JP2014153601A - Device for evaluating difficulty degree of infant vocabulary comprehension, infant vocabulary retrieval device, and infant vocabulary classification device, and method and program for the devices - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a new index for evaluating a degree of difficulty of vocabulary comprehension in a vocabulary learning process of an infant.SOLUTION: The device for evaluating a difficulty degree of infant vocabulary comprehension includes: a logistic function approximation section; a learning day-age calculating section; and a learning period calculating section. The logistic function approximation section uses a learning time xand a learning ratio f(x) of vocabulary i for a vocabulary learning process of multiple infants as input in order to approximate a relationship of the learning time xand the learning ratio f(x) for each vocabulary i in a logistic curve. The learning day-age calculating section uses the logistic function to obtain an α% learning day-age and a β% learning day-age which is smaller than the α. The learning period calculating section uses the α% learning day-age and the β% learning day-age as input and calculates the learning period of the vocabulary i in order to output the learning period as the comprehension difficulty degree of the vocabulary i.

Description

  The present invention relates to an infant vocabulary comprehension difficulty evaluation device, an infant vocabulary search device, an infant vocabulary classification device, and methods and programs thereof for evaluating lexical comprehension difficulty in an infant vocabulary learning process.

  Traditionally, in the field of developmental psychology, vocabulary features have been captured by large-scale group data using vocabulary checklists in questionnaire surveys based on parents' answers. In this method, as shown in Non-Patent Document 1, what is the “50% reached age” of each vocabulary calculated from the cross-sectional data of the vocabulary checklist method (when the 50% children understand the vocabulary) I examined it and made it the age of understanding vocabulary.

  Choosing words that can be easily understood or spoken by all infants is important in creating easy-to-understand infant materials. On the other hand, selecting words that cannot be understood or spoken by all infants is also important in adjusting the difficulty of the problem. However, until now, the information that all infants can understand or speak immediately has not been used as a characteristic of the word.

  Until now, the age at which words are acquired has been presented in a table as shown in FIG. FIG. 1 shows the percentage of infants who understand each vocabulary. The first line indicates the age at which each vocabulary is understood, and the second and subsequent lines indicate the percentage of infants who understand each vocabulary.

  Conventionally, from this table, for example, the age understood by 50% of infants is obtained, and the difficulty level of words is determined.

Tamako Komine, Toru Watanamaki, “Standard Research on Vocabulary Development in Japanese Children: From the Japanese MacArthur Infant Language Development Questionnaire”, Development and Nursing Research Institute 2008. vol.24,3-42. Tetsuo Kobayashi, Yasuhiro Minami, Masaaki Nagata, “Identification of Age of Early Childhood Acquisition Using Longitudinal and Cross-sectional Data,” 18th Annual Conference of the Language Processing Society, P2-3, 2012.

  However, in the past, there was no idea to use the word selection to investigate whether the target word is a word that can be easily understood by all the infants or can be spoken immediately from this table. The reason is that it is difficult to obtain a cumulative word acquisition probability distribution for understanding for all words.

  In the questionnaire to find out the vocabulary to understand, if the number of vocabulary that an infant thinks understood increases, it becomes difficult to obtain accurate data for each vocabulary. In other words, since the number of vocabulary to be understood tends to increase rapidly, it becomes ambiguous whether or not each vocabulary is understood. Therefore, as shown in FIG. 1 with a pattern, for example, “Yes”, “Kani”, “Kaba”, “Kame”, and the vocabulary that the infant to understand does not reach 50% is data. It often occurs on the top.

  The present invention has been made in view of such problems, and it is possible to evaluate the difficulty of understanding infant vocabulary based on a new idea that uses the spread (slope) of the distribution that can be calculated from the cumulative vocabulary acquisition probability distribution. It is an object of the present invention to provide an infant vocabulary understanding difficulty evaluation device, an infant vocabulary search device, an infant vocabulary classification device, and methods and programs thereof.

The infant vocabulary comprehension difficulty evaluation device of the present invention includes a logistic function approximating unit, a learning age calculating unit, and a learning period calculating unit. Logistic function approximation unit as input a plurality of infants vocabulary learning learning timing vocabulary i in the process x _i and learning rate f (x _i), and learning time x _i for each vocabulary i and learning rate f (x _i) Is approximated by a logistic curve. The acquisition day age calculating unit obtains an α% acquisition day age and a β% acquisition day age of β% smaller than the α using a logistic function. The learning period calculation unit calculates the learning period of the vocabulary i by using α% learning age and β% learning age as input, and outputs the learning period as an understanding difficulty level of the vocabulary i.

