENSE file for details.