Anomaly Detection and Outlier Handling in PHP Using Z-Score and Isolation Forest

Overview: In practical data processing, outliers often appear due to measurement errors, unpredictable events, or data source issues, negatively affecting analysis, model training, and prediction. This article introduces how to use PHP and machine‑learning techniques for anomaly detection and outlier handling.

1. Anomaly Detection Methods:

Various machine‑learning algorithms can be employed to detect outliers. Two common methods are described below.

1.1 Z‑Score Method:

The Z‑Score method is a statistical approach that calculates each data point’s deviation from the dataset mean to determine outliers. Steps:

1) Compute the mean and standard deviation of the dataset.

2) For each data point, calculate deviation = (data – mean) / std.

3) With a typical threshold of 3, mark points whose absolute deviation exceeds the threshold as outliers.

function zscore($data, $threshold){
    $mean = array_sum($data) / count($data);
    $std = sqrt(array_sum(array_map(function($x) use ($mean) { return pow($x - $mean, 2); }, $data)) / count($data));
    $result = [];
    foreach ($data as $value) {
        $deviation = ($value - $mean) / $std;
        if (abs($deviation) > $threshold) {
            $result[] = $value;
        }
    }
    return $result;
}

$data = [1, 2, 3, 4, 5, 100];
$threshold = 3;
$result = zscore($data, $threshold);

echo "异常值检测结果：" . implode(", ", $result);

1.2 Isolation Forest:

Isolation Forest is a tree‑based anomaly detection technique that builds random binary trees to assess the isolation depth of each point. Steps:

1) Randomly select a feature and a split point between its minimum and maximum values.

2) Partition the data recursively until each leaf contains a single point or a maximum tree depth is reached.

3) Compute the anomaly score from the path length; shorter paths indicate higher abnormality.

require_once('anomaly_detection.php');

$data = [1, 2, 3, 4, 5, 100];
$contamination = 0.1;
$forest = new IsolationForest($contamination);
$forest->fit($data);
$result = $forest->predict($data);

echo "异常值检测结果：" . implode(", ", $result);

2. Outlier Handling Methods:

After detecting outliers, they need to be processed. Two common approaches are presented.

2.1 Delete Outliers:

A simple method is to remove outliers directly based on detection results.

function removeOutliers($data, $threshold){
    $result = [];
    foreach ($data as $value) {
        if (abs($value) <= $threshold) {
            $result[] = $value;
        }
    }
    return $result;
}

$data = [1, 2, 3, 4, 5, 100];
$threshold = 3;
$result = removeOutliers($data, $threshold);

echo "异常值处理结果：" . implode(", ", $result);

2.2 Replace Outliers:

Another approach replaces outliers with a reasonable value such as the mean or median, preserving the overall distribution.

function replaceOutliers($data, $threshold, $replacement){
    $result = [];
    foreach ($data as $value) {
        if (abs($value) > $threshold) {
            $result[] = $replacement;
        } else {
            $result[] = $value;
        }
    }
    return $result;
}

$data = [1, 2, 3, 4, 5, 100];
$threshold = 3;
$replacement = 0;
$result = replaceOutliers($data, $threshold, $replacement);

echo "异常值处理结果：" . implode(", ", $result);

Conclusion:

This article presented methods for anomaly detection and outlier handling using PHP and machine‑learning techniques. By applying the Z‑Score method and Isolation Forest algorithm, outliers can be identified, and then either removed or replaced, helping to clean data, improve model accuracy, and enable more reliable analysis and prediction.

Java learning material

C language learning material

Frontend learning material

C++ learning material

PHP learning material