CN115630155B

CN115630155B - A multi-dimensional grading method and system for Chinese children's reading materials

Info

Publication number: CN115630155B
Application number: CN202211100230.6A
Authority: CN
Inventors: 袁曦临; 章敏
Original assignee: Southeast University
Current assignee: Southeast University
Priority date: 2022-09-08
Filing date: 2022-09-08
Publication date: 2025-06-17
Anticipated expiration: 2042-09-08
Also published as: CN115630155A

Abstract

The invention discloses a multidimensional grading method and system for Chinese children books, which comprehensively evaluates the reading difficulty of the Chinese children books from the perspective of cognition and emotion, carries out Tong Shu grading, generates grading codes comprising three dimensions of text cognition dimension, text emotion dimension and text function dimension, consists of a plurality of indexes, and forms multidimensional grading codes by a mathematical calculation method and classification judgment, and is used for measuring the language reading difficulty and emotion acceptance difficulty of the Chinese children books. And according to the difficulty result ranking, providing a plurality of hierarchical selection modes under the cognition guidance, emotion guidance and cognition emotion comprehensive guidance. The invention takes a classification method of the surface assembly as a guide, measures and classifies the reading and cognition difficulty and the text emotion complexity of books, and establishes a Chinese Tong Shuduo-dimensional classification system which accords with Chinese language characteristics and the local social culture environment.

Description

Multi-dimensional grading method and system for Chinese children's books

Technical Field

The invention belongs to the technical field of book grading evaluation, and particularly relates to a multi-dimensional grading system and method for Chinese children books.

Background

The development of children book publishing and digital media creates urgent needs for children ladder reading and grading reading guidance, and promotes the practical needs of the children reading grading evaluation and children book grading measurement methods. The existing foreign hierarchical reading systems such as Lexile hierarchical reading evaluation system, A-Z hierarchical reading system and the like are difficult to adapt to Chinese children books due to language differences, social system differences and cultural background differences, the hierarchical system and the hierarchical method cannot be simply transplanted, and the education policy of China requires to establish a Chinese ladder reading system which accords with the characteristics of Chinese children. The "obligation education Chinese course standard (2011 edition)" requires to design course targets from three aspects of knowledge and capability, process and method, emotion attitude and value, realizes the unification of tooliness and humanity, puts forward the overall requirements for children reading including the multi-aspect targets of children reading capability development, emotion education, moral education and the like, and emphasizes the effect of reading on the balanced development of children emotion and intelligence capability.

The existing grading standard system mainly belongs to 2 major categories, namely, objectively divides according to the difficulty of reading text content, does not consider the difference of emotion needs and aesthetic interests of children, and mainly depends on education, learning and subjective experience of reading specialists, and takes the age or grade of the children as a main basis of grading. The children have different growth and development, socialization degrees and cultural environments, so that the reading capability and reading interest among children are greatly different, the two types of grading reading standards mainly take age and grade as standards, language difficulty as main grading basis, single dimension and simple dividing mode, and the consideration of the emotion development requirement and the independent reading difference of the children is lacking. The problem of how to realize quantitative evaluation and system autonomous evaluation by Tong Shu classification is not solved due to the complexity of the main body related to children education and the diversity of the market demand scenes of children reading, and the recommendation form mainly judged by subjective experience of experts is still adopted at present.

Disclosure of Invention

The invention aims to solve the technical problems that a Chinese children book grading system is constructed from the multi-dimensions of text difficulty, subject type, text emotion and the like, a classification method of face-to-face assembly is used as a guide, and a Chinese Tong Shuduo-dimensional grading system which accords with Chinese language characteristics and social culture environment is established aiming at the evaluation and grading of book reading and cognition difficulty and text emotion complexity.

In order to solve the technical problems, the invention provides a multi-dimensional grading system and a multi-dimensional grading method for Chinese children books, which take text cognition, text emotion and text functions as guidance and quantify the reading difficulty of the Chinese children books in a multi-dimensional manner.

