Method of detecting abnormal behavior

• US 20070255542A1
• Filed: 07/06/2007
• Published: 11/01/2007
• Est. Priority Date: 02/18/2003
• Status: Abandoned Application

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Yamanishi, Kenji, Matsunaga, Yuko
• Application Number:
  US11/822,460
• Publication Number:
  US 20070255542A1
• Time in Patent Office: Days
• US Class Current: 703/2
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06F 18/295 Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models

Abstract

Supplied with a string of vector data as input data, a probabilistic distribution estimation apparatus estimates, by using a stochastic model having hidden variables, a probabilistic distribution in which each data occurs by successively reading the train of vector data. Specifically, the probabilistic distribution estimation apparatus reads values of parameters of the stochastic model having the hidden variables for a value of the input data, calculates, by using the stochastic model, a certainty in which the input data occurs, renews the parameters in response to new read data with past data forgotten, and produce several parameter'"'"'s values. By using the parameter'"'"'s values received from the probabilistic distribution estimation apparatus, an abnormality detection unit calculates an information amount of data as an abnormal behavior degree to produce the abnormal behavior degree.

Claims

1. An abnormal behavior detection apparatus comprising:

a plurality of probabilistic distribution estimation apparatuses each of which responds to, as input data, a string of vector data to estimate, using a stochastic model, a probabilistic distribution occurred in each data by successively reading said string of vector data, each of said probabilistic distribution estimation apparatuses comprising a parameter storage unit for storing all of parameters for the stochastic model having hidden variables, certainty calculation means for calculating, in response to said input data, a certainty where said input data occurs using said stochastic model by reading the parameters of said stochastic model from said parameter storage unit, and parameter renewal means for renewing contents of said parameter storage unit in accordance with new read data with past data forgotten by reading the certainty from said certainty calculation means and by reading each parameter of said stochastic model from said parameter storage unit; and

information amount standard calculation means for calculating, by using, in parallel, said plurality of probabilistic distribution estimation apparatuses for said stochastic models having different number of the states where the hidden variables can take, standard values of information amounts from the parameters of the probabilistic distributions estimated by the respective probabilistic distribution estimation apparatuses and the input data to produce, as an optimum value, the number of states where the hidden variables can take when the standard value of the information amount is the least.

2. An abnormal behavior detection apparatus according to claim 1, further comprising:

session means for processing the input data into the string of vector data.

3. An -abnormal behavior detection apparatus according to claim 1, wherein the time series model has a continuous time distribution and hidden variables.

4. An abnormal behavior detection apparatus according to claim 1, further comprising:

session means for processing the input data into the string of vector data.

5. An abnormal behavior detection apparatus according to claim 1, wherein a finite mixed distribution of hidden Markov models is used to estimate the probabilistic distribution occurred in each data.

6. An abnormal behavior detection apparatus according to claim 5, further comprising:

session means for processing the input data into the string of vector data.

7. An abnormal behavior detection program for making a computer operate as:

a plurality of probabilistic distribution estimation apparatuses each of which responds to, as input data, a string of vector data to estimate, using a stochastic model, a probabilistic distribution occurred in each data by successively reading said string of vector data, each of said probabilistic distribution estimation apparatuses comprising a parameter storage unit for storing all of parameters for the stochastic model having hidden variables, certainty calculation means for calculating, in response to said input data, a certainty where said input data occurs using said stochastic model by reading the parameters of said stochastic model from said parameter storage unit, and parameter renewal means for renewing contents of said parameter storage unit in accordance with new read data with past data forgotten by reading the certainty from said certainty calculation means and by reading each parameter of said stochastic model from said parameter storage unit; and

information amount standard calculation means for calculating, by using, in parallel, said plurality of probabilistic distribution estimation apparatuses for said stochastic models having different number of the states where the hidden variables can take, standard values of information amounts from the parameters of the probabilistic distributions estimated by the respective probabilistic distribution estimation apparatuses and the input data to produce, as an optimum value, the number of states where the hidden variables can take when the standard value of the information amount is the least.

8. An abnormal behavior detection program according to claim 7, wherein each of said probabilistic distribution estimation apparatuses comprises session means for processing the input data into the string of vector data.