Information processing apparatus and non-transitory computer readable medium

A non-transitory computer readable medium storing a program causing a computer to execute a process for information processing, the process comprising:

• acquiring question document contents and answer document contents, which are in a relationship of questions and answers, and calculating a feature amount of each of the answer document contents;
• generating distribution map information by plotting each of the answer document contents in a feature amount space on the basis of the calculated feature amount;
• calculating distances in the feature amount space between the answer document contents on the basis of the calculated feature amounts;
• classifying the answer document contents so that, in a case that there is the calculated distance equal to or greater than a value set in advance between a first set of the answer document contents and a second set of the answer document contents, first attribute information is added to the first set of the answer document contents, and second attribute information different from the first attribute information is added to the second set of the answer document contents;
• and
• classifying the question document contents on the basis of a classifying result of the classifying of the answer document contents so that, the first attribute information is added to a first set of the question document contents corresponding to the first set of the answer document contents and the second attribute information is added to a second set of the question document contents corresponding to the second set of the answer document contents.

A non-transitory computer readable medium storing a program causing a computer to execute a process for information processing, the process comprising:

• acquiring question document contents and answer document contents, which are in a relationship of questions and answers, and determination contents each indicating a point of view for generating the answer from the question, and classifying the answer document contents in which a first one of the determination contents is set to a first set of the answer document contents and a second one of the determination contents different from the first one of the determination contents is set to a second set of the answer document contents;
• calculating a first feature amount of each of the first set of the answer document contents and a second feature amount of each of the second set of the answer document contents;
• generating first distribution map information by plotting each of the first set of the answer document contents in a first feature amount space on the basis of the calculated first feature amount, and second distribution map information by plotting each of the second set of the answer document contents in a second feature amount space on the basis of the calculated second feature amount;
• calculating first distances in the first feature amount space between the first set of the answer document contents on the basis of the calculated first feature amounts and second distances in the second feature amount space between the second set of the answer document contents on the basis of the calculated second feature amounts;
• classifying the first set of the answer document contents so that, in a case that there is the first distance equal to or greater than a value set in advance between a first subset of the first set of the answer document contents and a second subset of the first set of the answer document contents, first attribute information is added to the first subset of the first set of the answer document contents and second attribute information different from the first attribute information is added to the second subset of the first set of the answer document contents, and classifying the second set of the answer document contents so that, in a case that there is the second distance equal to or greater than the value set in advance between a first subset of the second set of the answer document contents and a second subset of the second set of the answer document contents, third attribute information is added to the first subset of the second set of the answer document contents and fourth attribute information different from the third attribute information is added to the second subset of the second set of the answer document contents;
• and
• notifying one of the first and second determination contents, which reveals good classifying separation to the corresponding answer document contents, as a remarkable point for generating the answer from the question on the basis of classifying results of the classifying.

A non-transitory computer readable medium storing a program causing a computer to execute a process for information processing, the process comprising:

• acquiring question document contents and answer document contents, which are in a relationship of questions and answers, and determination contents each indicating a point of view for generating the answer from the question, and classifying the answer document contents in which a first one of the determination contents is set to a first set of the answer document contents and a second one of the determination contents different from the first one of the determination contents is set to a second set of the answer document contents;
• calculating a first feature amount of each of the first set of the answer document contents;
• generating first distribution map information by plotting each of the first set of the answer document contents in a first feature amount space on the basis of the calculated first feature amount;
• classifying the first set of the answer document contents so that, in a case that there is the first distance equal to or greater than a value set in advance between a first subset of the first set of the answer document contents and a second subset of the first set of the answer document contents, first attribute information is added to the first subset of the first set of the answer document contents and second attribute information different from the first attribute information is added to the second subset of the first set of the answer document contents;
• classifying a first set of the question document contents on the basis of a classifying result of the classifying of the first set of the answer document contents so that, the first attribute information is added to a first subset of the first set of the question document contents corresponding to the first subset of the first set of the answer document contents and the second attribute information is added to a second subset of the first set of the question document contents corresponding to the second subset of the first set of the answer document contents;
• and
• adding the first attribute information to the first subset of the first set of the question document contents;
• and the second attribute information to the second subset of the first set of the question document contents.