Firstly, the invention provides a multi-dimensional grading method for Chinese children's books, which comprises the following steps:

S1, defining an index system from three dimensions of text cognition, text emotion and text function, and determining an operation index under each dimension according to the grading dimension of the index system, wherein the operation index comprises language cognition difficulty, genre, image-text relationship, emotion polarity and emotion richness under the text emotion dimension and reading requirement type under the text function dimension;

s2, constructing and processing word lists, namely constructing a hierarchical word list, a hierarchical word list and a multidimensional emotion dictionary according to requirements;

S3, text processing, namely preprocessing, word segmentation processing and positive emotion and negative emotion statistical processing are carried out on the Chinese children books according to purposes and requirements;

S4, matching the text with the word list, namely matching the text subjected to word division processing with the hierarchical word list, and counting the number of Chinese characters of each level; matching the text after word segmentation with a hierarchical vocabulary and a multidimensional emotion dictionary respectively, and counting the number of words in each level and emotion word frequency in each dimension;

S5, judging classification types, namely judging the types of the bodies, the image-text relations and the reading requirement types of books according to classification rules, and determining classification identifiers of the books;

and S6, integrating the dimension indexes by adopting a segmentation marking method according to the principle of the classification method of the face-based assembly, namely compiling the calculation result of each index and the classification mark into a multi-dimensional classification code, and obtaining the Chinese children book difficulty score and category under each dimension.

In step S2, the construction of the grading word list and the grading word list is derived from the processing of the 'Chinese horizontal vocabulary and Chinese character grade outline (revision)' set by the examination center of the national Chinese horizontal examination Committee, wherein the processing of the word list comprises the steps of keeping the first grade word and the second grade word unchanged, merging the third grade word and the third grade word annex into the third grade word, merging the third grade word and the fourth grade annex into the third grade word, adding Chinese characters outside the fifth grade word recording word list, processing of the word list is to remove single syllable words in the word list, and word grades of the first, second, third and fourth grade words correspond to word difficulty grade coefficients cl, vl and cl are word difficulty grade coefficients respectively.

In step S2, the multi-dimensional emotion dictionary is constructed from a Chinese emotion vocabulary ontology library of university of great company, the subclasses are re-aggregated into the major classes, and major class identifications are correspondingly given, so that emotion words with auxiliary emotion classification are repeatedly classified into different emotion classifications.

On the other hand, the invention also provides a multi-dimensional grading system for the Chinese children's book, which comprises:

the index construction module is used for defining an index system from three dimensions of text cognition, text emotion and text function, and determining an operation index under each dimension according to the grading dimension of the index system, wherein the operation index comprises language cognition difficulty, genre, image-text relationship, emotion polarity and emotion richness under the text cognition dimension and reading requirement type under the text function dimension;

The vocabulary construction and processing module is used for constructing a hierarchical vocabulary, a hierarchical vocabulary and a multidimensional emotion dictionary according to the needs;

The text processing module is used for carrying out pretreatment, word segmentation and positive emotion and negative emotion statistical processing on the Chinese children books according to the purposes and the requirements;

the text and word list matching module is used for matching the text after word segmentation processing with the hierarchical word list, counting the number of Chinese characters of each level, simultaneously matching the text after word segmentation processing with the hierarchical word list, matching the text with a multidimensional emotion dictionary, counting the number of words of each level and the emotion word frequency of each dimension, and further obtaining an operation index of the book corresponding to each dimension;

The classification type judging module is used for judging the genre, the image-text relationship and the reading demand type of the book according to the classification rule and determining the classification identification of the book;

And the grading module integrates the dimension indexes by adopting a segmentation marking method according to the principle of a classification method of the face assembly, namely, the calculation result of each index and the classification mark are compiled into a multi-dimensional grading code, and the Chinese children book difficulty score and category under each dimension are obtained.

The invention adopts the technical proposal, and has the following beneficial effects compared with the prior art:

(1) The method comprises the steps of constructing a Chinese children book multidimensional grading index system based on the understanding and emotion considering visual angles, wherein the primary index content is divided into 3 dimensions, namely, the method comprises the steps of paying attention to language understanding difficulty of the traditional grading system, paying attention to the effect of reading on the balanced development of emotion and intelligence capability of children and the situation function of reading, and grading dimensions are diversified compared with the traditional grading standard and system.

(2) Based on Chinese characteristics and the reading requirements of localized children, the classification system is more refined and reasonable by adopting a classification and classification mode, and the multi-layer multi-dimensional classification of the Chinese children's books is realized, wherein the types of the text are used for correcting the prejudice effect caused by the types of the books, and the graph-text relationship highlights the important effect of the images on the understanding of the meaning of the books.

(3) The grading system does not depend on subjective evaluation and guidance recommendation of experts, the selection of grading indexes has quantitative operability, and the construction of grading word lists is based on 'Chinese horizontal vocabulary and Chinese character grade outline (revision)' with general meaning and standardization, and is based on independent free reading of children.

(4) The hierarchical system provides multiple hierarchical ordering and selection modes under the conditions of cognition guidance, emotion guidance and cognition emotion comprehensive guidance, can meet the reading selection requirements of different target guidance, and realizes quantitative evaluation and system autonomous evaluation.

