Method of predicting initial input of new product, system for predicting initial input of new product, and recording medium
First Claim
1. A method of predicting an initial input of a new product for predicting a sales volume or market share at the launch of a new product, comprising steps of:
- making more than one people numerically evaluate a plurality of past products and said new product, with respect to a plurality of factors which are considered to influence a sales volume or market share;
calculating comprehensive evaluations on said products and said new products for each one of said people based on numerical values with respect to said factors, calculating correlation coefficients between the comprehensive evaluations on said products by each one of said people and actual sales volumes or actual market shares of said products, obtaining relationships between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products, calculating a sales volume or market share of said new product for each one of said people based on the relationships and the comprehensive evaluations on said new product, layering the calculated sales volumes or market shares of said new product for said more than one people based on the correlation coefficients for said more than one people, calculating average values of the sales volumes or market shares of said new product and confidence intervals in the respective layers; and
predicting a sales volume or market share of said new product based on the average values and the confidence intervals for the respective layers.
1 Assignment
0 Petitions
Accused Products
Abstract
Based on numerical values with respect to factors influencing shares of existing products and a new product evaluated by more than one people, a structured neural network calculates predictive shares of the new product predicted by the respective persons (S6). Comprehensive evaluations on the respective products and the new product are calculated for each person, based on the numerical values with respect to the respective factors (S10). Correlation coefficients between the comprehensive evaluations on the respective products by the respective persons and the actual shares are calculated (S12), and relationships between the are obtained (S12). Predictive shares of the new product are calculated for the respective persons based on the relationships and the comprehensive evaluations on the new product (S16). The predictive shares are layered out in accordance with the correlation coefficients for the respective persons (S18). Average values of the predictive shares and confidence intervals are calculated for the respective layers (S20), and a share of the new product is predicted based on the average values and the confidence intervals for the respective layers and the calculation result obtained by the structured neural network.
51 Citations
18 Claims
-
1. A method of predicting an initial input of a new product for predicting a sales volume or market share at the launch of a new product, comprising steps of:
-
making more than one people numerically evaluate a plurality of past products and said new product, with respect to a plurality of factors which are considered to influence a sales volume or market share;
calculating comprehensive evaluations on said products and said new products for each one of said people based on numerical values with respect to said factors, calculating correlation coefficients between the comprehensive evaluations on said products by each one of said people and actual sales volumes or actual market shares of said products, obtaining relationships between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products, calculating a sales volume or market share of said new product for each one of said people based on the relationships and the comprehensive evaluations on said new product, layering the calculated sales volumes or market shares of said new product for said more than one people based on the correlation coefficients for said more than one people, calculating average values of the sales volumes or market shares of said new product and confidence intervals in the respective layers; and
predicting a sales volume or market share of said new product based on the average values and the confidence intervals for the respective layers. - View Dependent Claims (2, 8, 9)
-
-
3. A method of predicting an initial input of a new product for predicting a sales volume or market share at the launch of a new product, wherein
more than one people numerically evaluate a plurality of past products and said new product, with respect to a plurality of factors which are considered to influence a sales volume or market share, a structured neural network calculates a sales volume or market share of said new product for each one of said people based on numerical values on said products and said new products given by each one of said people with respect to said factors and actual sales volumes or actual market shares of said products, and obtains said factors contributing to the calculation and the magnitudes of said factors, comprehensive evaluations on said products and said new products for each one of said people are calculated based on the numerical values with respect to said factors, correlation coefficients are calculated between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products, the sales volumes or market shares of said new product calculated for said more than one people by said structured neural network are layered out based on the correlation coefficients for said more than one people, average values and confidence intervals of the sales volumes or market shares of said new product in the respective layers are calculated, and a sales volume or market share of said new product is predicted based on the average values and the confidence intervals for the respective layers, the sales volumes or market shares calculated for said more than one people by said structured neural network, and said factors contributing to the calculation and the magnitudes of said factors.