The infant vocabulary retrieval apparatus of the present invention includes an infant vocabulary understanding difficulty evaluation device and a vocabulary retrieval unit. The infant vocabulary comprehension difficulty level evaluation device calculates the comprehension difficulty level of the vocabulary i from a logistic curve that models the cumulative understanding rate f (x _i ). The vocabulary search unit searches the vocabulary i that matches the search condition by comparing the degree of understanding difficulty with the search condition input from the outside.

  The infant vocabulary classification device of the present invention includes the above-described infant vocabulary understanding difficulty evaluation device and a vocabulary classification unit. The vocabulary classification unit inputs the understanding difficulty level calculated by the vocabulary understanding difficulty level evaluation device and the α% acquisition date, and configures the understanding difficulty level and the α% acquisition date as a vector corresponding to the vocabulary i. The vocabulary i is classified according to the level of difficulty of understanding and the value of α% acquisition date.

  According to the infant vocabulary comprehension difficulty evaluation device of the present invention, it is possible to provide a new index for evaluating the comprehension difficulty of infant vocabulary from the cumulative vocabulary acquisition probability distribution. In addition, the infant vocabulary retrieval apparatus according to the present invention makes it possible to retrieve an infant vocabulary according to the degree of understanding difficulty by using the new index. Moreover, the infant word classification device of the present invention makes it possible to classify the infant vocabulary according to the degree of understanding difficulty. Thus, each device of the present invention has an effect that contributes to the field of infant vocabulary research and education.

The figure which shows the data which recorded the ratio of the infant who understands a certain vocabulary for every age. The figure which shows the example of the relationship between the acquisition time x when the acquisition vocabulary i is "Yes", and the cumulative utterance ratio f (x). The figure which shows the example of the relationship between the acquisition time x when the acquisition vocabulary i is "Baba", and the cumulative utterance rate f (x). The figure which shows the function structural example of the infant vocabulary understanding difficulty level evaluation apparatus 100 of this invention. The figure which shows the operation | movement flow of the infant vocabulary understanding difficulty level evaluation apparatus. The figure which shows the function structural example of the infant vocabulary understanding difficulty level evaluation apparatus 200 of this invention. The figure which shows the function structural example of the infant vocabulary understanding difficulty evaluation apparatus 300 of this invention. The figure which shows the function structural example of the infant vocabulary understanding difficulty level evaluation apparatus 400 of this invention. The figure which shows the operation | movement flow of the infant vocabulary understanding difficulty level evaluation apparatus 400. FIG. The figure which shows the operation | movement flow of the utterance vocabulary logistic function approximation part 410. FIG. The figure which shows the more specific functional structural example of the learning age calculation part 420. FIG. The figure which shows the operation | movement flow of the learning age calculation part 420. FIG. The figure which shows the function structural example of the infant vocabulary search apparatus 500 of this invention. The figure which shows the operation | movement flow of the infant vocabulary search apparatus 500. FIG. The figure which shows the function structural example of the infant vocabulary classification apparatus 600 of this invention. The figure which shows the operation | movement flow of the infant vocabulary classification apparatus 600.

  Embodiments of the present invention will be described below with reference to the drawings. The same reference numerals are given to the same components in a plurality of drawings, and the description will not be repeated.

[Invention]
Prior to the description of the embodiments, a new concept for evaluating the difficulty of understanding the infant vocabulary according to the present invention will be described. According to the method of evaluating the difficulty level of understanding the infant vocabulary according to the present invention, a learning rate probability distribution consisting of a relationship between a learning period xi of a vocabulary _i and a learning ratio f (x _i ) in a vocabulary learning process of a plurality of infants is expressed by a logistic curve. It is a method of approximating and evaluating the difficulty of understanding from the spread of the probability density.

  FIGS. 2 and 3 show examples of modeling using the cumulative utterance probability as the learning probability distribution by the equation (1). In modeling the function, a least square method or a likelihood maximization method is used. In this case, since optimization of a non-linear function is necessary, an optimum parameter is obtained using a technique such as the steepest descent method. Both methods are well known and will not be described in detail.