Drawings

FIG. 1 is a schematic flow chart of the method of the present invention.

FIG. 2 is a schematic diagram of the hierarchical index system of the present invention.

Fig. 3 is a schematic diagram of a multi-dimensional hierarchical code structure and sample based on a facet-based partitioning classification method according to the present invention.

Detailed Description

The following describes the embodiments of the present invention in further detail with reference to the accompanying drawings.

The invention provides a multi-dimensional grading method for Chinese children books, which comprises the following steps:

(1) Defining an index system:

The index system is defined as the definition of related information and concept of the index system, and the index system information records the overall description information and creation information of the index system, including the determination of grading purpose, the definition of grading dimension and the definition of basic information of the index system.

The design principle of the index system is emphasized in the principle of child home position, the principle of individual difference, the principle of child education, the principle of reading cognition and emotion experience, the principle of autonomous reading and teaching, and the principle of scene reality and functional diversity application.

The method is characterized by comprising the steps of carrying out book classification around the reading difficulty of the multi-dimensional comprehensive evaluation Chinese children books, helping children to read and select and constructing a ladder reading system which accords with the characteristics of Chinese children, and constructing a Chinese children book classification index system according to the related design principle of the invention, wherein the evaluation dimension comprises 3 aspects of text cognition, text emotion and text function. The index system design process is shown in figure 1.

(2) Constructing grading indexes:

The invention takes the grading dimension as the grading index of the traction selection scheme layer, and comprises 3 aspects of language recognition difficulty, text emotion complexity, function type and the like of Chinese children books.

According to the hierarchical dimension of the index system, determining operation indexes in each dimension, including language cognition difficulty, genre, image-text relationship in cognition dimension, emotion polarity and emotion richness in emotion dimension, and reading requirement type in text function dimension, wherein the specific index system is shown in figure 2.

(3) Designing a hierarchical expression mode:

The classification method of the facet group is a literature classification method compiled according to the analysis and comprehensive principles of concepts, and the facet class table consists of a plurality of groups of facets. A group face is a set of categories generated by dividing a subject area by a single series of classification criteria, i.e., a set of simple concepts that represent attributes of an aspect of a class of things. Each set of facets may also be divided into multiple sub-facets using the same series of finer criteria. The "colon taxonomies" Colon Classification created by the indian musician Ruan Gangna is the most well known facet group classification method, its main body is basic class table and facet class table, and the segmentation mark system is adopted, i.e. the class number is formed from several segments with independent meaning, and it can express not only a theme concept, but also each group of facets and theme factors constituting the theme concept in the form of segmentation. It is called "colon-out Classification" because it adopts the composition symbol ":".

In the embodiment, 6 sub-surface grading indexes of text cognition, text emotion, language cognition difficulty under 3-dimensional sub-surfaces of text function, text type, image-text relationship, emotion polarity, emotion richness, reading requirement type and the like are integrated through a sub-surface assembly classification method, each index calculation result and classification mark are compiled into a multi-dimensional grading code, each sub-surface is connected by a 'connection' to represent a parallel relationship, and the sub-surfaces are connected by a 'connection'. The specific construction mode of the multidimensional hierarchical code is shown in fig. 3.

(4) Vocabulary construction and processing:

in the embodiment, the construction of the hierarchical word list and the hierarchical word list is derived from the treatment of the ' Chinese horizontal vocabulary and Chinese character class outline ', which are formulated in 1992 by the office examination center of the national Chinese horizontal examination Committee, and are used as standard of Chinese language skills and levels, and the programming principle is that ' high-frequency words are selected, and words with wide distribution and high use degree are simultaneously selected at the same time. The Chinese horizontal vocabulary and Chinese character class outline record 4 class Chinese vocabularies of A, B, C and D. The first and second vocabulary are common words, and the third Ding Liangji vocabulary is difficult words. With the development of society and the evolution of language, the outline of 1994 was revised for 5 years. Finally, outline (revision) records 2905 Chinese characters commonly used and 8822 words commonly used, wherein:

(1) 800 first-level words (800 Chinese characters are most commonly used), 804 second-level words, 601 third-level words and 700 third-level words;

(2) Class a 1033, class b 2028, class c 2022, class t 3569.

In the embodiment, the processing of the word list comprises the steps of keeping the first level word and the second level word unchanged, merging the third level word and the third level word annex into the third level word, adding Chinese characters outside the fifth level word recording word list, and processing the word list to remove monosyllabic words in the word list, wherein word grades of the first, second, third, fourth and fifth words correspond to word difficulty coefficients cl and vl respectively. cl is the Chinese character difficulty level coefficient, vl is the vocabulary difficulty level coefficient.