An information processing apparatus, comprising:

• a feature amount calculating portion that acquires question document contents and answer document contents, which are in a relationship of questions and answers, and calculates a feature amount of each of the answer document contents;
• a distribution map generating portion generates distribution map information by plotting each of the answer document contents in a feature amount space on the basis of the calculated feature amount;
• a distance calculation portion that calculates distances in the feature amount space between the answer document contents on the basis of the calculated feature amounts;
• and
• a classifying portion that classifies the answer document contents so that, in a case that there is the calculated distance equal to or greater than a value set in advance between a first set of the answer document contents and a second set of the answer document contents, first attribute information is added to the first set of the answer document contents, and second attribute information different from the first attribute information is added to the second set of the answer document contents, wherein
• the classifying portion that classifies the question document contents on the basis of a classifying result of the classifying of the answer document contents so that, the first attribute information is added to a first set of the question document contents corresponding to the first set of the answer document contents and the second attribute information is added to a second set of the question document contents corresponding to the second set of the answer document contents.

An information processing apparatus, comprising:

• a determination content classifying portion that acquires question document contents and answer document contents, which are in a relationship of questions and answers, and determination contents each indicating a point of view for generating the answer from the question, and classifies the answer document contents in which a first one of the determination contents is set to a first set of the answer document contents and a second one of the determination contents different from the first one of the determination contents is set to a second set of the answer document contents;
• a feature amount calculating portion that calculates a first feature amount of each of the first set of the answer document contents and a second feature amount of each of the second set of the answer document contents;
• a distribution map generating portion that generates first distribution map information by plotting each of the first set of the answer document contents in a first feature amount space on the basis of the calculated first feature amount, and second distribution map information by plotting each of the second set of the answer document contents in a second feature amount space on the basis of the calculated second feature amount;
• a distance calculating portion that calculates first distances in the first feature amount space between the first set of the answer document contents on the basis of the calculated first feature amounts and second distances in the second feature amount space between the second set of the answer document contents on the basis of the calculated second feature amounts;
• a classifying portion that classifies the first set of the answer document contents so that, in a case that there is the first distance equal to or greater than a value set in advance between a first subset of the first set of the answer document contents and a second subset of the first set of the answer document contents, first attribute information is added to the first subset of the first set of the answer document contents and second attribute information different from the first attribute information is added to the second subset of the first set of the answer document contents, and classifies the second set of the answer document contents so that, in a case that there is the second distance equal to or greater than the value set in advance between a first subset of the second set of the answer document contents and a second subset of the second set of the answer document contents, third attribute information is added to the first subset of the second set of the answer document contents and fourth attribute information different from the third attribute information is added to the second subset of the second set of the answer document contents;
• and
• a determination content notifying portion that notifies one of the first and second determination contents, which reveals good classifying separation to the corresponding answer document contents, as a remarkable point for generating the answer from the question on the basis of classifying results of the classifying.

An information processing apparatus, comprising:

• a determination content classifying portion that acquires question document contents and answer document contents, which are in a relationship of questions and answers, and determination contents each indicating a point of view for generating the answer from the question, and classifies the answer document contents in which a first one of the determination contents is set to a first set of the answer document contents and a second one of the determination contents different from the first one of the determination contents is set to a second set of the answer document contents;
• a feature amount calculating portion that calculates a first feature amount of each of the first set of the answer document contents;
• a distribution map generating portion that generates first distribution map information by plotting each of the first set of the answer document contents in a first feature amount space on the basis of the calculated first feature amount;
• a classifying portion that classifies the first set of the answer document contents so that, in a case that there is the first distance equal to or greater than a value set in advance between a first subset of the first set of the answer document contents and a second subset of the first set of the answer document contents, first attribute information is added to the first subset of the first set of the answer document contents and second attribute information different from the first attribute information is added to the second subset of the first set of the answer document contents, wherein the classifying portion classifies a first set of the question document contents on the basis of a classifying result of the classifying of the first set of the answer document contents so that, the first attribute information is added to a first subset of the first set of the question document contents corresponding to the first subset of the first set of the answer document contents and the second attribute information is added to a second subset of the first set of the question document contents corresponding to the second subset of the first set of the answer document contents;
• and
• a attribute information adding unit that adds the first attribute information to the first subset of the first set of the question document contents;
• and the second attribute information to the second subset of the first set of the question document contents.