An example in which the cumulative utterance rate f (x _i ) of the vocabulary is modeled by Equation (1) is shown in FIGS. FIG. 2 shows the cumulative utterance ratio of an infant when the acquired vocabulary is “Yes”. FIG. 3 shows the cumulative utterance ratio of an infant when the acquired vocabulary is “Baba”.

  The logistic curves shown in FIG. 2 and FIG. 3 are S-shaped curves, and show a change in which the increase amount becomes gentle again after the slope gradually increases after the gentle increase at first. The parameter c is the slope of the cumulative utterance ratio of the vocabulary, and b represents the offset amount of the logistic curve. a is a value that determines the upper limit of the cumulative utterance rate. As c increases, the slope of the central portion of the S-curve becomes steep, and as b increases, the S-curve moves to the larger acquisition time x.

  That is, if the slope of the central portion of the S-shaped curve is steep, it can be understood that the word is easy to learn and spoke at a short age. It is possible to evaluate the difficulty level of vocabulary acquisition by the size of the parameter c.

  In addition, it is difficult to obtain accurate data on the cumulative understanding ratio regarding the acquired vocabulary for the reasons described above. Therefore, in the present invention, attention is paid to the point that the utterance is performed after understanding the vocabulary. In other words, even if there are many vocabulary words on the data that do not reach 50% of infants who understand the vocabulary and then utter the vocabulary, the difficulty level of the vocabulary is accurately calculated. The device is made possible. Detailed description will be described later.

[Infant vocabulary comprehension difficulty evaluation device]
FIG. 4 shows a functional configuration example of the infant vocabulary comprehension difficulty evaluation device 100 of the present invention. The operation flow is shown in FIG. The infant vocabulary comprehension difficulty evaluation device 100 includes a logistic function approximation unit 110, a learning age calculation unit 120, a learning period calculation unit 130, and a control unit 140. The infant vocabulary comprehension difficulty evaluation device 100 is realized by, for example, reading a predetermined program into a computer composed of a ROM, a RAM, a CPU, and the like, and executing the program by the CPU.

Logistic function approximation unit 110 is input with multiple infants vocabulary learning learning timing vocabulary i in the process _{x i} and learning rate f _{(x i),} timing learning for each vocabulary i _{x i} a learning rate f _(x i) Is approximated by a logistic curve (step S110). The logistic curve is, for example, a function shown in the above equation (1).

  The learning day age calculation unit 120 uses the logistic curve approximated by the logistic function approximation unit 110 to obtain an α% learning day age and a β learning day age of β% smaller than α% (step S120). Here, for example, α% is 50%, and β% is 20%, for example. Any combination of α% and β% may be used as long as the relationship of α> β is maintained. Here, since data including 50% or less is included as data (Fig. 1), considering reliability, 20% where data exists in all vocabulary i and 50% which half of infants learn Using.

  In addition, the coefficient a in the formula (1) is a value between 0 and 1, and generally the value of the function in the formula (1) is less than this coefficient. Therefore, the value of 50% is the percentage indicated by the coefficient. It is more appropriate to set the value less than the value (for example, less than 70% when a = 0.7). Further, 20% is desirably set to 10% or more which does not become smaller than the lower inflection point of the logistic curve (S-shaped curve).

  The learning period calculation unit 130 calculates the learning period of the vocabulary i as (α−β) using the α% learning age and the β learning age as inputs, and outputs the learning period as an understanding difficulty level of the vocabulary i (step) S130). The operations in steps S110 to S130 are repeated until the processing for all the vocabulary i is completed (No in step S140). The control unit 140 controls this repetitive operation.

  According to the infant vocabulary comprehension difficulty evaluation device 100 shown in the first embodiment, a new index for evaluating the comprehension difficulty of an infant vocabulary can be provided. Note that there are two types of cumulative vocabulary acquisition probability distributions consisting of acquisition time and acquisition ratio input to the infant vocabulary understanding difficulty evaluation device 100. An infant vocabulary comprehension difficulty evaluation device that limits the cumulative vocabulary acquisition probability distribution to be used will now be described as a modification.

