Sentence classification apparatus, sentence classification method, and sentence classification program

Abstract

A sentence classification apparatus, a sentence classification method, and a sentence classification program that can appropriately classify sentences by adjusting parameters are provided. The sentence classification apparatus includes: a case sentence obtention unit for obtaining plural case sentences which are associated with effect values that are values obtained by evaluating effects; a case value creation unit for creating case values obtained by numerizing the case sentences for each of parameters with different values; a correlation coefficient calculation unit for calculating a correlation coefficient between the case values and the effect values for each of the values of the parameters; and a parameter selection unit for selecting a parameter among the parameters with different values on the basis of the correlation coefficient.

1 . A sentence classification apparatus for classifying sentences into classified groups, comprising:
 a case sentence obtention unit for obtaining a plurality of case sentences which are associated with effect values that are values obtained by evaluating effects;   a case value creation unit for creating case values obtained by numerizing the case sentences for each of parameters with different values;   a correlation coefficient calculation unit for calculating a correlation coefficient between the case values and the effect values for each of the values of the parameters; and   parameter selection unit for selecting a parameter among the parameters with different values on the basis of the correlation coefficient,   wherein the case value creation unit clusters sentences into a plurality of clusters,   wherein the correlation coefficient calculation unit calculates a correlation coefficient for each cluster, and   wherein the parameter selection unit selects the parameter on the basis of the correlation coefficient for each cluster.   
     
     
         2 . The sentence classification apparatus according to  claim 1 ,
 wherein the parameter selection unit selects the parameter on the basis of the number of correlation coefficients that exceeds a certain threshold for each parameter.   
     
     
         3 . The sentence classification apparatus according to  claim 1 ,
 wherein the parameter selection unit selects the parameter on the basis of the average value of the correlation coefficients calculated for each parameter.   
     
     
         4 . The sentence classification apparatus according to  claim 1 , further comprising:
 a classification outcome display unit for displaying the outcome obtained by classifying the plurality of case sentences using the parameter selected by the parameter selection unit.   
     
     
         5 . The sentence classification apparatus according to  claim 4 ,
 wherein the classification outcome display unit displays representative sentences that are sentences representing classified groups respectively.   
     
     
         6 . The sentence classification apparatus according to  claim 5 ,
 wherein the classification outcome display unit includes a selection unit for selecting a classified group among the classified groups and displays a case sentence included in the classified group selected by the selection unit.   
     
     
         7 . The sentence classification apparatus according to  claim 1 ,
 wherein the case sentence obtention unit obtains a case sentence that meets a condition set on a condition setting screen where a condition for selecting the parameter is set.   
     
     
         8 . The sentence classification apparatus according to  claim 1 ,
 wherein the parameter selection unit selects a parameter that makes the absolute value of the correlation coefficient maximum.   
     
     
         9 . The sentence classification apparatus according to  claim 1 ,
 wherein the case sentences include condition sentences showing the conditions for the cases and outcome sentences showing the outcomes of the cases respectively; and   wherein the case value creation unit creates condition values obtained by numerizing the condition sentences and outcome values obtained by numerizing the outcome sentences.   
     
     
         10 . The sentence classification apparatus according to  claim 8 ,
 wherein the case value creation unit clusters the condition sentences and the outcome sentences into the clusters of the condition sentences and the clusters of the outcome sentences respectively, and at the same time, extracts the clusters of the condition sentences associated with the clusters of the outcome sentences respectively,   wherein the correlation coefficient calculation unit calculates the correlation coefficient for each of the extracted clusters of the condition sentences, and   wherein the parameter selection unit selects the parameter that makes the correlation coefficients between effect values associated with the clusters of the outcome sentences and the clusters of the condition sentences maximum.   
     
     
         11 . The sentence classification apparatus according to  claim 1 ,
 wherein the case value creation unit creates a word vector database in which words are associated with word vectors respectively for each of the parameters with different values on the basis of the case sentences obtained by the case sentence obtention unit and creates the case values using the word vector database.   
     
     
         12 . A sentence classification method for classifying sentences into classified groups, comprising:
 a case sentence obtention step for obtaining a plurality of case sentences which are associated with effect values that are values obtained by evaluating effects;   a case value creation step for creating case values obtained by numerizing the case sentences for each of parameters with different values;   a correlation coefficient calculation step for calculating a correlation coefficient between the case values and the effect values for each of the values of the parameters; and   parameter selection unit step for selecting a parameter among the parameters with different values on the basis of the correlation coefficient, wherein the case value creation unit clusters sentences into a plurality of clusters,   wherein the correlation coefficient calculation unit calculates a correlation coefficient for each cluster, and   wherein the parameter selection unit selects the parameter on the basis of the correlation coefficient for each cluster.   
     
     
         13 . A non-transitory computer readable medium that stores a sentence classification program for making a computer classify sentences into classified groups, the program causes the computer to execute steps comprising:
 a case sentence obtention step for obtaining a plurality of case sentences which are associated with effect values that are values obtained by evaluating effects;   a case value creation step for creating case values obtained by numerizing the case sentences for each of parameters with different values;   a correlation coefficient calculation step for calculating a correlation coefficient between the case values and the effect values for each of the values of the parameters; and   parameter selection unit step for selecting a parameter among the parameters with different values on the basis of the correlation coefficient, wherein the case value creation unit clusters sentences into a plurality of clusters,   wherein the correlation coefficient calculation unit calculates a correlation coefficient for each cluster, and   wherein the parameter selection unit selects the parameter on the basis of the correlation coefficient for each cluster.