Apparatus and program for detecting abnormal behavior

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.

1 . An abnormal behavior detection apparatus comprising: 
 a probabilistic distribution estimation apparatus for responding 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, said probabilistic distribution estimation apparatus 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 having the respective hidden variables from said parameter storage unit;    posteriori probability calculation means for calculating a posteriori probability of the state corresponding to the hidden variables by using the parameters of the probabilistic distribution estimated by said probabilistic distribution estimation apparatus;    a reference data input unit for inputting data different from said input data; and    posteriori probability variation degree calculation means for calculating a variation of the posteriori distribution and outputting it by using the posteriori distribution of the state corresponding to the hidden variables calculated by said posteriori distribution calculation means on the basis of the data read out of said reference data input unit and by using the posteriori distribution of the state corresponding to the hidden variables calculated by said posteriori distribution calculation mean on the basis of the new read data.    
   
   
       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 1 , 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 probabilistic distribution estimation apparatus for responding 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, said probabilistic distribution estimation apparatus 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 having the respective hidden variables from said parameter storage unit;    posteriori probability calculation means for calculating a posteriori probability of the state corresponding to the hidden variables by using the parameters of the probabilistic distribution estimated by said probabilistic distribution estimation apparatus;    a reference data input unit for inputting data different from said input data; and    posteriori probability variation degree calculation means for calculating a variation of the posteriori distribution and outputting it by using the posteriori distribution of the state corresponding to the hidden variables calculated by said posteriori distribution calculation means on the basis of the data read out of said reference data input unit and by using the posteriori distribution of the state corresponding to the hidden variables calculated by said posteriori distribution calculation mean on the basis of the new read data.    
   
   
       8 . An abnormal behavior detection apparatus according to  claim 7 , wherein said probabilistic distribution estimation apparatus further comprises session means for processing the input data into the string of vector data.