In the embodiment, the construction of the multidimensional emotion dictionary is derived from a Chinese emotion vocabulary ontology library of university of great company, the subclasses are recombined into the major classes, major class marks are correspondingly given, and emotion words with auxiliary emotion classification are repeatedly classified into different emotion classifications.

The university of the great company Chinese emotion vocabulary ontology library is a Chinese emotion classification ontology resource which is arranged and marked by the university of the great company information retrieval research laboratory under the guidance of Lin Hongfei professor, and is constructed on the basis of a psychologist Ekman emotion classification system, and finally emotion in the vocabulary ontology is totally divided into 7 major categories and 21 minor categories. The resource describes a Chinese word or phrase from different angles, including information such as word part of speech category, emotion strength and polarity. The Chinese emotion vocabulary ontology can be used for solving the problem of multi-category emotion classification and also can be used for solving the problem of general tendency analysis.

(5) Text processing:

The text processing comprises text preprocessing, word segmentation processing, positive emotion and negative emotion statistical processing, wherein the text preprocessing is to delete the content such as the preamble, copyright information, author information and postscript of a Chinese children book, the main text content of the book is reserved as a subsequent operation object, the word segmentation processing is to cut the preprocessed text into a set of Chinese character blocks by using a tool, the word segmentation processing is to cut the preprocessed text into a set of word blocks by using the tool, and the positive emotion and negative emotion statistical processing is to perform two-dimensional emotion analysis on the preprocessed text by using a tool platform.

(6) Matching the text with the word list;

The text after word segmentation is matched with a hierarchical word list through a python program code, the number of Chinese characters of each level is counted, the text after word segmentation is matched with the hierarchical word list through the python program code, the number of words of each level is counted, and the text after word segmentation is matched with a multidimensional emotion dictionary through the python program code, so that emotion word frequency of each dimension is obtained.

(7) And judging the classification category, and determining the classification identification of the book by judging the genre, the graph-text relationship and the reading requirement type of the Chinese children book according to the classification rule.

The types of the literary composition are classified into I type, II type and III type,

Wherein, I is poetry type cultural relics, II is other cultural relics, III is information description cultural relics.

The graph-text relationship is classified as P, pt, tp.GR (X) and T,

Wherein P is pure picture type, pt is pattern-to-pattern type, tp.GR (X) is pattern-to-pattern type, X is = Σgris gf, gr is the proportion of single picture to single page, gf is the function of picture, and Tpure Wen Zixing.

The type of reading requirement is classified as PC, LA, TI, DE, SM,

The PC is personal maintenance, mainly comprises famous celebrities, biography and other works, LA is literature appreciation, mainly comprises children literature works, novels, fairy tales and other works with specific plots and strong situation sense, or poetry, free words and artistic aesthetic works, TI is tool information, comprises science popularization works, social sciences knowledge works, natural science works, tool guides and other works, DE is discipline education works, mainly refers to teaching materials and externally related reference coaching books, SM is social moral works, and mainly comprises social norms, legal treatments and ideological and political education works.

(8) Multidimensional hierarchical mode:

The invention comprises a multidimensional grading system and a plurality of grading selection modes, wherein the plurality of grading modes are based on the grading system, are used for measuring the language reading difficulty and emotion receiving difficulty of the Chinese children's book through mathematical calculation and category judgment based on specific indexes in the system, and comprise a cognitive guiding mode, an emotion guiding mode and a cognitive emotion comprehensive guiding mode. The cognitive guiding mode comprises the steps of calculating a difficulty score based on language difficulty indexes and text characteristic data of Chinese children books, performing value-added sequencing on score results of the Chinese children books to determine a reading difficulty level sequence under cognitive guiding, the emotion guiding mode comprises the steps of calculating emotion polarity and emotion richness scores based on emotion dimension indexes and text emotion word characteristic data of the Chinese children books, performing value-added sequencing on score results of the Chinese children books to determine a reading difficulty level sequence under emotion guiding, and the cognitive emotion comprehensive mode comprises the step-added sequencing of normalized summation results based on language difficulty and emotion richness comprehensive scores of the Chinese children books, and determining a comprehensive difficulty level sequence of the books.

The text language difficulty readability formula is as follows:

RL=∑cl*cr+∑vl*vr,

Wherein RL is language difficulty, cl is Chinese character difficulty level coefficient, cr is the proportion of the Chinese characters in the level, vl is vocabulary difficulty level coefficient, vr is the proportion of the vocabulary in the level.

