Object evaluation method, device, apparatus, storage medium, and capability evaluation device

Abstract

The application provides an object evaluation method, device, equipment, storage medium and capability evaluation equipment. The method comprises the following steps: obtaining each evaluation result, wherein the evaluation result is determined by evaluating the capability of an evaluated object in each evaluation dimension; determining the correlation between each evaluation dimension and the specific aspect capability according to the capability difference degree of different types of objects in the preset multiple types of objects in the same evaluation dimension and the capability similarity of the same type of objects in the preset multiple types of objects in different evaluation dimensions; and classifying the evaluated object into one type of object in the preset multiple types of objects according to the evaluation result and the correlation between each evaluation dimension and the specific aspect capability. The above processing process can more accurately evaluate the specific aspect capability of the evaluated object.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to an object evaluation method, apparatus, device, storage medium, and capability evaluation equipment. Background Technology

[0002] Assessing specific abilities of a subject is essential for understanding its development in those areas and for identifying areas of weakness. For example, evaluating the abilities of intelligent robots or similar devices in language, reasoning, reading, and motor skills allows us to understand their skill levels in these areas and identify areas for improvement, enabling us to develop targeted and effective improvement plans.

[0003] Currently, conventional competency assessment schemes typically involve first determining which assessment dimensions to use to evaluate specific competencies, then assessing the competencies of the assessed individual in each dimension, and finally determining the competency level of the assessed individual in that specific area based on the assessment results of each dimension.

[0004] The accuracy of the above-mentioned competency assessment schemes is low, and they usually cannot obtain accurate assessment results. Summary of the Invention

[0005] Based on the above-mentioned technological status, this application proposes an object evaluation method, apparatus, equipment, storage medium, and capability evaluation device, which can obtain capability evaluation results more accurately.

[0006] To improve the accuracy of the evaluation results, this application proposes the following technical solution:

[0007] An object evaluation method, comprising:

[0008] Each evaluation result is obtained by assessing the capabilities of the evaluated object in each evaluation dimension, wherein each evaluation dimension is an evaluation dimension used to evaluate specific aspects of capabilities.

[0009] Based on the difference in capabilities of different types of objects in the same evaluation dimension across various evaluation dimensions, and the similarity in capabilities of the same types of objects in different evaluation dimensions across various evaluation dimensions, the correlation between each evaluation dimension and the specific aspect capability is determined; wherein, the specific aspect capability of the same types of objects in the preset multiple types of objects is the same, and the specific aspect capability of different types of objects in the preset multiple types of objects is different.

[0010] Based on the evaluation results and the correlation between each evaluation dimension and the specific aspect of capability, the evaluated object is classified into one of the preset multiple types of objects.

[0011] An object evaluation device, comprising:

[0012] The data acquisition unit is used to acquire various evaluation results, which are determined by evaluating the capabilities of the evaluated object in various evaluation dimensions. The various evaluation dimensions are various evaluation dimensions used to evaluate specific aspects of capabilities.

[0013] The calculation processing unit is used to determine the correlation between each evaluation dimension and the specific aspect ability based on the ability difference degree of different types of objects in the same evaluation dimension in each preset multiple types of objects, and the ability similarity degree of the same types of objects in different evaluation dimensions in each preset multiple types of objects; wherein, the specific aspect ability of the same types of objects in the preset multiple types of objects is the same, and the specific aspect ability of different types of objects in the preset multiple types of objects is different.

[0014] The classification processing unit is used to classify the evaluated object into one of the preset multiple types of objects based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of the capability.

[0015] An electronic device, comprising:

[0016] Memory and processor;

[0017] The memory is connected to the processor and is used to store computer programs;

[0018] The processor is used to implement the above-described object evaluation method by running the program in the memory.

[0019] A competency assessment device, comprising:

[0020] Input / output devices, and a processor connected to said input / output devices;

[0021] The input / output device is used to output the graphical interface of the evaluation task and the audio data of the evaluation task, and to collect evaluation data when the evaluated object performs the evaluation task, the evaluation data including the voice data and / or behavioral data of the evaluated object when performing the evaluation task;

[0022] The processor is configured to score the evaluation data through various evaluation dimensions to obtain evaluation results for each evaluation dimension; each evaluation dimension is a dimension used to evaluate specific aspects of ability; based on the difference in ability of different types of objects in the same evaluation dimension among various preset types of objects, and the similarity in ability of the same types of objects in different evaluation dimensions among various preset types of objects, the processor determines the correlation between each evaluation dimension and the specific aspect of ability; wherein, the specific aspect of ability of the same types of objects in the preset multiple types of objects is the same, and the specific aspect of ability of different types of objects in the preset multiple types of objects is different; based on the evaluation results and the correlation between each evaluation dimension and the specific aspect of ability, the evaluated object is classified into one type of object among the preset multiple types of objects.

[0023] A storage medium storing a computer program, which, when executed by a processor, implements the above-described object evaluation method.

[0024] The object evaluation method proposed in this application can accurately grasp the correlation between different evaluation dimensions and specific aspects of capabilities by measuring the difference in capabilities of different types of objects across various evaluation dimensions, as well as the similarity in capabilities of objects of the same type across various evaluation dimensions. This allows for a better understanding of the role of different evaluation dimensions in evaluating specific aspects of capabilities. Based on this, embodiments of this application classify the evaluated object into one type of object from among multiple object types according to the correlation between each evaluation dimension and specific aspects of capabilities, and the evaluation results corresponding to each evaluation dimension. The above-described object classification process comprehensively considers the evaluation results of each evaluation dimension and the role of each evaluation dimension in evaluating specific aspects of capabilities, thus making the classification results more accurate, i.e., more accurately determining the specific aspects of capabilities of the evaluated object.

[0025] Furthermore, based on the similarities and differences in the abilities of different groups across various assessment dimensions, this application scientifically and accurately grasps the magnitude of the role of each assessment dimension in assessing specific aspects of ability. Moreover, by classifying the assessed objects based on specific aspects of ability through the assessment results of each dimension and the correlation between each assessment dimension and specific aspects of ability, the classification results can be made more interpretable. For example, by comparing the assessment results of each assessment dimension and the correlation between each assessment dimension and specific aspects of ability, it can be known in which dimensions the assessed objects are better, in which dimensions their abilities need to be improved, and in which dimensions the assessment results lead to the direction of the classification results of the assessed objects. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0027] Figure 1 A flowchart illustrating an object evaluation method provided in an embodiment of this application;

[0028] Figure 2 A schematic diagram showing the distribution of evaluation scores for the first type of object and the second type of object in the perception dimension, as provided in the embodiments of this application;

[0029] Figure 3 A schematic diagram of a comprehensive scoring system based on exploratory factor analysis provided for embodiments of this application;

[0030] Figure 4 This is a schematic diagram of the structure of an object evaluation device provided in an embodiment of this application;

[0031] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application;

[0032] Figure 6 This is a schematic diagram of the structure of a capability assessment device provided in an embodiment of this application. Detailed Implementation

[0033] The technical solution of this application embodiment is applicable to capability assessment application scenarios. By adopting the technical solution of this application embodiment, the basic assessment results of capability assessment can be post-processed to determine the correlation between each assessment dimension and the capability to be assessed. Then, by using the assessment results of each assessment dimension and the correlation between each assessment dimension and the capability to be assessed, the assessed object can be classified and the capability level of the assessed object can be determined. This solution can more accurately assess the capability of the assessed object.