-
4. A method of predicting an initial input of a new product for predicting a sales volume or market share at the launch of a new product, wherein
more than one people numerically evaluate a plurality of past products and said new product, with respect to a plurality of factors which are considered to influence a sales volume or market share, a structured neural network calculates a sales volume or market share of said new product for each one of said people based on numerical values on said products and said new products given by each one of said people with respect to said factors and actual sales volumes or actual market shares of said products, and obtains said factors contributing to the calculation and the magnitudes of said factors, comprehensive evaluations on said products and said new products for each one of said people are calculated based on the numerical values with respect to said factors, relationships are calculated between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products, the sales volumes or market shares of said new product are calculated for said more than one people based on the relationships and the comprehensive evaluations on said new product, and a sales volume or market share of said new product is predicted based on the calculated sales volumes or market shares of said new product for said more than one people, the sales volumes or market shares calculated for said more than one people by said structured neural network, and said factors contributing to the calculation and the magnitudes of said factors.
-
6. A method of predicting an initial input of a new product for predicting a sales volume or market share at the launch of a new product, wherein
more than one people numerically evaluate a plurality of past products and said new product, with respect to a plurality of factors which are considered to influence a sales volume or market share, a structured neural network calculates a sales volume or market share of said new product for each one of said people based on numerical values on said products and said new products given by each one of said people with respect to said factors and actual sales volumes or actual market shares of said products, and obtains said factors contributing to the calculation and the magnitudes of said factors, comprehensive evaluations on said products and said new products for each one of said people are calculated based on the numerical values with respect to said factors, correlation coefficients are calculated between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products, relationships are calculated between the comprehensive evaluations by each one of said people and the actual sales volumes or actual market shares of said products, the sales volumes or market shares of said new product are calculated for said more than one people based on the relationships and the comprehensive evaluations on said new product, the calculated sales volumes or market shares of said new product for said more than one people are layered out based on the correlation coefficients for said more than one people, average values and confidence intervals of the sales volumes or market shares of said new product in the respective layers are calculated, and a sales volume or market share of said new product is predicted based on the average values and the confidence intervals for the respective layers, the sales volumes or market shares calculated for said more than one people by said structured neural network, and said factors contributing to the calculation and the magnitudes of said factors.
-
15. A computer readable recording medium which stores a program for causing a computer to predict a sales volume or market share at the launch of a new product, said recording medium storing a program including:
-
a procedure for causing a computer to input data which are evaluations in the form of numerical values made by more than one people on a plurality of past products and said new product, with respect to a plurality of factors, including actual sales volume or actual market shares of said products, which are considered to influence a sales volume or market share;
a procedure for causing a computer to calculate comprehensive evaluations on said products and said new products for each one of the people based on the numerical values with respect to said factors;
a procedure for causing a computer to calculate correlation coefficients between said comprehensive evaluations on said products by each one of the people and the actual sales volumes or actual market shares of said products;
a procedure for causing a computer to calculate relationships between said comprehensive evaluations on said products by each one of the people and the actual sales volumes or actual market shares of said products;
a procedure for causing a computer to calculate the sales volumes or market shares of said new product for the more than one people, based on said relationships and said comprehensive evaluations on said new product;
a procedure for causing a computer to layer out the sales volumes or market shares of said new product calculated through said procedure for the more than one people, based on said correlation coefficients for the more than one people;
a procedure for causing a computer to calculate average values and confidence intervals of the sales volumes or market shares for the respective layers which are created through said procedure; and
a procedure for causing a computer to output the average values and the confidence intervals in the respective layers calculated through said procedure.