[Modification 1]
FIG. 6 shows a functional configuration example of the infant vocabulary comprehension difficulty evaluation device 200 of the present invention. The infant vocabulary comprehension difficulty evaluation device 200 uses the cumulative vocabulary acquisition probability distribution to be used as a ratio that an infant understands a word (infant word comprehension ratio). The infant vocabulary comprehension difficulty evaluation device 200 is obtained by replacing the logistic function approximating unit 110 of the infant vocabulary comprehension difficulty evaluating device 100 with a logistic function approximating unit 210, and a learning age calculating unit 120 and a learning period calculating unit 130. Are the same as is apparent from the reference signs.

Logistic function approximation unit 210, understanding the time _{x i} vocabulary i mastery time _{x i,} as learning rate f _{'(x i)} the cumulative understanding ratio f' _{(x i),} draw a logistic curve for each understood vocabulary i following Model the function of the expression,

Comprehension vocabulary parameters a _i ′, b _i ′, and c _i ′ are calculated.

[Modification 2]
FIG. 7 shows a functional configuration example of the infant vocabulary comprehension difficulty evaluation device 300 of the present invention. The infant vocabulary comprehension difficulty level evaluation apparatus 300 uses the cumulative vocabulary acquisition probability distribution to be used as the ratio of infant utterances (infant word utterance ratio). The infant vocabulary comprehension difficulty level evaluation device 300 is obtained by replacing only the logistic function approximation unit 110 of the infant vocabulary understanding difficulty level evaluation device 100 with a logistic function approximation unit 310.

Logistic function approximation unit 310, the utterance timing vocabulary i mastery timing _{x i x i,} as learning rate f _{(x i)} the cumulative speech rate f _{(x i),} the function to draw a logistic curve for each utterance vocabulary i Expression (1) is modeled, and utterance vocabulary parameters a _i , b _i , and c _i are calculated.

  Similar to the infant vocabulary understanding difficulty evaluation device 100, the infant vocabulary understanding difficulty evaluation devices 200 and 300 shown as modifications can provide a new index for evaluating the understanding difficulty of infant vocabulary. However, as mentioned above, it is difficult to investigate the vocabulary understood by young children. Then, the infant vocabulary comprehension difficulty evaluation apparatus 400 of the present invention devised so that the comprehension difficulty level can be calculated even when there are many words that the infant to understand does not reach 50% will be described as a second embodiment. .

FIG. 8 shows a functional configuration example of the infant vocabulary comprehension difficulty evaluation device 400 of the present invention. The infant vocabulary comprehension difficulty evaluation device 400 obtains the acquisition time x _i of the utterance vocabulary i and the cumulative utterance rate f (x _i ), the acquisition time xi of the understanding vocabulary _i and the cumulative understanding rate f ′ (x _i ). The input age is estimated as the input date. Here, since it is difficult to use the constraint of f (x _i ) ≦ f ′ (x _i ), the constraint of a _i ≦ a ′ _i is used instead.

[Cumulative utterance rate]
The cumulative utterance rate f (x _i ) is obtained by answering a questionnaire related to vocabulary described in Non-Patent Document 1 and grasping a vocabulary newly spoken by an infant. Specifically, about 100 mothers answer the above questionnaire every few months in a period from a little before the time when the infant starts to learn words to about 3 years old. Table 1 exemplifies the cumulative utterance rate f (x _i ) of an infant who utters the vocabulary “Ah”.

The data shown in Table 1 is obtained for all vocabulary spoken by the infant.

[Cumulative understanding ratio]
The cumulative understanding ratio f ′ (x _i ) is a ratio at which the infant understands the vocabulary, and is acquired in the same manner as the cumulative utterance ratio f (x _i ). The table exemplifies a cumulative understanding ratio f ′ (x _i ) of an infant who understands the vocabulary “Ah”.

The data shown in Table 2 is obtained for all vocabulary understood by young children. The vocabulary to understand and the vocabulary to utter are the same vocabulary i.

  The infant vocabulary understanding difficulty evaluation device 400 (FIG. 8) includes an utterance vocabulary logistic function approximation unit 410, an utterance vocabulary parameter storage unit 450, an understanding vocabulary logistic function approximation unit 460, an understanding vocabulary parameter storage unit 470, and a learning date. An age calculation unit 420, a learning period calculation unit 130, and a control unit 440 are provided. The operation flow is shown in FIG. The infant vocabulary comprehension difficulty level evaluation apparatus 400 is realized by reading a predetermined program into a computer including, for example, a ROM, a RAM, and a CPU, and executing the program by the CPU.