[0034] Typically, assessing a subject's ability in a specific area requires evaluation across multiple subdivided dimensions to ensure comprehensiveness. For instance, assessing reasoning ability usually involves multiple dimensions such as text-image reasoning, image combination, relationship analysis, and deductive calculation. Each dimension's evaluation is completed, and the results are then combined to arrive at the overall assessment of the subject's reasoning ability. Similarly, assessing athletic ability typically involves multiple dimensions such as long-distance running, sprinting, high jump, long jump, flexibility testing, and strength testing. Each dimension's evaluation is completed, and the results are then combined to arrive at the overall assessment of the subject's athletic ability.

[0035] The common ability assessment schemes mentioned above usually involve simply summing up the assessment results of each assessment dimension to obtain the final assessment score, and directly determining the ability level of the assessed individual based on the final assessment score.

[0036] The above evaluation scheme cannot accurately grasp the degree of influence of each evaluation dimension on the final evaluation result, which will lead to inaccurate final evaluation results and make it difficult to scientifically and objectively distinguish the ability levels of different evaluated objects.

[0037] Based on the above problems, the inventors of this application propose an object evaluation scheme that can accurately grasp the correlation between each evaluation dimension and the ability to be evaluated. The evaluation results of each evaluation dimension and the correlation between each evaluation dimension and the ability to be evaluated are used together to classify the ability of the object being evaluated, thereby more accurately determining the level of the ability of the object to be evaluated.

[0038] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] Exemplary methods

[0040] This application first proposes an object evaluation method, which can be executed by a computer device, such as a computer, a computer processor, a smart terminal device, a server, etc.

[0041] See Figure 1 As shown, the method includes:

[0042] S101. Obtain the results of each evaluation.

[0043] The evaluation results are determined by assessing the capabilities of the evaluated object in each evaluation dimension, where each evaluation dimension is a specific dimension used to evaluate capabilities in a particular area.

[0044] Specifically, the objects being evaluated can be any living or non-living, tangible or intangible object possessing a specific ability. For example, if the specific ability is motor ability, the objects being evaluated could be living organisms, robots, vehicles, or moving mechanical parts; if the specific ability is language ability, the objects being evaluated could be intelligent robots, humans, or neural network language models.

[0045] The aforementioned specific abilities can be any kind of ability, such as language ability, reasoning ability, reading ability, motor ability, etc.

[0046] The aforementioned assessment dimensions refer to the various subdivisions used to assess specific abilities. For example, if the specific ability being assessed is reasoning ability, the corresponding assessment dimensions could be dimensions such as text-image reasoning, graphic combination, relationship analysis, and deductive calculation. By assessing abilities in these dimensions, the overall reasoning ability of the assessed individual can be comprehensively reflected. If the specific ability being assessed is athletic ability, the corresponding assessment dimensions could be dimensions such as long-distance running, sprinting, high jump, long jump, flexibility testing, and strength testing. By assessing abilities in these dimensions, the overall athletic ability of the assessed individual can be comprehensively reflected. If the specific ability being assessed is reading ability, the corresponding assessment dimensions could be dimensions such as perception, attention, and memory. By assessing the perception, attention, and memory of the assessed individual, the overall reading ability of the assessed individual can be comprehensively reflected.

[0047] In the actual implementation of the technical solutions of this application, the various evaluation dimensions used to assess specific capabilities can be predetermined according to industry practices. For example, common industry methods for evaluating specific capabilities can be statistically analyzed to determine which dimensions are used in the industry to evaluate specific capabilities, and these dimensions can then be determined as the evaluation dimensions for assessing specific capabilities.

[0048] The aforementioned acquisition of various evaluation results can refer to obtaining evaluation results determined by assessing the capabilities of the evaluated object in various evaluation dimensions. These evaluation results can be qualitative, such as excellent, good, acceptable, or poor, or quantitative, such as evaluation scores. When acquiring these evaluation results, they can be pre-obtained evaluation results received or read during the execution of the technical solutions in the embodiments of this application. These evaluation results are determined by assessing the capabilities of the evaluated object in various evaluation dimensions; that is, one evaluation result is obtained for each evaluation dimension.

[0049] Alternatively, the acquisition of each evaluation result can also be achieved by evaluating the capabilities of the evaluated object through each evaluation dimension when executing the technical solution of the embodiments of this application, that is, evaluating the capabilities of the evaluated object in each evaluation dimension and obtaining the evaluation result corresponding to each evaluation dimension.

[0050] The specific process of evaluating the capabilities of the evaluated object through different evaluation dimensions to obtain evaluation results can refer to conventional evaluation schemes.

[0051] S102. Based on the difference in capabilities of different types of objects in the same evaluation dimension in each of the preset multiple types of objects, and the similarity in capabilities of the same types of objects in different evaluation dimensions in each of the preset multiple types of objects, determine the correlation between each evaluation dimension and the specific aspect capability.

[0052] Among the preset multiple types of objects, objects of the same type have the same specific aspect capabilities, while objects of different types have different specific aspect capabilities.

[0053] In this application embodiment, multiple different types of objects are defined in advance based on the differences in specific aspects of their capabilities, wherein the specific aspects of the capabilities of each type of object are different from each other.

[0054] Furthermore, different types of objects exhibit significant differences in capabilities in specific aspects; for example, the degree of difference in capabilities among different types of objects in these specific aspects is greater than a predetermined degree of difference. Objects of the same type, on the other hand, refer to a group of objects that possess the same capabilities in specific aspects.

[0055] For example, assuming the specific ability mentioned above is reading ability, then different types of objects are respectively objects with normal reading ability and objects with abnormal reading ability. Objects of the same type are either objects with normal reading ability or objects with abnormal reading ability.

[0056] It is understandable that the above-mentioned objects of the same or different types represent a class of objects with the same or different levels of capabilities in a specific aspect.

[0057] This application's embodiments measure the capability differences of different types of objects in the same evaluation dimension across various evaluation dimensions, as well as the capability similarity of the same type of objects across different evaluation dimensions.

[0058] Among them, the ability difference of different types of objects in the same evaluation dimension in each evaluation dimension can reflect the ability difference of groups with different abilities in a certain aspect in a certain evaluation dimension, thereby determining the degree of influence of the ability of the evaluated object in a certain evaluation dimension on the evaluated object's ability in the aforementioned specific aspect.

[0059] For example, suppose we assess a subject's reading ability using three dimensions: perception, attention, and memory. Regarding the attention dimension, analyzing the attention abilities of a group with normal reading ability and a group with abnormal reading ability reveals a significant difference between the two groups; specifically, the attention ability of the group with normal reading ability is significantly higher than that of the group with abnormal reading ability. This demonstrates that the level of a subject's attention ability has a substantial impact on their overall reading ability.