-
-
16. A computer readable recording medium which stores a program for causing a computer to predict a sales volume or market share at the launch of a new product, said recording medium storing:
-
a program for causing a computer to execute a procedure for entering data which are evaluations in the form of numerical values made by more than one people on a plurality of past products and said new product, with respect to a plurality of factors, including actual sales volumes or actual market shares of said products, which are considered to influence a sales volume or market share;
a structured neural network which calculates sales volumes or market shares of said new product for said more than one people based on the numerical values with respect to said factors given on said products and said new product given by said more than one people and the actual sales volumes or actual market shares of said products, said structured neural network obtaining said factors contributing to the calculation and the magnitudes of said factors; and
a program for causing a computer to execute the following procedures which are;
a procedure of calculating comprehensive evaluations on said products and said new products for each one of said people based on the numerical values with respect to said factors;
a procedure of calculating correlation coefficients between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products;
a procedure of layering the sales volumes or market shares of said new product calculated by said structured neural network for said more than one people, based on the correlation coefficients for said more than one people;
a procedure of calculating average values and confidence intervals of the sales volumes or market shares for the respective layers which are created through said procedure; and
a procedure of outputting the average values and the confidence intervals in the respective layers calculated through said procedure, the sales volumes or market shares calculated for said more than one people by said structured neural network, and said factors contributing to the calculation and the magnitudes of said factors.
-
-
17. A computer readable recording which stores a program which causes a computer to predict a sales volume or market share at the launch of a new product, said recording medium storing:
-
a program for causing a computer to execute a procedure for entering data which are evaluations in the form of numerical values made by more than one people on a plurality of past products and said new product, with respect to a plurality of factors, including actual sales volumes or actual market shares of said products, which are considered to influence a sales volume or market share;
a structured neural network which calculates sales volumes or market shares of said new product for said more than one people based on the numerical values with respect to said factors given on said products and said new product given by said more than one people and the actual sales volumes or actual market shares of said products, said structured neural network obtaining said factors contributing to the calculation and the magnitudes of said factors; and
a program for causing a computer to execute the following procedures which are;
a procedure of calculating comprehensive evaluations on said products and said new products for each one of said people based on the numerical values with respect to said factors;
a procedure of obtaining relationships between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products;
a procedure of calculating the sales volumes or market shares of said new product for said more than one people based on the relationships and the comprehensive evaluations on said new product; and
a procedure of outputting the sales volumes or market shares of said new product calculated for said more than one people through said procedure, the sales volumes or market shares calculated for said more than one people by said structured neural network, and said factors contributing to the calculation and the magnitudes of said factors.
-
-
18. A computer readable recording medium which stores a program which causes a computer to predict a sales volume or market share at the launch of a new product, said recording medium storing:
-
a program for causing a computer to execute a procedure for entering data which are evaluations in the form of numerical values made by more than one people on a plurality of past products and said new product, with respect to a plurality of factors, including actual sales volumes or actual market shares of said products, which are considered to influence a sales volume or market share;
a structured neural network which calculates sales volumes or market shares of said new product for said more than one people based on the numerical values with respect to said factors given on said products and said new product given by said more than one people and the actual sales volumes or actual market shares of said products, said structured neural network obtaining said factors contributing to the calculation and the magnitudes of said factors; and
a program for causing a computer to execute the following procedures which are;
a procedure of calculating comprehensive evaluations on said products and said new products for each one of said people based on the numerical values with respect to said factors;
a procedure of calculating correlation coefficients between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products;
a procedure of obtaining relationships between the comprehensive evaluations on said products by each one of said people and the actual sales volumes or actual market shares of said products;
a procedure of calculating the sales volumes or market shares of said new product for said more than one people, based on the relationships and the comprehensive evaluations on said new product;
a procedure of layering the sales volumes or market shares of said new product calculated through said procedure for said more than one people, based on the correlation coefficients for said more than one people;
a procedure of calculating average values and confidence intervals of the sales volumes or market shares for the respective layers which are created through said procedure; and
a procedure of outputting the average values and the confidence intervals in the respective layers, the sales volumes or market shares calculated for said more than one people by said structured neural network, and said factors contributing to the calculation and the magnitudes of said factors.
-
Specification