18438717. Point Anomaly Detection simplified abstract (Google LLC)
Contents
Point Anomaly Detection
Organization Name
Inventor(s)
Zichuan Ye of Mountain View CA (US)
Jiashang Liu of Kirkland WA (US)
Forest Elliott of Mountain View CA (US)
Amir Hormati of Mountain View CA (US)
Mingge Deng of Kirkland WA (US)
Point Anomaly Detection - A simplified explanation of the abstract
This abstract first appeared for US patent application 18438717 titled 'Point Anomaly Detection
The method described in the abstract involves using data processing hardware to detect anomalous point data values in a set of point data values based on a user query.
- The method includes training a model with the set of point data values to determine anomalous values.
- For each point data value in the set, the method calculates the variance using the trained model.
- If the calculated variance value meets a threshold, the respective point data value is identified as anomalous.
- The method then reports the identified anomalous point data value back to the user.
Potential Applications: - Anomaly detection in various data sets such as sensor data, financial data, or network traffic. - Quality control in manufacturing processes by identifying faulty data points. - Fraud detection in financial transactions by flagging suspicious data points.
Problems Solved: - Efficiently detecting anomalous data points in large datasets. - Providing real-time feedback to users on potentially problematic data values. - Improving overall data quality and accuracy in various applications.
Benefits: - Enhances data analysis by pinpointing outliers and anomalies. - Helps in identifying potential issues or irregularities in data sets. - Enables proactive decision-making based on accurate and reliable data.
Commercial Applications: Title: "Advanced Anomaly Detection Method for Data Processing Systems" This technology can be utilized in industries such as finance, healthcare, and manufacturing for data quality assurance, fraud detection, and process optimization. The market implications include improved efficiency, reduced risks, and enhanced decision-making capabilities.
Questions about Anomaly Detection Method: 1. How does the trained model determine the variance value for each point data value? 2. What are the key factors that determine the threshold value for identifying anomalous data points?
Frequently Updated Research: Stay updated on advancements in machine learning algorithms for anomaly detection and data processing systems to enhance the efficiency and accuracy of anomaly detection methods.
Original Abstract Submitted
A method includes receiving a point data anomaly detection query from a user. The query requests the data processing hardware to determine a quantity of anomalous point data values in a set of point data values. The method includes training a model using the set of point data values. For at least one respective point data value in the set of point data values, the method includes determining, using the trained model, a variance value for the respective point data value and determining that the variance value satisfies a threshold value. Based on the variance value satisfying the threshold value, the method includes determining that the respective point data value includes an anomalous point data value. The method includes reporting the determined anomalous point data value to the user.