The utterance vocabulary logistic function approximation unit 410 receives the utterance time x _i (acquisition time) and the cumulative utterance rate f (x _i ) of the utterance vocabulary i in the vocabulary learning process of a plurality of infants, and generates a logistic curve for each utterance vocabulary i. The function of the expression (1) to be drawn is modeled, and the utterance vocabulary parameters a _i , b _i and c _i are calculated (step S410).

FIG. 10 shows step S410 in detail. The utterance vocabulary logistic function approximating unit 410 models the cumulative utterance ratio f (x _i ) using the equation (1) (step S4100). When modeling, the parameter a _i may be a _i > 1. Since the cumulative utterance rate f (x _i ) does not exceed 1, in this case (Yes in step S 4101), parameters b _i and c _i are recalculated with a _i = 1 (step S 4102). Speech vocabulary parameters _{a i,} b _{i, c} i is stored in the speech vocabulary parameter storage unit 450 (step S450).

The understanding vocabulary logistic function approximating unit 460 draws a logistic curve for each understanding vocabulary i, using as input the understanding time xi of the same vocabulary i as the utterance vocabulary _i and the cumulative understanding ratio f ′ (x _i ) in the vocabulary learning process. The function of the equation (2) is modeled, and the understanding vocabulary parameters a _i ', b _i ', c _i 'are calculated (step S460). The understanding vocabulary parameters a _i ′, b _i ′, and c _i ′ are stored in the understanding vocabulary parameter storage unit 470 (step S470).

The learning age calculation unit 420 receives both parameters stored in the utterance vocabulary parameter storage unit 450 and the understanding vocabulary parameter storage unit 470 as input, and based on the relationship between the understanding and the utterance, the understanding vocabulary reaches 50% of the understanding age Vocabulary 50% acquisition age and comprehension vocabulary 20% acquisition age are calculated (step S420). The calculation of the 50% learning age is f ′ (x _i ) = 0.5, and the calculation of the 20% learning age is f ′ (x _i ) = 0.2. For example, the Newton method is used. Alternatively, other optimization methods may be used.

  The learning period calculation unit 130 calculates the acquisition period of the vocabulary i using 50% acquisition age and 20% acquisition age as inputs, and outputs the acquisition period as an understanding difficulty level of the vocabulary i (step S130). The control unit 440 repeats the processing from step S410 to S130 until the processing is completed for all the vocabulary i (step S440).

  By operating as described above, the infant vocabulary understanding time estimation device 100 according to the present invention has 50 vocabulary of understanding vocabulary even if there is no day-age data to understand more than 50% in the questionnaire result data for examining the vocabulary to understand. % Acquisition date can be estimated.

[Learning age calculation part]
FIG. 11 shows a more specific functional configuration example of the learning age calculation unit 420. The operation flow is shown in FIG.

The learning age calculation unit 420 includes parameter comparison setting means 421 and learning age calculation means 422. The parameter comparison setting unit 421 receives the utterance vocabulary parameter a _i stored in the utterance vocabulary parameter storage unit 450 and the understanding vocabulary parameter a _i ′ stored in the understanding vocabulary parameter storage unit 470 as inputs, and the size of both parameters. Are compared (step S4211).

When a _i ′ <a _i (Yes in step S4211), recalculation of the understanding vocabulary logistic function approximation according to equation (2) is performed with a _i ′ = a _i (step S4212). If a _i ′ ≧ a _i and a _i ′> 1 (Yes in step S4213), recalculation of the understanding vocabulary logistic function approximation by equation (2) is performed with a _i ′ = 1 (step S4214). The parameter comparison setting means 421 outputs the comparison result of the two parameters a _i and a _i ′ and the understanding vocabulary parameters a _i ′, b _i ′ and c _i ′ re-approximate the logistic function to the acquired age calculation means 422. .

