Information processing device, method of processing information and storage medium

• US 9,104,641 B2
• Filed: 12/07/2010
• Issued: 08/11/2015
• Est. Priority Date: 12/15/2009
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Aoki, Kenji, Nakata, Takayuki
• Primary Examiner(s): Ngo, Chuong D
• Application Number:
  US13/515,554
• Publication Number:
  US 20120331024A1
• Time in Patent Office: 1,708 Days
• Field of Search: 708/200
• US Class Current: 1/1
• CPC Class Codes: G01S 19/396 Determining accuracy or reliability of position or pseudorange measurements
• 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

Abstract

An information processing device calculates a mean and a variance of a response variable for each range determined based on a partition candidate being information on partitioning a domain of an explanatory variable into a plurality of ranges based on observed data and the mean and the variance are dependent on the explanatory variable, selects a specific partition candidate among a plurality of the partition candidates using an information criterion calculated based on the observed data and the calculated mean and variance of the response variable, and calculates a smooth function representing the variance of the response variable in the domain of the explanatory variable and a smooth function representing the mean of the response variable in the domain of the explanatory variable based on the mean and the variance calculated for each range determined based on the specific partition candidate.

Claims

1. An information processing device, comprising:

a memory storing instructions; and

one or more processors configured to process the instructions to;

calculate a mean and a variance of a response variable for each range which is determined based on a partition candidate being information on partitioning a domain of an explanatory variable into a plurality of ranges based on observed data including information representing the explanatory variable and information representing the response variable being an observed value in terms of the explanatory variable and the mean and the variance are dependent on the explanatory variable;

select a specific partition candidate among a plurality of the partition candidates by using an information criterion which is calculated based on the observed data and the calculated mean and variance of the response variable; and

calculate a smooth function representing the variance of the response variable in the domain of the explanatory variable and a smooth function representing the mean of the response variable in the domain of the explanatory variable based on the mean and the variance calculated for each range determined based on the specific partition candidate.

2. The information processing device according to claim 1, wherein in the calculating the mean and the variance of the response variable, the mean and the variance of each range are of normal distribution and the mean and the variance of the response variable are calculated for the each range determined by the partition candidate.

3. The information processing device according to claim 1, wherein in the selecting the specific partition candidate, the partition candidate in which the calculated information criterion is the smallest is selected as the specific partition candidate.

4. The information processing device according to claim 1, wherein in the calculating the smooth function, the smooth function is calculated based on a position of the explanatory variable in each range and the calculated mean and variance for each range determined by using the specific partition candidate.

5. The information processing device according to claim 1, wherein in the calculating the smooth function, the smooth function is calculated based on dispersion of the explanatory variable in each range and the calculated mean and variance for each range determined by using the specific partition candidate.

6. The information processing device according to claim 1, wherein in the calculating the smooth function, a Kernel function is calculated with respect to the calculated mean and variance for each range determined based on the specific partition candidate, and the smooth function is calculated based on the Kernel function.

7. The information processing device according to claim 1, wherein the one or more processors are further configured to process the instructions to:

calculate the partition candidate based on the observed data, information representing the domain of explanatory variable, the maximum value for the number of partitions in a case of partitioning the domain of the explanatory variable into a plurality of ranges and the minimum block for each range, wherein

in the calculating the mean and the variance of the response variable, the mean and the variance of the response variable are calculated for each range determined based on the specific partition candidate.

8. The information processing device according to claim 1, wherein the information criterion is either of Akaike'"'"'s information criterion and Bayesian information criterion.

9. An information processing device, comprising:

parameter estimation means for calculating a mean and a variance of a response variable for each range which is determined based on a partition candidate being information on partitioning a domain of an explanatory variable into a plurality of ranges based on observed data including information representing the explanatory variable and information representing the response variable being an observed value in terms of the explanatory variable and the mean and the variance are dependent on the explanatory variable;

optimum partition specification means for selecting a specific partition candidate among a plurality of the partition candidates by using an information criterion which is calculated based on the observed data and the calculated mean and variance of the response variable; and

smoothing means for calculating a smooth function representing the variance of the response variable in the domain of the explanatory variable and a smooth function representing the mean of the response variable in the domain of the explanatory variable based on the mean and the variance calculated for each range determined based on the specific partition candidate.

10. A method of processing information, comprising:

calculating, using one or more processors, a mean and a variance of a response variable for each range which is determined based on a partition candidate being information on partitioning a domain of an explanatory variable into a plurality of ranges based on observed data including information representing the explanatory variable and information representing the response variable being an observed value in terms of the explanatory variable and the mean and the variance are dependent on the explanatory variable;

selecting, using the one or more processors, a specific partition candidate among a plurality of the partition candidates by using an information criterion which is calculated based on the observed data and the calculated mean and variance of the response variable; and

calculating, using the one or more processors, a smooth function representing the variance of the response variable in the domain of the explanatory variable and a smooth function representing the mean of the response variable in the domain of the explanatory variable based on the mean and the variance calculated for each range determined based on the specific partition candidate.

11. A non-transitory storage medium that stores a computer program to enable a computer to execute a process, the process comprising:

a parameter estimation process that calculates a mean and a variance of a response variable for each range which is determined based on a partition candidate being information on partitioning a domain of an explanatory variable into a plurality of ranges based on observed data including information representing the explanatory variable and information representing the response variable being an observed value in terms of the explanatory variable and the mean and the variance are dependent on the explanatory variable;

an optimum partition specification process that selects a specific partition candidate among a plurality of the partition candidates by using an information criterion which is calculated based on the observed data and the calculated mean and variance of the response variable; and

a smoothing process that calculates a smooth function representing the variance of the response variable in the domain of the explanatory variable and a smooth function representing the mean of the response variable in the domain of the explanatory variable based on the mean and the variance calculated for each range determined based on the specific partition candidate.

12. An information processing device, comprising:

a memory storing instructions; and

one or more processors configured to process the instructions to;

input observed data including information representing an explanatory variable and information representing an response variable being an observed value in terms of the explanatory variable and the mean and the variance are dependent on the explanatory variable;

calculate a mean and a variance of the response variable for each range which is determined based on a partition candidate being information on partitioning a domain of the explanatory variable into a plurality of ranges based on the input observed data;

select a specific partition candidate among a plurality of the partition candidates by using an information criterion which is calculated based on the observed data and the calculated mean and variance of the response variable;

transform from the input obtained data to a first smooth function representing the variance of the response variable in the domain of the explanatory variable and a second smooth function representing the mean of the response variable in the domain of the explanatory variable based on the mean and the variance calculated for each range determined based on the specific partition candidate; and

output the first smooth function and the second smooth function.