[0060] The similarity of abilities of the same type of objects across different evaluation dimensions can reflect the similarity of abilities of a group with the same abilities in a specific aspect across different evaluation dimensions. This allows us to determine the correlation between different evaluation dimensions when evaluating a specific aspect of ability, that is, to determine the degree of influence of different evaluation dimensions on evaluating a specific aspect of ability.

[0061] For example, suppose we assess a subject's reading ability using three dimensions: perception, attention, and memory. For the attention dimension, analyzing the attention abilities of groups with normal and abnormal reading abilities reveals a high similarity between the attention abilities of the group with normal reading abilities and their perception and memory abilities. Similarly, the attention abilities of the group with abnormal reading abilities show a high similarity between their perception and memory abilities. This indicates a high correlation between the subject's attention ability and their perception and memory abilities. The degree of correlation between a particular assessment dimension and other assessment dimensions directly determines the extent to which that dimension influences the assessment of that specific aspect of the subject's ability.

[0062] When a certain evaluation dimension is highly correlated with other evaluation dimensions, the change in the level of the ability of that evaluation dimension is the same as the change in the level of the ability of other evaluation dimensions. In other words, when the level of the ability of that evaluation dimension changes, the level of the ability of other evaluation dimensions also changes accordingly.

[0063] For example, taking attention ability as an example again, if it is highly correlated with perception and memory abilities, then when a subject's attention ability is high, their perception and memory abilities are also high, thus indicating that the subject's reading ability is high; conversely, when a subject's attention ability is low, their perception and memory abilities are also low, thus indicating that the subject's reading ability is low.

[0064] In summary, by measuring the difference in ability of different types of objects in the same evaluation dimension across various evaluation dimensions, and the similarity in ability of the same type of objects across different evaluation dimensions across various evaluation dimensions, we can determine the degree of influence of each evaluation dimension used to evaluate a specific aspect of ability on the evaluation of that specific aspect of ability.

[0065] Based on the aforementioned evaluation dimensions for assessing specific capabilities, and the degree of influence of each evaluation dimension on the assessment of specific capabilities, embodiments of this application determine the correlation between each evaluation dimension and the aforementioned specific capabilities. This correlation can also be regarded as the evaluation weight corresponding to each evaluation dimension for assessing specific capabilities.

[0066] For example, when a certain evaluation dimension has a high degree of influence on the evaluation of a specific aspect of ability, its correlation with the aforementioned specific aspect of ability is high, and its corresponding evaluation weight is large. When a certain evaluation dimension has a low degree of influence on the evaluation of a specific aspect of ability, its correlation with the aforementioned specific aspect of ability is low, and its corresponding evaluation weight is small.

[0067] Therefore, the degree of correlation between each evaluation dimension and the aforementioned specific aspects of ability reflects the extent to which the evaluation dimension has an impact on the specific aspects of the evaluated object's ability.

[0068] The correlation between the evaluation dimensions and the specific capabilities mentioned above, and the specific correspondence between them and their corresponding evaluation weight values, can be flexibly set based on the above basic ideas.

[0069] In the actual implementation of the technical solution of this application embodiment, after it is clear which specific aspect of capability is being evaluated and the various evaluation dimensions used to evaluate that specific aspect of capability, the correlation between each evaluation dimension and the specific aspect of capability can be determined by measuring the capability difference of different types of objects in the same evaluation dimension in each of the various preset types of objects, and the capability similarity of the same types of objects in different evaluation dimensions in each of the various preset types of objects. That is, the evaluation weight corresponding to each evaluation dimension is determined.

[0070] S103. Based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of the capability, classify the evaluated object into one of the preset multiple types of objects.

[0071] Specifically, based on the evaluation results and the correlation between each evaluation dimension and specific aspects of ability, the evaluated object is classified according to the specific aspects of ability, and it is determined which type of object among the aforementioned preset types of objects it belongs to, thereby determining the level of the evaluated object's specific aspects of ability.

[0072] As an exemplary implementation, a classification model can be pre-trained. This model takes as input the evaluation results of the evaluated object in each evaluation dimension and the correlation information between each evaluation dimension and specific aspects of ability. It can classify the evaluated object into one of a number of preset types of objects.

[0073] As another exemplary implementation, the comprehensive evaluation result of the specific aspect ability of the evaluated object can be determined first based on the evaluation results and the correlation between each evaluation dimension and the specific aspect ability; then, based on the comprehensive evaluation result of the specific aspect ability of the evaluated object, the evaluated object can be classified into one of the preset multiple types of objects.

[0074] For example, when each evaluation result is a score corresponding to a specific evaluation dimension, the correlation between each evaluation dimension and the specific ability can be used as its corresponding evaluation weight. The scores for each evaluation dimension can then be weighted and summed, or the scores for each evaluation dimension can be weighted and summed before averaging to obtain the comprehensive evaluation score for the specific ability. Then, based on the comprehensive evaluation score of the evaluated object for the specific evaluation aspect, the evaluated object can be classified into one of several preset object types.

[0075] It is understandable that in the above comprehensive evaluation score, the evaluation scores of different evaluation dimensions have different weights. Therefore, the comprehensive evaluation score is more objective and can more accurately reflect the level of ability in a specific aspect.

[0076] As described above, the object evaluation method proposed in this application can accurately grasp the correlation between different evaluation dimensions and specific aspects of capabilities by measuring the difference in capabilities of different types of objects across various evaluation dimensions, as well as the similarity in capabilities of objects of the same type across various evaluation dimensions. This allows for a better understanding of the impact of different evaluation dimensions on evaluating specific aspects of capabilities. Based on this, the embodiments of this application classify the evaluated object into one type of object from among multiple object types according to the correlation between each evaluation dimension and specific aspects of capabilities, and the evaluation results corresponding to each evaluation dimension. The above-described object classification process comprehensively considers the evaluation results of each evaluation dimension and the impact of each evaluation dimension on evaluating specific aspects of capabilities, thus making the classification results more accurate, i.e., more accurately determining the specific capabilities of the evaluated object.

[0077] Furthermore, based on the similarities and differences in the abilities of different groups across various assessment dimensions, this application scientifically and accurately grasps the magnitude of the role of each assessment dimension in assessing specific aspects of ability. Moreover, by classifying the assessed objects based on specific aspects of ability through the assessment results of each dimension and the correlation between each assessment dimension and specific aspects of ability, the classification results can be made more interpretable. For example, by comparing the assessment results of each assessment dimension and the correlation between each assessment dimension and specific aspects of ability, it can be known in which dimensions the assessed objects are better, in which dimensions their abilities need to be improved, and in which dimensions the assessment results lead to the direction of the classification results of the assessed objects.

[0078] The following section uses reading ability assessment as an example to introduce the specific processing procedure for classifying the assessed objects according to the object assessment method proposed in the embodiments of this application.

[0079] First, this application's embodiments study existing reading ability assessment schemes and propose a multi-dimensional, multi-task reading ability assessment task system.