The text emotion polarity formula is:

Wherein SP is emotion polarity, ps is positive emotion sum in the text, nss is negative emotion sum in the text, absolute value is taken, and the value range is SP is more than or equal to 0.

The text emotion richness formula is as follows:

N=∑Ni,(i=1,2,3,4,5,6,7)

Wherein Ni represents emotion word frequency in each dimension of emotion, namely emotion, happiness, anger, sadness, fear, aversion and convulsion in the text, and the emotion word frequency is counted by matching with an emotion dictionary; N represents the total amount of text emotion words.

The cognitive emotion comprehensive difficulty score value is as follows:

R=RL+SD,

Wherein RL is language difficulty; SD is text emotion richness, RL and SD values are normalized, the RL value is equal to the sum of the ratios of word average difficulty and highest difficulty, and the normalized SD value is equal to the ratio of SD to extremum 1.

The following further details the implementation steps of the solution according to the invention in connection with a specific embodiment:

The books selected in the embodiment are all derived from a 'school student reading guidance catalog (2020 edition)' which is developed and released by the education material development center of the basic education course of the education department, and comprise a Chinese child book A, a Chinese child book B, a Chinese child book C, a Chinese child book D and a Chinese child book E. The method comprises the following specific steps:

Firstly, defining an index system;

The index system refers to an organism consisting of a plurality of individual indices with inherent links. Therefore, defining the index system refers to completing the definition of the related information and the concept of the index system, and the index system information records the overall description information and the creation information of the index system, including determining the classification purpose, defining the classification dimension and defining the basic information of the index system.

(1) Determining a grading purpose;

the grading purpose is to comprehensively evaluate the reading difficulty of the Chinese children books A, B, C, D and E in a multi-dimensional mode, and to sort the books A, B, C, D and E in a grading manner in various modes.

(2) Explicitly evaluating the dimension;

the hierarchical dimension is created according to the hierarchical object, the hierarchical purpose, and the overall characteristics of the present invention, and thus the present invention creates 3 dimensions altogether for text cognition, text emotion, and text function.

(3) Defining basic information of an index system;

The basic information is the general feature description of the index system, and comprises evaluation industry, industry field, evaluation description, evaluation content, evaluation purpose and the like. The inventor defines the basic information, the evaluation object of the index system is a published Chinese children's book, the reading difficulty of the Chinese children's book is to be evaluated from text cognition, text emotion and text function in a multi-dimensional mode, a ladder reading system of the children's book is formed, and the localized development of the grading reading is promoted.

Step two, constructing grading indexes;

The invention relates to a multi-dimensional grading system and an evaluation method of a Chinese children book, which take grading dimension as a grading index of a traction selection scheme layer, and comprise 3 aspects of language recognition difficulty, text emotion complexity, function type and the like of the Chinese children book, namely 6 indexes of language recognition difficulty, genre, image-text relationship, emotion polarity, emotion richness and reading requirement type, wherein the language recognition difficulty, emotion polarity and emotion richness are quantitative grading indexes, and the genre, image-text relationship and reading requirement type are qualitative classification indexes. The specific contents are shown in tables 1 and 2.

Table 1 quantitative grading index of Chinese children's book

Table 2 qualitative classification index of Chinese children's book

Thirdly, designing a grading expression mode;

The method comprises the steps of integrating 6 sub-surface grading indexes such as text cognition, text emotion, language cognition difficulty under 3-dimensional sub-surface of a text function, genre, picture-text relationship, emotion polarity, emotion richness, reading requirement type and the like through a sub-surface assembly classification method, compiling each index calculation result and classification identification into a multi-dimensional grading code, wherein the sub-surfaces are connected by a 'connection' to represent a parallel relationship, and the sub-surfaces are connected by a 'connection'.

Fourthly, constructing and processing word list;

The construction of the grading word list and the grading word list is derived from the processing of Chinese horizontal vocabulary and Chinese character grade outline (revision), the processing of the word list comprises the steps of keeping the first grade word and the second grade word unchanged, merging the third grade word and the third grade word annex into the third grade word, merging the fourth grade word and the fourth grade word annex into the fourth grade word, adding Chinese characters outside the fifth grade word recording word list, and the processing of the word list comprises the step of removing monosyllabic words in the word list, wherein the word grades of the first, second, third, fourth and fifth words correspond to word difficulty coefficients cl and vl respectively. The method comprises the steps of (1) c, setting up a multi-dimensional emotion dictionary, wherein c is a Chinese character difficulty level coefficient, vl is a vocabulary difficulty level coefficient, setting up a multi-dimensional emotion dictionary from a university Chinese emotion vocabulary ontology library of the university of great company, re-aggregating subclasses of the multi-dimensional emotion dictionary into the subclasses, correspondingly giving a subclass mark, and repeatedly classifying emotion words with auxiliary emotion classification into different emotion classifications.

