18053738. Anomaly Detection with Local Outlier Factor simplified abstract (Google LLC)

From WikiPatents
Jump to navigation Jump to search

Anomaly Detection with Local Outlier Factor

Organization Name

Google LLC

Inventor(s)

Xi Cheng of Kirkland WA (US)

Zichuan Ye of Mountain View CA (US)

Peng Lin of Mountain View CA (US)

Jiashang Liu of Kirkland WA (US)

Amir Hormati of Mountain View CA (US)

Mingge Deng of Kirkland WA (US)

Anomaly Detection with Local Outlier Factor - A simplified explanation of the abstract

This abstract first appeared for US patent application 18053738 titled 'Anomaly Detection with Local Outlier Factor

Simplified Explanation

The abstract describes a method for anomaly detection using a local outlier factor (LOF) algorithm. Here is a simplified explanation of the abstract:

  • The method receives a query from a user requesting anomaly detection in a dataset.
  • A model is trained using the dataset, specifically using the LOF algorithm.
  • For each example in the dataset, a local deviation score is determined based on its features using the trained model.
  • If the local deviation score satisfies a threshold, the example is considered anomalous.
  • The method reports the anomalous examples to the user.

Potential Applications

This technology can be applied in various domains where anomaly detection is crucial, such as:

  • Cybersecurity: Identifying unusual network traffic or malicious activities.
  • Fraud detection: Detecting fraudulent transactions or activities.
  • Manufacturing: Identifying anomalies in production processes to prevent defects or failures.
  • Healthcare: Detecting abnormal patterns in patient data for early disease diagnosis.
  • Financial markets: Identifying unusual trading patterns or market manipulation.

Problems Solved

The method addresses the following problems in anomaly detection:

  • Efficient detection: The LOF algorithm allows for efficient identification of anomalies in large datasets.
  • Automated detection: The method automates the process of anomaly detection, reducing the need for manual inspection.
  • Real-time detection: Anomalies can be detected in real-time, enabling timely responses to potential threats or issues.
  • Accurate detection: The trained model provides accurate anomaly detection based on the local deviation scores.

Benefits

This technology offers several benefits:

  • Improved security: Anomalies can be quickly identified, enhancing security measures and reducing potential risks.
  • Cost savings: Automated anomaly detection reduces the need for manual inspection, saving time and resources.
  • Early detection: Anomalies can be detected early, allowing for prompt actions to mitigate potential damages.
  • Scalability: The method can handle large datasets efficiently, making it suitable for various industries and applications.


Original Abstract Submitted

A method for anomaly detection includes receiving an anomaly detection query from a user. The anomaly detection query requests data processing hardware determine one or more anomalies in a dataset including a plurality of examples. Each example in the plurality of examples is associated with one or more features. The method includes training a model using the dataset. The trained model is configured to use a local outlier factor (LOF) algorithm. For each respective example of the plurality of examples in the dataset, the method includes determining, using the trained model, a respective local deviation score based on the one or more features. The method includes determining that the respective local deviation score satisfies a deviation score threshold and, based on the location deviation score satisfying the threshold, determining that the respective example is anomalous. The method includes reporting the respective anomalous example to the user.

Retrieved from "http://wikipatents.org/index.php?title=18053738._Anomaly_Detection_with_Local_Outlier_Factor_simplified_abstract_(Google_LLC)&oldid=24294"