[0080] Specifically, literature on reading ability assessment and reading disorder screening was retrieved from literature databases. The search employed a combination of subject terms and free-form terms. Chinese search terms included reading difficulty, reading disorder, screening, and classification; English search terms included reading disorder, reading difficulty, poor read, developmental dyslexia, screening, and Chinese.

[0081] Then, the retrieved literature was screened to select those containing experimental data and research group information from experimental studies, those including the tasks and standards used in the assessments, and those involving the examination of reading comprehension, perception, attention, and memory. Furthermore, the selected literature was not a review article or a duplicate publication. Based on this, the selected qualified literature was coded, with the coding information including the following three aspects: First, the title, author, region, and publication date; second, the basic characteristics of the research subjects, such as the intelligence level, age, and screening criteria of each group; and third, the outcome indicators and measurement data of interest to the research, such as the screening tasks and their metrics.

[0082] Secondly, using the aforementioned document coding as input, a meta-analysis method was used to statistically analyze the assessment dimensions and assessment tasks, thereby determining a multi-task assessment task system corresponding to the three dimensions of perception, attention, and memory.

[0083] The assessment task system consists of multiple assessment tasks that evaluate abilities across various dimensions. For example, some tasks assess abilities in the perception dimension, some in the attention dimension, and some in the memory dimension.

[0084] Furthermore, this application embodiment also verifies the effectiveness of the aforementioned assessment task system. Specifically, firstly, various reliability analyses (homogeneity reliability analysis (Cronbach's α reliability coefficient method), split-half reliability analysis, test-retest reliability analysis, etc.) and validity analyses (criterion validity analysis, construct validity analysis, single-item and total-sum correlation validity analysis, etc.) are used to conduct a multi-dimensional analysis of the internal consistency, stability, and reliability of the aforementioned reading ability assessment task system; then, variance tests are used to analyze the aforementioned reading ability assessment task system to evaluate whether its test results are statistically significant; finally, structural equation modeling is used to conduct factor analysis on the aforementioned reading ability assessment task system to deeply analyze the effect of individual indicators on the overall system and the interrelationships between individual indicators, thereby optimizing the standardized assessment dimensions and corresponding assessment tasks for different groups.

[0085] Through the above validity verification, it can be guaranteed that the final evaluation task put into application is scientific, reasonable, and effective in evaluating reading ability.

[0086] In practical assessments of reading ability, evaluations can be conducted from three dimensions: perception, attention, and memory. Alternatively, any two dimensions can be selected, or only one dimension can be used. However, it should be understood that the more comprehensive the assessment dimensions, the more accurate the evaluation of reading ability. In this embodiment, we will use a three-dimensional reading ability assessment as an example to introduce the technical solution of this embodiment.

[0087] In addition, the present application embodiment has specially designed the display format of the above-mentioned evaluation task system, so that the task display method of the evaluation task system is more natural, and the test subject completes the evaluation task in a natural interactive process.

[0088] For example, in the aforementioned reading ability assessment, some test subjects (such as school-aged children) may experience anxiety, fear, and worry about poor assessment results. Therefore, this embodiment of the application can effectively help test subjects relax by setting up gamified scenarios, allowing them to naturally complete the assessment process during gameplay. At the same time, gamified scenarios can effectively attract the attention of test subjects, achieving the goal of intervention and training through assessment.

[0089] Currently, the relatively mature theoretical foundation of gamification is the ARCS motivation theory. The ARCS model focuses on how to motivate students through instructional design. "A" stands for attention, which involves attracting students' attention at the beginning of the learning activity through the virtual context of the game. "R" stands for relevance, meaning the game content should be related to familiar things, examples, or concepts for school-aged children, and the game should also be relevant to their actual needs. "C" stands for confidence, guiding school-aged children to correctly view the game's outcome and strengthening their belief in successfully completing the assessment. "S" stands for satisfaction, where students experience mental pleasure and a sense of accomplishment from acquiring skills in gamified learning. Therefore, this application, supported by ARCS theory, proposes interface interaction design principles based on general perception and cognitive abilities:

[0090] Interface interaction design based on the perceptual level: For school-aged children, it is difficult to form top-down information processing during gamified interface interactions. Their main perception comes from the stimulation of the game's rules, art resources, and buttons. Therefore, this application's embodiments design the game's icons, interface types, and interactive operation forms to ensure that when children interact with gamified tasks, their core perceptual experience comes from the game's visual perceptual stimulation.

[0091] Interface interaction design based on attention level: Since the focus of attention development varies among school-aged children at different stages (e.g., 10-12 years old emphasizes attention allocation, while 12-15 years old emphasizes attention shifting), this application embodiment stimulates children through engaging interactions, sound effects, and animated visuals. It also enhances children's concentration by appropriately improving the coherence and flow of the reading comprehension assessment interaction process. Furthermore, sound effects, motion effects, and color changes can also effectively focus children's attention to a certain extent.

[0092] Memory-based interface interaction design: This application embodiment organizes and categorizes the functional entry points in the reading ability assessment interface for school-aged children, highlighting the core functions, which is more in line with the memory and cognitive characteristics of children, and will also help improve children's operational efficiency and interaction in the main interface and other interfaces from the perspective of memory and cognition.

[0093] After the above design, the reading ability assessment task system will be more natural and conducive to obtaining more objective assessment results.

[0094] The aforementioned evaluation task system can be set up in any terminal product or on a cloud server, while a client is set up for users to log in on any terminal to conduct the evaluation.

[0095] Once the evaluation task system, consisting of the aforementioned multiple evaluation tasks, is determined, the test subject can execute the evaluation task system, that is, complete the tasks in the evaluation task system, thereby obtaining the evaluation results corresponding to each evaluation dimension.

[0096] In order to measure the difference in ability of different types of objects in the same evaluation dimension in various evaluation dimensions, and the similarity in ability of the same type of objects in different evaluation dimensions in various evaluation dimensions, this application embodiment pre-organizes a set number of objects with normal reading ability as the first type of objects, and organizes the same number of objects with abnormal reading ability as the second type of objects.

[0097] Then, the first type of objects and the second type of objects respectively perform the above-described evaluation task system, obtaining the evaluation scores for each object in the first type of objects across each evaluation dimension, and obtaining the evaluation scores for each object in the second type of objects across each evaluation dimension. That is, the evaluation scores for each object in the first type of objects across the three dimensions of perception, attention, and memory are obtained, and the evaluation scores for each object in the second type of objects across the three dimensions of perception, attention, and memory are obtained. It can be understood that the evaluation scores for the first and second type of objects in each dimension reflect their respective ability levels in each dimension.

[0098] Then, for each evaluation dimension, the difference between the evaluation score distribution of a set number of first-type objects and the evaluation score distribution of a set number of second-type objects is calculated to obtain the ability difference degree of the first-type objects and the second-type objects in the same evaluation dimension in each evaluation dimension.

[0099] Specifically, for each of the three dimensions of perception, attention, and memory, the difference in the distribution of the evaluation scores of each type of object and the second type of object in that dimension can be calculated using JS divergence to obtain the difference in ability between the first type of object and the second type of object in that evaluation dimension.