Fifthly, text processing;

The method comprises the steps of preprocessing Chinese children books A, B, C, D and E, deleting the content such as the preamble, copyright information, author information and postscript of books A, B, C, D, E, reserving main text content as a subsequent operation object, utilizing tools to divide the preprocessed texts of the Chinese children books A, B, C, D and E into a set of Chinese character blocks and a set of word blocks, and importing the preprocessed texts of the Chinese children books A, B, C, D and E into a NLPIR analysis platform to count positive emotion and negative emotion values of the books A, B, C, D, E.

Sixthly, matching the text with the word list;

The method comprises the steps of respectively matching a Chinese child book A, a Chinese child book B, a Chinese child book C, a Chinese child book D and a Chinese child book E with a hierarchical word list through a python program code, and counting the number of Chinese characters of each level of the book A, B, C, D, E;

Seventh, classifying the classification judgment;

The classification identification of the books A, B, C, D, E is determined by manually browsing the Chinese children books A, the Chinese children books B, the Chinese children books C, the Chinese children books D and the Chinese children books E according to the basic classification rules.

Eighth step, multidimensional grading mode;

The invention comprises a cognitive guiding mode, an emotion guiding mode and a cognitive emotion comprehensive guiding mode. The cognitive guiding mode comprises the steps of calculating a difficulty score based on language difficulty indexes and text characteristic data of Chinese children books, performing value-added sequencing on score results of the Chinese children books to determine a reading difficulty level sequence under cognitive guiding, the emotion guiding mode comprises the steps of calculating emotion polarity and emotion richness scores based on emotion dimension indexes and text emotion word characteristic data of the Chinese children books, performing value-added sequencing on score results of the Chinese children books to determine a reading difficulty level sequence under emotion guiding, and the cognitive emotion comprehensive mode comprises the step-added sequencing of normalized summation results based on language difficulty and emotion richness comprehensive scores of the Chinese children books, and determining a comprehensive difficulty level sequence of the books.

(1) Cognitive guided mode;

Based on all the steps, finally obtaining the language difficulty scores of the Chinese children books A, B, C, D and E in the cognitive dimension of 3.96, 4.17, 4.88, 4.95 and 5.68 (two decimal places are reserved), and performing value-added sorting on the 5 books as shown in the table 3, so that the reading difficulty of the Chinese children book A is lowest under the cognitive direction, and then B, C, D, E is sequentially carried out. The best choice for children is from A, the requirement of the book on the language decoding capability of children is lowest, the reading fluency is also best for children, and then children can choose to read B, C, D, E orderly under the gradual lifting difficulty according to the capability, the targets, the interests and the like. Meanwhile, the multidimensional grading code display A, B, C is of the same genre type, A, C, D, E is of the same picture-text relationship, and an adjusting effect is provided for reading difficulty sequencing.

Table 3 cognitively oriented Chinese children book difficulty score and grading code

(2) Emotion guiding mode;

Based on all the steps, finally obtaining emotion polarity values of 1.97, 0.70, 1.45 and 1.11 (two decimal places are reserved) of the Chinese child book A, the Chinese child book B, the Chinese child book C, the Chinese child book D and the Chinese child book E in emotion dimensions, and performing value-added sequencing on the 5 books as shown in a table 4. Wherein B, C emotion polarity values are smaller than 1, the whole emotion tends to be negative, and the whole emotion tends to be positive when E, D, A emotion polarity values are larger than 1. The emotion polarity value determines emotion trend and distribution characteristics of the Chinese children books, and indicates emotion basic tones of the books by taking 1 as a limit. In this hierarchical mode, the child's reading choices are not fully ordered by the duty cycle, but are related to the child's emotional need. The child book A with the highest emotion polarity value can be selected when parents or teachers wish the child to show more positive optimistic attitudes or the child to perform more pleasant and relaxed reading activities, and the child book B or C can be selected when parents or teachers wish the child to increase self-thinking and thinking about complex negative emotion. In general, reading choices based on emotion polarity take actual emotion needs into account.