The learning age calculation means 422 is stored in the comprehension vocabulary parameter storage unit 470, the comparison result of both parameters a _i , a _i ′, the recalculated comprehension vocabulary parameters a _i ′, b _i ′, c _i ′. With the understanding vocabulary parameters a _i ′, b _i ′, and c _i ′ as inputs, the re-approximate understanding vocabulary parameters a _i ′, b _i ′, c _i ′ or the understanding vocabulary parameter storage 470 A 50% learning age and a 20% learning age are calculated using any one of the understanding vocabulary parameters a _i ′, b _i ′, and c _i ′ (step S422).

The learning age calculation means 422 determines that the understanding vocabulary parameter a _i ′ is smaller than the utterance vocabulary parameter a _i (Yes in step S4221) or the understanding vocabulary parameter a _i ′ is greater than 1 (Yes in step S4222). Then, using the comprehension vocabulary parameters a _i ′, b _i ′, c _i ′ re-approximate by the parameter comparison setting means 421, the 50% learning age and the 20% learning age are calculated (step S4223).

When the understanding vocabulary parameter a _i ′ is not less than the size of the utterance vocabulary parameter a _i and not more than a _i ′ (No in step S4222), the understanding vocabulary parameter a _i ′, b stored in the understanding vocabulary parameter storage unit 470 is used. _A 50% acquisition day and a 20% acquisition day are calculated using _i ′ and c _i ′ (step S4224). By operating in this way, it is possible to estimate the age at which 50% of the understanding vocabulary is acquired even if there is no age data to be understood that exceeds 50%.

[Specific examples of learning age]
Table 3 shows the results of obtaining the 50% acquisition date and the 20% acquisition date with the infant vocabulary understanding difficulty evaluation device 400 described above. The data was a check of the vocabulary spoken by the infant using the words and gestures (448 words) and words and grammar (711 words) of the Macarthur Infant Language Development Questionnaire, which was answered by 1699 people. Was used.

  Table 3 shows 50% acquisition age, 20% acquisition age, and acquisition period of the utterance acquisition date and the understanding acquisition date for the infant vocabulary classified as social languages. Examining words with an acquisition period of 80 days or less, although not shown in Table 3, “Thank you”, “Thank you”, “Feast”, “No”, “Give me”, “Please”, “Yes”, “ This includes words such as “Baibai”. These words are words that all infants acquire simultaneously.

  Examining vocabulary with a learning period of 130 days or more, the words such as “Ah”, “Garger”, “Kokekoko”, etc. as shown in Table 3 apply. These words require 130 days or more after 20% of infants learn until 50% of them learn. Thus, the infant vocabulary comprehension difficulty evaluation device of the present invention can provide a new index for quantitatively analyzing an infant vocabulary.

[Application Example 1]
By using the infant vocabulary comprehension difficulty evaluation device 100, 200, 300, 400 of the present invention, an infant vocabulary retrieval device that retrieves infant vocabulary according to the retrieval condition can be configured. FIG. 13 shows a functional configuration example of the infant vocabulary retrieval apparatus 500 of the present invention. The operation flow is shown in FIG.

  The infant vocabulary search device 500 includes an infant vocabulary understanding difficulty evaluation device 200, a vocabulary search unit 510, and a control unit 540. The infant vocabulary comprehension difficulty evaluation device 200 is as described above (FIG. 6). The vocabulary search unit 510 receives the understanding difficulty level of the vocabulary i output from the infant vocabulary understanding difficulty level evaluation device 200 and the search condition input from the outside, and compares the understanding difficulty level and the search condition for each vocabulary i. Thus, the vocabulary i that matches the search condition is searched (step S130).

  In the search condition, a learning period and a learning age are set. For example, if the learning period is set to 130 or more, the words “Ah”, “Garger”, and “Kokekokko” shown in Table 3 above are output as searched vocabularies. In addition, although it demonstrated by the example combined with the infant vocabulary comprehension difficulty evaluation apparatus 200, you may use any of above-mentioned apparatuses for an infant vocabulary comprehension difficulty evaluation apparatus.

[Application 2]
As an infant vocabulary search device, utterance vocabulary parameters and comprehension vocabulary parameters are stored as a database in association with vocabulary i for a number of infant words, accept arbitrary search conditions from the user, and output words that match the search conditions It is also conceivable to use a device.

[Application Example 3]
FIG. 15 shows a functional configuration example of the infant vocabulary classification apparatus 600 of the present invention. The operation flow is shown in FIG. The infant vocabulary classification device 600 includes the infant vocabulary understanding difficulty evaluation device 100, a vocabulary classification unit 610, and a control unit 640.