[0100] Taking the sensory dimension as an example, see [link to relevant documentation]. Figure 2 As shown, by statistically analyzing the evaluation score distribution of Type 1 objects with normal reading ability and Type 2 objects with abnormal reading ability in the perception dimension, the difference in evaluation scores between the two in this dimension can be determined, and the difference in ability between the two in this dimension can be determined as W1.

[0101] Similarly, for the attention and memory dimensions, the differences in ability between the first type of object and the second type of object in these dimensions, W2 and W3, can also be determined through the above processing.

[0102] On the other hand, for each type of object, the similarity of the evaluation scores of each type of object in each evaluation dimension is calculated to determine the ability similarity of the type of objects in each evaluation dimension.

[0103] For example, using the Pearson correlation coefficient, the similarity of the assessment scores of the first type of object in the two dimensions of perception and attention is calculated and denoted as a12. The similarity of the assessment scores of the first type of object in the two dimensions of perception and memory is calculated and denoted as a13. The similarity of the assessment scores of the first type of object in the two dimensions of attention and memory is calculated and denoted as a23.

[0104] Similarly, following the above technical solution, for the second type of object, the similarity of the evaluation scores of each second type of object across each evaluation dimension is calculated to determine the ability similarity of the second type of object in each evaluation dimension, denoted as b12, b13, and b23, respectively. Here, b12 represents the similarity of the evaluation scores of the second type of object in the dimensions of perception and attention, b13 represents the similarity of the evaluation scores of the second type of object in the dimensions of perception and memory, and b23 represents the similarity of the evaluation scores of the second type of object in the dimensions of attention and memory.

[0105] The similarity of the evaluation scores of the first type of object and the second type of object across each evaluation dimension reflects the similarity of their capabilities across each evaluation dimension.

[0106] Based on this, when the object evaluation method proposed in the embodiments of this application is implemented, and the correlation degree between each evaluation dimension and the specific aspect of the capability is determined, the correlation degree between each evaluation dimension and the specific aspect of the capability can be calculated based on the capability difference degree of the first type of object and the second type of object in the same evaluation dimension, the capability similarity of the first type of object in each evaluation dimension, and the capability similarity of the second type of object in each evaluation dimension.

[0107] Specifically, in the reading ability assessment, for each of the three assessment dimensions—perception, attention, and memory—the correlation between the dimension and reading ability is determined through the following processing:

[0108] The correlation between the assessment dimension and reading ability is calculated based on the similarity between the ability of the first type of object in this assessment dimension and the ability of the first type of object in other assessment dimensions, the similarity between the ability of the second type of object in this assessment dimension and the ability of the second type of object in other assessment dimensions, and the difference between the abilities of the first type of object and the second type of object in this assessment dimension.

[0109] For example, the average similarity between the ability of the first type of object in this evaluation dimension and the ability of the first type of object in other evaluation dimensions is calculated based on the similarity between the ability of the second type of object in this evaluation dimension and the ability of the second type of object in other evaluation dimensions.

[0110] Then, the correlation between the assessment dimension and reading ability is determined by multiplying the average similarity between the ability in this assessment dimension and the ability in other assessment dimensions, and the difference between the abilities of the first type of object and the second type of object in this assessment dimension.

[0111] For example, taking the perception dimension as an example, based on the similarity a12 and a13 between the ability of the first type of object in the perception dimension and the ability of the first type of object in the attention and memory dimensions, the similarity b12 and b13 between the ability of the second type of object in the perception dimension and the ability of the second type of object in the attention and memory dimensions, and the difference W1 between the abilities of the first type of object and the second type of object in the perception dimension, the correlation K1 between this assessment dimension and reading ability is calculated.

[0112] The specific process is as follows: First, based on the similarity a12 and a13 between the abilities of the first type of object in the perception dimension and the abilities of the first type of object in the attention and memory dimensions, and the similarity b12 and b13 between the abilities of the second type of object in the perception dimension and the abilities of the second type of object in the attention and memory dimensions, calculate the average similarity between the abilities in the perception dimension and the abilities in the attention and memory dimensions. For example, calculate the harmonic mean of a12 and a13 and b12 and b13 to obtain the average similarity N1 between the abilities in the perception dimension and the abilities in the attention and memory dimensions.

[0113]

[0114] Then, the product of N1 and the difference W1 between the abilities of the first type of object and the second type of object in the perceptual dimension is determined as the correlation K1 between the perceptual dimension and reading ability:

[0115] K1 = N1 * W1

[0116] Similarly, following the above processing, we can calculate the correlation K2 between attention dimension and reading ability, and the correlation K3 between memory dimension and reading ability.

[0117] When assessing a subject's reading ability, the subject performs the aforementioned assessment task system for evaluating reading ability. After the subject completes each task, an assessment score A1 is obtained for the corresponding perception dimension, an assessment score A2 for the corresponding attention dimension, and an assessment score A3 for the corresponding memory dimension.

[0118] Then, based on the assessment scores A1 for the corresponding perception dimension, A2 for the corresponding attention dimension, and A3 for the corresponding memory dimension, as well as the correlation K1 between the perception dimension and reading ability, the correlation K2 between the attention dimension and reading ability, and the correlation K3 between the memory dimension and reading ability, the assessment subject is classified into reading ability into either the first type of subject or the second type of subject.

[0119] For example, the evaluation scores A1 corresponding to the perception dimension, A2 corresponding to the attention dimension, and A3 corresponding to the memory dimension of the evaluation object, as well as the correlation K1 between the perception dimension and reading ability, the correlation K2 between the attention dimension and reading ability, and the correlation K3 between the memory dimension and reading ability, are input into a pre-trained reading ability classification model to classify the evaluated object into a first-type object or a second-type object.

[0120] Alternatively, when the aforementioned assessment results are scores for each corresponding assessment dimension, the correlation K1 between the perception dimension and reading ability, the correlation K2 between the attention dimension and reading ability, and the correlation K3 between the memory dimension and reading ability are used as assessment weights for the perception dimension, attention dimension, and memory dimension, respectively. Based on the assessment weights for each of the perception dimension, attention dimension, and memory dimension obtained through the above processing, and the aforementioned assessment scores, the comprehensive reading ability assessment score for the assessed object can be calculated. Then, based on the comprehensive reading ability assessment score of the assessed object, it is classified into either a first-type object or a second-type object.

[0121] For example, based on the evaluation weights corresponding to each evaluation dimension and the scores of each evaluation, the scores of each evaluation are weighted and summed, and then the average value is calculated as the comprehensive evaluation score.

[0122] For example, for the assessment scores of the above-mentioned subject in the perception, attention, and memory dimensions, the average of the weighted sums of the three is calculated according to the following formula, which is taken as the subject's comprehensive reading ability assessment score A:

[0123] A = (A1K1 + A2K2 + A3K3) / 3

[0124] The overall assessment score A indicates the level of reading ability of the person being assessed. A higher A value indicates stronger reading ability, and vice versa. Based on the overall assessment score A, it can be determined whether the person's reading ability is normal or abnormal, thus classifying them into either Type I or Type II.