TABLE 4 emotion-oriented Chinese children's book emotion polarity score and grading code

Based on all the steps, finally obtaining the emotion richness values of the Chinese children books A, B, C, D and E in emotion dimensions of 0.64, 0.73, 0.75, 0.61 and 0.69 (two decimal places are reserved), and performing value-added sequencing on the 5 books as shown in the table 5. The reading difficulty of the Chinese children books displayed by the emotion richness values is D, A, E, B, C in sequence, so that children can start reading selection by the books D, compared with other books, the emotion dimension of the books D is relatively simple, and text emotion mainly comprises 'good' (57%) and 'happy' (23%), so that understanding is easy. For children with strong reading and understanding ability and emotion perception ability, C with emotion richness of 0.75 is more suitable, and the contents of the texts comprise 'good, bad, happy, fun, sad, frighten and anger' with the ratios of about 24.7%, 39%, 15.5%, 5%, 10.7%, 3.1% and 2%, so that the needs of the children on diversified emotions can be met.

TABLE 5 emotion oriented Chinese children book emotion richness score and grading code

(3) Cognitive emotion comprehensive guiding mode;

Based on all the steps, finally obtaining the reading difficulty scores of the Chinese children books A, B, C, D and E under the comprehensive dimension of the cognitive emotion of the Chinese children books, wherein the scores are respectively 1.16, 1.28, 1.39, 1.26 and 1.44 (two decimal places are reserved), and performing value-added sequencing on the 5 books as shown in a table 6, so that the reading difficulty of the Chinese children books A is lowest when the cognitive emotion is comprehensively guided, and then the reading difficulty scores are D, B, C, E in turn. Under the interaction of the cognitive reading difficulty and the emotion reading difficulty, the cognitive emotion comprehensive difficulty adjusts the book rank ordering under a single dimension. For example, the cognitive dimension reading difficulty rank-emotion dimension reading difficulty rank-cognitive emotion comprehensive difficulty rank is 2-4-3, and for example, the cognitive dimension reading difficulty rank-emotion dimension reading difficulty rank-cognitive emotion comprehensive difficulty rank is 4-1-2. In cognitive emotion integrated mode, children's reading may be selected according to A, D, B, C, E sequences.

Table 6 cognitive emotion comprehensive oriented Chinese children book comprehensive difficulty score and grading code

The foregoing is only a partial embodiment of the present invention, and it should be noted that it will be apparent to those skilled in the art that modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the present invention.

Claims

1. A multi-dimensional grading method for Chinese children's books, characterized by comprising the following steps:

S1. Define the index system from the three dimensions of text cognition, text emotion, and text function; according to the hierarchical dimensions of the index system, determine the operational indicators under each dimension, including: language cognition difficulty, style type, and image-text relationship under the text cognition dimension, emotional polarity and emotional richness under the text emotion dimension, and reading demand type under the text function dimension;

S2, vocabulary construction and processing: construct hierarchical word lists, hierarchical vocabulary lists, and multi-dimensional sentiment dictionaries;

S3. Text processing: pre-processing, word segmentation, character segmentation, and positive and negative sentiment statistics of Chinese children's books to retain the main part;

S4, matching text with vocabulary: matching the text after word segmentation with the graded vocabulary, and counting the number of Chinese characters at each level; matching the text after word segmentation with the graded vocabulary and the multidimensional sentiment dictionary, counting the number of words at each level and the frequency of sentiment words at each dimension, and obtaining the operation index of the reader under each dimension;

S5. Classification category determination: according to the classification rules, determine the style type, image-text relationship and reading demand type of the book, and determine the classification mark of the reading book;

S6. Based on the principle of facet group classification, the segmented marking system is adopted to integrate the indicators of each dimension, that is, the calculation results of each indicator and the classification mark are compiled into a multi-dimensional classification code to obtain the difficulty score and category of Chinese children's books under each dimension;

In step S5,

The stylistic types are classified as follows:

Class I, Class II, Class III,

Among them, Category I is poetry; Category II is other literary narratives; Category III is informational writing;

The relationship between images and texts is classified as follows:

P, Pt, Tp.GR(X), T,

Among them, P is pure picture type; Pt is picture with text type; Tp.GR(X) is text with picture type, X = ∑gr*gf, gr is the proportion of a single illustration on a single page, gf is the function of the illustration; T is pure text type;

Reading needs are classified into:

PC, LA, TI, DE, SM,

Among them, PC is for personal cultivation, LA is for literary appreciation, TI is for tool information; DE is for subject education, and SM is for social ethics.