  The infant vocabulary comprehension difficulty evaluation device 100 is as described above (FIG. 4). The vocabulary classification unit 610 is configured with the learning period calculated by the infant vocabulary comprehension difficulty evaluation device 100 and the α% acquisition date as input, and the acquisition period and α% acquisition date as a vector corresponding to the vocabulary i. Then, the vocabulary i is classified according to the value of the acquisition period and the α% acquisition age (step S610).

  If α% is 50%, for example, α means the average of the Gaussian distribution. The acquisition period is a value corresponding to the variance (σ) of the Gaussian distribution. Therefore, the infant vocabulary classification apparatus 600 can also classify the infant vocabulary into several Gaussian distributions.

  Various existing methods can be used as the clustering method for classification. Since the infant vocabulary classification apparatus 600 can classify the infant vocabulary into several clusters, the learning words can be grouped according to the development of the infant.

  In addition, such a clustering method may handle both speech and understanding parameters at the same time. By doing so, it is also possible to classify word clusters considering both utterance and understanding at the same time.

  When the processing means in the above apparatus is realized by a computer, the processing contents of the functions that each apparatus should have are described by a program. Then, by executing this program on the computer, the processing means in each apparatus is realized on the computer.

  The program describing the processing contents can be recorded on a computer-readable recording medium. As the computer-readable recording medium, for example, any recording medium such as a magnetic recording device, an optical disk, a magneto-optical recording medium, and a semiconductor memory may be used. Specifically, for example, as a magnetic recording device, a hard disk device, a flexible disk, a magnetic tape or the like, and as an optical disk, a DVD (Digital Versatile Disc), a DVD-RAM (Random Access Memory), a CD-ROM (Compact Disc Read Only) Memory), CD-R (Recordable) / RW (ReWritable), etc., magneto-optical recording media, MO (Magneto Optical disc), etc., semiconductor memory, EEP-ROM (Electronically Erasable and Programmable-Read Only Memory), etc. Can be used.

  This program is distributed by selling, transferring, or lending a portable recording medium such as a DVD or CD-ROM in which the program is recorded. Further, the program may be distributed by storing the program in a recording device of a server computer and transferring the program from the server computer to another computer via a network.

  Each means may be configured by executing a predetermined program on a computer, or at least a part of these processing contents may be realized by hardware.

  The present invention can be used in the field of customized education for young children.

Claims (11)

Logistic as input a plurality of infants vocabulary learning acquisition and learning time x _i vocabulary i ratio in the process f (x _i), the relationship between the learning time for each vocabulary i x _i and the learning rate f (x _i) A logistic function approximator approximating with a curve;
Using the above-mentioned logistic function, an acquisition day age calculating part for obtaining an α% acquisition day age and a β% acquisition day age of β% smaller than the α%,
A learning period calculation unit that calculates the learning period of the vocabulary i using the α% learning date and the β% learning date as inputs, and outputs the learning period as an understanding difficulty level of the vocabulary i;
An infant vocabulary comprehension difficulty evaluation device. In the infant vocabulary comprehension difficulty evaluation apparatus according to claim 1,
The logistic function approximation part is
The learning time _{x i} is the vocabulary i understood timing _{x i} of, as the learning rate f _{'(x i)} is the cumulative understood fraction f' _{(x i),} a function of the following equation to draw a logistic curve for each understood vocabulary i Model

An infant vocabulary comprehension difficulty evaluation device, which is a logistic function approximating unit for calculating comprehension vocabulary parameters a _i ′, b _i ′, and c _i ′. In the infant vocabulary comprehension difficulty evaluation apparatus according to claim 1,
The logistic function approximation part is
The learning time x _i is the utterance time x _{i of} the vocabulary i, the learning rate f (x _i ) is the cumulative utterance rate f ′ (x _i ), and a function of the following equation is drawn for each utterance vocabulary i: Model,

An infant vocabulary comprehension difficulty evaluation device, which is a logistic function approximating unit for calculating utterance vocabulary parameters a _i , b _i and c _i . In the infant vocabulary comprehension difficulty evaluation apparatus according to claim 1,
The logistic function approximation part is
The learning time _{x i} is the vocabulary i understood timing _{x i} of, as the learning rate f _{'(x i)} is the cumulative understood fraction f' _{(x i),} a function of the following equation to draw a logistic curve for each understood vocabulary i Model