[0125] As an exemplary implementation, this application embodiment scores the evaluation data through various evaluation dimensions when obtaining various evaluation results, thereby obtaining the evaluation results corresponding to each evaluation dimension.

[0126] The aforementioned evaluation data includes the voice data and / or behavioral data of the evaluated object when performing the evaluation task.

[0127] The aforementioned voice data includes voice data of the evaluated subject answering questions or reading certain content in response to the evaluation task requirements when performing the evaluation task.

[0128] The aforementioned behavioral data includes the operational behavior data of the evaluated object in response to the requirements of the evaluation task.

[0129] Based on the above evaluation data, the evaluation results for each evaluation dimension can be obtained by scoring each dimension.

[0130] As an optional implementation method, the embodiments of this application first extract the evaluation features corresponding to each evaluation dimension from the evaluation data.

[0131] For example, in reading ability assessment, corresponding assessment features are extracted for each assessment dimension.

[0132] For the perception dimension, the main focus is on the measurement of cognitive dimensions. Information such as the accuracy and response time of figural reasoning and elimination tasks can be extracted from the evaluation data as evaluation features. In addition, information such as the accuracy and response time of ranking and discrimination of different targets in auditory tasks can also be extracted as evaluation features.

[0133] For the attention dimension, the main focus is on time-related metrics. Information such as reaction time for a quick naming task and pause time between words can be extracted from the evaluation data as evaluation features.

[0134] For the memory dimension, the main focus is on the indicators of literacy. Information such as the scores of high and low frequency words, as well as the accuracy of initials, finals, and tones, can be extracted from the evaluation data as evaluation features.

[0135] Then, a correlation analysis is performed on the evaluation features corresponding to each evaluation dimension, and based on the correlation between the evaluation features corresponding to each evaluation dimension, the evaluation features corresponding to each evaluation dimension are divided into evaluation feature classes.

[0136] Finally, by scoring the evaluation feature classes corresponding to each evaluation dimension, the evaluation scores for each evaluation dimension are obtained.

[0137] Specifically, for each evaluation dimension, this application embodiment constructs a comprehensive scoring system using exploratory factor analysis. Under this system, evaluation features are automatically classified, and scoring is performed using the classified clustering features.

[0138] See Figure 3 As shown, for a certain evaluation dimension, assuming that evaluation features 1-N are extracted, the scoring system will first perform correlation analysis on the N evaluation features, cluster the evaluation features with strong correlation into the same evaluation feature class, and then score each evaluation feature class to obtain the evaluation result corresponding to the dimension.

[0139] Following the above processing, the evaluation features corresponding to the three evaluation dimensions of perception, attention, and memory are processed separately to obtain the evaluation scores corresponding to each of the three evaluation dimensions.

[0140] As an optional implementation, embodiments of this application pre-train a speech evaluation model to perform phoneme-level error detection and scoring processing on the speech data of the evaluated object when performing an evaluation task, and to determine the evaluation features contained in the speech data for evaluating specific aspects of ability.

[0141] Taking reading ability assessment as an example, based on the assessment task system for assessing reading ability described in the above embodiments, a large number of assessment subjects are organized to perform the assessment task extraction, and the voice data of these assessment subjects when performing the assessment task system is obtained as voice data samples.

[0142] Then, the annotation personnel annotate these speech data samples, mainly annotating the corresponding timestamp information at the initials, finals, tones, and word levels.

[0143] Secondly, using these labeled speech data samples, the transformer-based speech evaluation model was fine-tuned and optimized to perform phoneme-level error detection and scoring on the input speech data. After training, the model possesses the ability to perform phoneme error detection and oral scoring suitable for reading comprehension assessment.

[0144] In the above processing, when the evaluation data is the speech data of the evaluated object performing the evaluation task, the speech data of the evaluated object performing the evaluation task is input into the above speech evaluation model to obtain the evaluation features corresponding to each evaluation dimension. These evaluation features can correspond to any evaluation dimension and can represent the processed information through speech evaluation, such as high and low frequency word scores, initial and final accuracy, tone accuracy, and the accuracy and time of sorting and distinguishing different targets in auditory tasks.

[0145] Exemplary device

[0146] Corresponding to the object evaluation method described above, this application also provides an object evaluation apparatus, see [link to relevant documentation]. Figure 4 As shown, the device includes:

[0147] The data acquisition unit 100 is used to acquire various evaluation results. The various evaluation results are determined by evaluating the ability of the evaluated object in various evaluation dimensions. The various evaluation dimensions are various evaluation dimensions used to evaluate specific aspects of ability.

[0148] The calculation processing unit 110 is configured to determine the correlation between each evaluation dimension and the specific aspect ability based on the ability difference degree of different types of objects in the same evaluation dimension in each preset multiple types of objects, and the ability similarity degree of the same types of objects in different evaluation dimensions in each preset multiple types of objects; wherein, the specific aspect ability of the same types of objects in the preset multiple types of objects is the same, and the specific aspect ability of different types of objects in the preset multiple types of objects is different.

[0149] The classification processing unit 120 is used to classify the evaluated object into one of the preset multiple types of objects based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of the ability.

[0150] As an optional implementation, when the different types of objects include a first type of object and a second type of object, the step of determining the correlation between each evaluation dimension and the specific aspect capability based on the capability difference degree of different types of objects in the same evaluation dimension in each preset multiple types of objects, and the capability similarity degree of the same type of objects in different evaluation dimensions in each preset multiple types of objects, includes:

[0151] Based on the difference in ability between the first type of object and the second type of object in the same evaluation dimension in each evaluation dimension, the similarity in ability of the first type of object in each evaluation dimension, and the similarity in ability of the second type of object in each evaluation dimension, the correlation between each evaluation dimension and the specific aspect of ability is calculated respectively.

[0152] The specific capabilities of the first type of object differ from those of the second type of object.

[0153] As an optional implementation, based on the capability difference of the first type of object and the second type of object in the same evaluation dimension in each evaluation dimension, the capability similarity of the first type of object in each evaluation dimension, and the capability similarity of the second type of object in each evaluation dimension, the correlation degree between each evaluation dimension and the specific aspect capability is calculated, including:

[0154] For each evaluation dimension, the correlation between the evaluation dimension and the specific aspect of the ability is calculated based on the similarity between the ability of the first type of object in this evaluation dimension and the ability of the first type of object in other evaluation dimensions, the similarity between the ability of the second type of object in this evaluation dimension and the ability of the second type of object in other evaluation dimensions, and the difference between the abilities of the first type of object and the second type of object in this evaluation dimension.

[0155] As an optional implementation, the correlation between the evaluation dimension and the specific aspect capability is calculated based on the similarity between the capabilities of the first type of object in this evaluation dimension and the capabilities of the first type of object in other evaluation dimensions, the similarity between the capabilities of the second type of object in this evaluation dimension and the capabilities of the second type of object in other evaluation dimensions, and the difference between the capabilities of the first type of object and the second type of object in this evaluation dimension. This includes:

[0156] Based on the similarity between the ability of the first type of object in this evaluation dimension and the ability of the first type of object in other evaluation dimensions, and the similarity between the ability of the second type of object in this evaluation dimension and the ability of the second type of object in other evaluation dimensions, calculate and determine the average similarity between the ability in this evaluation dimension and the ability in other evaluation dimensions.