2. A multi-dimensional grading method for Chinese children's reading books as described in claim 1, characterized in that, in step S2, the construction of the graded character table and the graded vocabulary table is derived from the processing of the "Outline of Chinese Proficiency Vocabulary and Chinese Character Levels (Revised Edition)" formulated by the Examination Center of the National Chinese Proficiency Examination Committee Office; wherein the processing of the character table includes keeping Class A characters and Class B characters unchanged, merging Class C characters and Class C character appendix into Class C characters, merging Class D characters and Class D character appendix into Class D characters, and adding Class E characters to include Chinese characters outside the character table; the processing of the vocabulary table is to remove monosyllabic words in the vocabulary table; the character and word levels of A, B, C, D, and E correspond to character and word difficulty coefficients cl and vl respectively, cl is the Chinese character difficulty level coefficient, and vl is the vocabulary difficulty level coefficient.

3. A multidimensional grading method for Chinese children's books as described in claim 1, characterized in that, in step S2, the construction of the multidimensional sentiment dictionary is derived from the Chinese sentiment vocabulary ontology library of Dalian University of Technology, and the sentiment words with auxiliary sentiment classification are repeatedly classified into different sentiment classifications by re-aggregating their subcategories into major categories and assigning corresponding major category identifiers.

4. A multi-dimensional grading method for Chinese children's books as described in claim 1, characterized in that, in step S6, the facets of the grading code of the Chinese children's books are connected with ":" to indicate a parallel relationship, and the sub-facets are connected with "-"; according to the scores of the language difficulty, emotional polarity, emotional richness, and cognitive and emotional comprehensive difficulty of the children's books, corresponding combinations are formed to form the reading difficulty ranking of the Chinese children's books under each mode.

5. A multi-dimensional grading method for Chinese children's books as claimed in claim 4, characterized in that the calculation formula for the language difficulty and readability of the books is:

RL＝∑cl*cr+∑vl*vr，

Among them, RL is the language difficulty, cl is the Chinese character difficulty level coefficient, cr is the proportion of Chinese characters at this level; vl is the vocabulary difficulty level coefficient, vr is the proportion of vocabulary at this level. The difficulty level coefficients of Chinese characters and vocabulary are defined by sorting and constructing a graded character table and a graded word table.

6. A multi-dimensional grading method for Chinese children's books as claimed in claim 4, characterized in that the calculation formula of the sentiment polarity of the books is:

Among them, SP is the sentiment polarity; pss is the sum of positive sentiment in the text; nss is the sum of negative sentiment in the text, taking its absolute value; the value range is SP≥0.

7. A multi-dimensional grading method for Chinese children's books as claimed in claim 4, characterized in that the calculation formula for the emotional richness of the books is:

N＝∑Ni，i＝1,2,3,4,5,6,7

Among them, Ni represents the frequency of emotional words in the text under the dimensions of happiness, good, anger, sadness, fear, hate, and surprise, which is counted by matching with the emotional dictionary; N represents the total number of emotional words in the text.

8. A multi-dimensional grading method for Chinese children's books as claimed in claim 4, characterized in that the calculation formula for the cognitive-emotional comprehensive difficulty score is:

R＝RL+SD，

Among them, RL is the language difficulty; SD is the emotional richness of the text; the RL and SD values are normalized, and the RL value is equal to the sum of the ratios of the average difficulty of characters and words to the highest difficulty; the normalized SD value is equal to the ratio of SD to the extreme value 1.

9. A multi-dimensional grading system for Chinese children's books, characterized by comprising:

The indicator construction module is used to define the indicator system from the three dimensions of text cognition, text emotion, and text function. According to the hierarchical dimensions of the indicator system, the operating indicators under each dimension are determined, including: language cognition difficulty, style type, and image-text relationship under the text cognition dimension, emotional polarity and emotional richness under the text emotion dimension, and reading demand type under the text function dimension;

The vocabulary building and processing module is used to build a hierarchical word list, a hierarchical vocabulary list, and a multi-dimensional sentiment dictionary;

The text processing module is used to perform preprocessing, word segmentation, character segmentation, and positive and negative sentiment statistical processing on Chinese children's books to retain the main part;

The text and vocabulary matching module is used to match the text after word segmentation with the graded word table, count the number of Chinese characters at each level, and match the text after word segmentation with the graded word table and the multi-dimensional sentiment dictionary respectively, count the number of words at each level and the frequency of sentiment words at each dimension; and then obtain the operation index of the reader under each dimension;

The classification category determination module is used to determine the book's genre, image-text relationship, and reading demand type based on the classification rules, and determine the classification identification of the book;

The grading module integrates the indicators of each dimension using a segmented labeling system based on the principle of faceted group classification. That is, the calculation results of each indicator and the classification mark are compiled into a multi-dimensional grading code to obtain the difficulty score and category of Chinese children's books under each dimension.