Understanding vocabulary logistic function approximation means for calculating understanding vocabulary parameters a _i ′, b _i ′, c _i ′;
Understanding vocabulary parameter storage means for storing the above understanding vocabulary parameters a _i ′, b _i ′, c _i ′;
The learning time x _i is the utterance time x _{i of} the vocabulary i, the learning rate f (x _i ) is the cumulative utterance rate f ′ (x _i ), and a function of the following equation is drawn for each utterance vocabulary i: Model,

Logistic function approximation means for calculating utterance vocabulary parameters a _i , b _i , c _i ;
Utterance vocabulary parameter storage means for storing the utterance vocabulary parameters a _i , b _i and c _i ;
With
The acquisition period calculation unit
Based on the result of comparing the size of the understanding vocabulary parameter a _i ′ and the utterance vocabulary parameter a _{i using} both parameters stored in the understanding vocabulary parameter storage unit and the utterance vocabulary parameter storage unit as input Calculate α% acquisition day and β% acquisition day using either vocabulary parameters a _i ′, b _i ′, c _i ′ or stored understanding vocabulary parameters a _i ′, b _i ′, c _i ′ What to do,
Infant vocabulary comprehension difficulty evaluation device characterized by being. An infant vocabulary comprehension difficulty evaluation device according to claim 2;
A vocabulary search unit that searches the vocabulary i that matches the search condition by comparing the comprehension difficulty of the vocabulary i output from the vocabulary understanding difficulty evaluation device with a search condition input from the outside;
An infant vocabulary search device comprising: The infant vocabulary comprehension difficulty evaluation device according to claim 1,
The learning period and the α% acquisition date calculated by the vocabulary understanding difficulty evaluation device are input, and the acquisition period and the α% acquisition date are configured as a vector corresponding to the vocabulary i, A vocabulary classification unit that classifies the vocabulary i according to the level of difficulty of understanding and the value of the α% acquisition date;
An infant vocabulary classification device comprising: Logistic as input a plurality of infants vocabulary learning acquisition and learning time x _i vocabulary i ratio in the process f (x _i), the relationship between the learning time for each vocabulary i x _i and the learning rate f (x _i) Logistic function approximation process that approximates with a curve,
Using the above-mentioned logistic function, an acquisition date calculation process for obtaining an α% acquisition day age and a β% acquisition day age of β% smaller than the α%,
The learning period calculation process of calculating the learning period of the vocabulary i using the α% learning date and the β% learning date as input, and outputting the learning period as an understanding difficulty level of the vocabulary i;
Infant vocabulary comprehension difficulty evaluation method. In the infant vocabulary comprehension difficulty evaluation method according to claim 7,
The logistic function approximation process is
The learning time _{x i} is the vocabulary i understood timing _{x i} of, as the learning rate f _{'(x i)} is the cumulative understood fraction f' _{(x i),} a function of the following equation to draw a logistic curve for each understood vocabulary i Model

An infant vocabulary comprehension difficulty evaluation method, which is a logistic function approximation process for calculating comprehension vocabulary parameters a _i ′, b _i ′, and c _i ′. A method for evaluating the difficulty level of understanding the infant vocabulary according to claim 8,
A vocabulary search process for searching for the vocabulary i that matches the search condition by comparing the understanding difficulty level of the vocabulary i output from the vocabulary understanding difficulty evaluation device with a search condition input from the outside;
An infant vocabulary search method comprising: The infant vocabulary comprehension difficulty evaluation method according to claim 7,
The learning period calculated by the vocabulary understanding difficulty evaluation method and the α% acquisition date are input, and the acquisition period and the α% acquisition date are configured as a vector corresponding to the vocabulary i. A vocabulary classification process for classifying the vocabulary i according to the degree of understanding difficulty and the value of the α% acquisition date;
An infant vocabulary classification method comprising:   To operate a computer as the infant vocabulary comprehension difficulty evaluation device according to any one of claims 1 to 4, the infant vocabulary search device according to claim 5, or the infant vocabulary classification device according to claim 6. Program.

Images