[0157] The correlation between the assessment dimension and the specific aspect ability is determined by multiplying the average similarity between the ability in this assessment dimension and the ability in other assessment dimensions, and the difference between the abilities of the first type of object and the second type of object in this assessment dimension.

[0158] As an optional implementation, obtaining the various evaluation results includes:

[0159] The evaluation data is scored by various evaluation dimensions to obtain the evaluation results for each evaluation dimension; wherein, the evaluation data includes the voice data and / or behavioral data of the evaluated object when performing the evaluation task.

[0160] As an optional implementation, the step of scoring the evaluation data through various evaluation dimensions to obtain evaluation results for each evaluation dimension includes:

[0161] Evaluation features corresponding to each evaluation dimension were extracted from the evaluation data.

[0162] A correlation analysis was conducted on the evaluation features corresponding to each evaluation dimension, and based on the correlation between the evaluation features corresponding to each evaluation dimension, the evaluation features corresponding to each evaluation dimension were divided into evaluation feature classes.

[0163] By scoring the evaluation feature classes corresponding to each evaluation dimension, the evaluation results for each evaluation dimension are obtained.

[0164] As an optional implementation, when the evaluation data is the speech data of the evaluated object performing the evaluation task, the step of extracting evaluation features corresponding to each evaluation dimension from the evaluation data includes:

[0165] The evaluation data is input into a pre-trained speech evaluation model to obtain evaluation features corresponding to each evaluation dimension.

[0166] The speech evaluation model is used to perform phoneme-level error detection and scoring on the speech data of the evaluated object when it performs the evaluation task, and to determine the evaluation features contained in the speech data for evaluating the specific aspects of the ability.

[0167] As an optional implementation, there are multiple evaluation tasks, wherein the multiple evaluation tasks perform capability evaluation on each evaluation dimension.

[0168] As an optional implementation, classifying the evaluated object into one of the preset multiple object types based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect capability includes:

[0169] Based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of the ability, a comprehensive evaluation result for the specific aspect of the evaluated object is determined.

[0170] Based on the comprehensive evaluation results of the specific aspects of the evaluated object, the evaluated object is classified into one of the preset multiple types of objects.

[0171] As an optional implementation, the various evaluation dimensions include at least two of the following: perception dimension, attention dimension, and memory dimension;

[0172] The specific abilities mentioned include reading ability.

[0173] As an optional implementation, the capability difference between the first type of object and the second type of object in the same evaluation dimension across all evaluation dimensions is determined through the following process:

[0174] Obtain the evaluation scores for each evaluation dimension for a set number of first-type objects and a set number of second-type objects, respectively;

[0175] For each evaluation dimension, the difference between the evaluation score distribution of the set number of first-type objects and the evaluation score distribution of the set number of second-type objects is calculated to obtain the capability difference degree of the first-type objects and the second-type objects in the same evaluation dimension in each evaluation dimension.

[0176] As an optional implementation, the ability similarity of the first type of object in each evaluation dimension and the ability similarity of the second type of object in each evaluation dimension are determined by the following process:

[0177] Obtain the evaluation scores for each evaluation dimension for a set number of first-type objects and a set number of second-type objects, respectively;

[0178] The ability similarity of the first type of objects in each evaluation dimension is determined by calculating the similarity of their evaluation scores for each evaluation dimension.

[0179] Furthermore, by calculating the similarity of the evaluation scores of each second type of object to each evaluation dimension, the ability similarity of the second type of object in each evaluation dimension is determined.

[0180] The object evaluation apparatus provided in this embodiment belongs to the same concept as the object evaluation method provided in the above embodiments of this application. It can execute the object evaluation method provided in any of the above embodiments of this application and has the corresponding functional modules and beneficial effects of the method. Technical details not described in detail in this embodiment can be found in the specific processing content of the object evaluation method provided in the above embodiments of this application, and will not be repeated here.

[0181] Exemplary electronic devices

[0182] Another embodiment of this application also provides an electronic device, see [link to relevant documentation] Figure 5 As shown, the device includes:

[0183] Memory 200 and processor 210;

[0184] The memory 200 is connected to the processor 210 and is used to store programs;

[0185] The processor 210 is used to implement the object evaluation method disclosed in any of the above embodiments by running the program stored in the memory 200.

[0186] Specifically, the aforementioned electronic device may also include: a bus, a communication interface 220, an input device 230, and an output device 240.

[0187] The processor 210, memory 200, communication interface 220, input device 230, and output device 240 are interconnected via a bus. Among them:

[0188] A bus can include a pathway for transmitting information between various components of a computer system.

[0189] The processor 210 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0190] Processor 210 may include a main processor, as well as a baseband chip, modem, etc.

[0191] The memory 200 stores a program that executes the technical solution of this invention, and may also store an operating system and other key business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory 200 may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0192] Input device 230 may include a device for receiving user input data and information, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor.

[0193] Output device 240 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.

[0194] Output device 240 can be used to output the graphical interface of the evaluation task and the audio data of the evaluation task, for example, displaying the graphical interface of the evaluation task on a screen and outputting the audio of the evaluation task through a speaker; input device 230 can be used to collect evaluation data when the evaluated object performs the evaluation task, the evaluation data including the voice data and / or behavioral data of the evaluated object when performing the evaluation task, for example, collecting the behavioral data of the evaluated object operating on the touch screen when performing the evaluation task through the touch module of the touch screen, and collecting the voice data of the evaluated object when performing the evaluation task through a microphone (voice input device).

[0195] The communication interface 220 may include a device that uses any transceiver to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0196] The processor 210 executes the program stored in the memory 200 and calls other devices, which can be used to implement the various steps of any of the object evaluation methods provided in the above embodiments of this application.

[0197] Another embodiment of this application also proposes a capability assessment device, see [link to relevant documentation]. Figure 6 As shown, the device includes:

[0198] Input / output device 300, and processor 310 connected to said input / output device;

[0199] The input / output device 300 is used to output the graphical interface of the evaluation task and the audio data of the evaluation task, and to collect the evaluation data when the evaluated object performs the evaluation task, the evaluation data including the voice data and / or behavioral data of the evaluated object when performing the evaluation task;

[0200] The processor 310 is configured to score the evaluation data through various evaluation dimensions to obtain evaluation results corresponding to each evaluation dimension; each evaluation dimension is a different evaluation dimension used to evaluate specific aspects of ability; based on the difference in ability of different types of objects in the same evaluation dimension among the various preset types of objects, and the similarity in ability of the same types of objects in different evaluation dimensions among the various preset types of objects, the correlation between each evaluation dimension and the specific aspect of ability is determined respectively; wherein, the specific aspect of ability of the same types of objects in the various preset types of objects is the same, and the specific aspect of ability of different types of objects in the various preset types of objects is different; based on the evaluation results and the correlation between each evaluation dimension and the specific aspect of ability, the evaluated object is classified into one type of object among the various preset types of objects.

