Abstract:
A response generation apparatus ensures accurate output. A computer stores graph knowledge including a response generation module generating a response to an input document including a plurality of sentences, the graph knowledge database includes graph data that manages a structure of each type of graph knowledge, and the response generation module generates a first graph knowledge from each of the sentences; searches a second graph knowledge similar to each of the plurality of first graph knowledge while referring to the graph data on the basis of the plurality of first graph knowledge; identifies the plurality of second graph knowledge included in a dense location where a density of the second graph knowledge is high in a graph space; searches third graph knowledge for generating the response while referring to the graph data on the basis of the identified second graph knowledge; and generates the response using the third graph knowledge.
Abstract:
A computer system includes a storage device and a processor configured to operate in accordance with command codes stored in the storage device. The processor places a plurality of feature vectors including feature vectors each provided with a label in a feature space. The processor determines forces among the plurality of feature vectors determined from repulsions and attractions between feature vectors. A repulsion is defined to be larger when a distance between the feature vectors is shorter, and an attraction is defined to be larger when the feature vectors have a predetermined relationship. The processor rearranges the plurality of feature vectors to reduce potential energy of the feature space determined by the forces among the plurality of feature vectors. The processor outputs at least a part of the plurality of rearranged feature vectors as data to be used for data analysis.
Abstract:
Sentence polarity determination is used to assess whether a sentence is an affirmative expression or a negative expression, and is applied for reputation analysis, etc. Polarity determination determines whether an input sentence is affirmative or negative. When some subject is being talked about, it is sometimes desired to determine whether what is being referred to in the sentence is affirmative or negative, rather than the polarity of the sentence per se. The present invention provides a method for determining the polarity of the sentence by applying a recursive polarity rule based on a dependency structure of the sentence, taking into consideration the portion of the sentence that is being referred to. Use of a recursive rule makes it possible to prevent the number of rules from becoming huge, and thereby to perform efficient polarity determination in terms of memory amount and calculation amount. The length of the dependency needed for polarity determination can also be efficiently controlled.
Abstract:
Provided is a system which controls information presented to a user referring to contents including a plurality of elements, outputs the contents to an output device, obtains bioinformation of the user at a time when each of the elements included in the contents is output to the output device, calculates an understanding level of the user with respect to each of the elements included in the contents, based on each piece of the obtained bioinformation, determines the understanding type based on the calculated understanding level and a difficulty level indicated by the difficulty level information of the elements of which the understanding level is equal to or lower than a first threshold value, in a case where it is determined that the user does not understand the contents based on the calculated understanding level, and outputs the presented information which corresponds to the determined understanding type to the output device.
Abstract:
A system for evaluating actions that a plurality of target objects performs to achieve communication, the system comprising at least one computer, the at least one computer is configured to: obtain a plurality of signals used for calculating a plurality of state values for evaluating an internal state of a target object in a case where the target object performed the action, from the plurality of target objects; calculate a state feature amount made up of the plurality of state values using the plurality of signals; and generate display information on the basis of the state feature amount, the display information for displaying a positional relationship of the internal state of each of the plurality of target objects in an evaluation space for visually illustrating a similarity of the internal state of the each of the plurality of target objects.
Abstract:
A computer system includes a storage device and a processor configured to operate in accordance with command codes stored in the storage device. The processor places a plurality of feature vectors including feature vectors each provided with a label in a feature space. The processor determines forces among the plurality of feature vectors determined from repulsions and attractions between feature vectors. A repulsion is defined to be larger when a distance between the feature vectors is shorter, and an attraction is defined to be larger when the feature vectors have a predetermined relationship. The processor rearranges the plurality of feature vectors to reduce potential energy of the feature space determined by the forces among the plurality of feature vectors. The processor outputs at least a part of the plurality of rearranged feature vectors as data to be used for data analysis.
Abstract:
In order to provide an interactive interface that arouses a user's sympathy and realizes a continuous interaction having much information, an interactive system of providing the interactive interface of performing an interaction with a user includes a computer and a measurement device of measuring a signal related to a user's speech. The computer calculates a first feature amount indicating feature of a user action in the speech of the user, based on the signal measured by the measurement device, calculates a second feature amount that is a control value of a response action to the user performed by the interactive interface, based on the first feature amount, calculates a control time of adjusting a timing of controlling the response action depending on the type of the response action, and controls the interactive interface, based on the second feature amount and the control time.
Abstract:
An information processing system includes a storing unit that stores evaluation data in which a plurality of subjects are associated with a plurality of evaluation expressions respectively and evaluation expression relationship data indicating relationships between the evaluation expressions, a question generating unit that generates and outputs questions on the basis of the evaluation data and the evaluation expression relationship data, and a matching unit that outputs information relating to the subject included in the evaluation data on the basis of responses when the responses with respect to the questions are input.
Abstract:
There is provided a system in which empowerment is performed by outputting conversation information to the user, the system including: a computer including a processor, a memory, and an interface; and a measuring device that measures signals of a plurality of types, wherein the processor calculates values of conversation parameters of a plurality of attributes for evaluating a state of a user who performs the empowerment on the basis of a plurality of signals measured by the measuring device, the processor selects a selection parameter which is a conversation parameter of a change target on the basis of the values of the conversation parameters of the plurality of attributes, the processor decides conversation information for changing a value of the selection parameter, and the processor outputs the decided conversation information to the user.
Abstract:
The present invention solves a problem that there may be a case that an estimated value of regression cannot be calculated depending on a discrimination result when a regression method is applied after a discrimination method, and has a purpose to obtain an estimating equation with high accuracy even when the number of sample groups to which the regression method is applied is small.An estimating equation that satisfies the regression and discrimination at the same time can be obtained by combining a discrimination evaluation function that evaluates discrimination accuracy and a regression evaluation function that evaluates regression accuracy, calculating a combination evaluation function, and optimizing the combination evaluation function.