[0201] As an optional implementation, the input / output device 300 includes a touch screen, a microphone, and a speaker;

[0202] The touch screen is used to display the graphical interface of the evaluation task and to collect behavioral data of the evaluated object on the touch screen when performing the evaluation task.

[0203] The speaker is used to output the audio data of the evaluation task;

[0204] The microphone is used to collect voice data of the evaluated object when it performs the evaluation task.

[0205] As an optional implementation, when the different types of objects include a first type of object and a second type of object, the step of determining the correlation between each evaluation dimension and the specific aspect capability based on the capability difference degree of different types of objects in the same evaluation dimension in each preset multiple types of objects, and the capability similarity degree of the same type of objects in different evaluation dimensions in each preset multiple types of objects, includes:

[0206] Based on the difference in ability between the first type of object and the second type of object in the same evaluation dimension in each evaluation dimension, the similarity in ability of the first type of object in each evaluation dimension, and the similarity in ability of the second type of object in each evaluation dimension, the correlation between each evaluation dimension and the specific aspect of ability is calculated respectively.

[0207] The specific capabilities of the first type of object differ from those of the second type of object.

[0208] As an optional implementation, based on the capability difference of the first type of object and the second type of object in the same evaluation dimension in each evaluation dimension, the capability similarity of the first type of object in each evaluation dimension, and the capability similarity of the second type of object in each evaluation dimension, the correlation degree between each evaluation dimension and the specific aspect capability is calculated, including:

[0209] For each evaluation dimension, the correlation between the evaluation dimension and the specific aspect of the ability is calculated based on the similarity between the ability of the first type of object in this evaluation dimension and the ability of the first type of object in other evaluation dimensions, the similarity between the ability of the second type of object in this evaluation dimension and the ability of the second type of object in other evaluation dimensions, and the difference between the abilities of the first type of object and the second type of object in this evaluation dimension.

[0210] As an optional implementation, the correlation between the evaluation dimension and the specific aspect capability is calculated based on the similarity between the capabilities of the first type of object in this evaluation dimension and the capabilities of the first type of object in other evaluation dimensions, the similarity between the capabilities of the second type of object in this evaluation dimension and the capabilities of the second type of object in other evaluation dimensions, and the difference between the capabilities of the first type of object and the second type of object in this evaluation dimension. This includes:

[0211] Based on the similarity between the ability of the first type of object in this evaluation dimension and the ability of the first type of object in other evaluation dimensions, and the similarity between the ability of the second type of object in this evaluation dimension and the ability of the second type of object in other evaluation dimensions, calculate and determine the average similarity between the ability in this evaluation dimension and the ability in other evaluation dimensions.

[0212] The correlation between the assessment dimension and the specific aspect ability is determined by multiplying the average similarity between the ability in this assessment dimension and the ability in other assessment dimensions, and the difference between the abilities of the first type of object and the second type of object in this assessment dimension.

[0213] As an optional implementation, the step of scoring the evaluation data through various evaluation dimensions to obtain evaluation results for each evaluation dimension includes:

[0214] Evaluation features corresponding to each evaluation dimension were extracted from the evaluation data.

[0215] A correlation analysis was conducted on the evaluation features corresponding to each evaluation dimension, and based on the correlation between the evaluation features corresponding to each evaluation dimension, the evaluation features corresponding to each evaluation dimension were divided into evaluation feature classes.

[0216] By scoring the evaluation feature classes corresponding to each evaluation dimension, the evaluation results for each evaluation dimension are obtained.

[0217] As an optional implementation, when the evaluation data is the speech data of the evaluated object performing the evaluation task, the step of extracting evaluation features corresponding to each evaluation dimension from the evaluation data includes:

[0218] The evaluation data is input into a pre-trained speech evaluation model to obtain evaluation features corresponding to each evaluation dimension.

[0219] The speech evaluation model is used to perform phoneme-level error detection and scoring on the speech data of the evaluated object when it performs the evaluation task, and to determine the evaluation features contained in the speech data for evaluating the specific aspects of the ability.

[0220] As an optional implementation, there are multiple evaluation tasks, wherein the multiple evaluation tasks perform capability evaluation on each evaluation dimension.

[0221] As an optional implementation, classifying the evaluated object into one of the preset multiple object types based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect capability includes:

[0222] Based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of the ability, a comprehensive evaluation result for the specific aspect of the evaluated object is determined.

[0223] Based on the comprehensive evaluation results of the specific aspects of the evaluated object, the evaluated object is classified into one of the preset multiple types of objects.

[0224] As an optional implementation, the various evaluation dimensions include at least two of the following: perception dimension, attention dimension, and memory dimension;

[0225] The specific abilities mentioned include reading ability.

[0226] As an optional implementation, the capability difference between the first type of object and the second type of object in the same evaluation dimension across all evaluation dimensions is determined through the following process:

[0227] Obtain the evaluation scores for each evaluation dimension for a set number of first-type objects and a set number of second-type objects, respectively;

[0228] For each evaluation dimension, the difference between the evaluation score distribution of the set number of first-type objects and the evaluation score distribution of the set number of second-type objects is calculated to obtain the capability difference degree of the first-type objects and the second-type objects in the same evaluation dimension in each evaluation dimension.

[0229] As an optional implementation, the ability similarity of the first type of object in each evaluation dimension and the ability similarity of the second type of object in each evaluation dimension are determined by the following process:

[0230] Obtain the evaluation scores for each evaluation dimension for a set number of first-type objects and a set number of second-type objects, respectively;

[0231] The ability similarity of the first type of objects in each evaluation dimension is determined by calculating the similarity of their evaluation scores for each evaluation dimension.

[0232] Furthermore, by calculating the similarity of the evaluation scores of each second type of object to each evaluation dimension, the ability similarity of the second type of object in each evaluation dimension is determined.

[0233] The aforementioned electronic devices and capability assessment devices belong to the same technical concept as the object assessment method provided in the above embodiments of this application. They can execute the object assessment method provided in any of the above embodiments of this application and have the corresponding functional modules and beneficial effects of the method. Technical details not described in detail in this embodiment can be found in the specific processing content of the object assessment method provided in the above embodiments of this application, and will not be repeated here.

[0234] Exemplary computer program products and storage media

[0235] In addition to the methods and devices described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the object evaluation method described in the "Exemplary Methods" section of this specification.

[0236] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0237] Furthermore, embodiments of this application may also be storage media storing a computer program, which is executed by a processor in the steps of the object evaluation method described in the "Exemplary Methods" section of this specification.

[0238] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0239] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For apparatus embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0240] The steps in the methods of the various embodiments of this application can be adjusted, merged, or deleted in order according to actual needs, and the technical features described in each embodiment can be replaced or combined.

[0241] The modules and sub-modules in the various embodiments of the present application's devices and terminals can be merged, divided, and deleted according to actual needs.

[0242] It should be understood that the disclosed terminals, devices, and methods can be implemented in other ways, given the several embodiments provided in this application. For example, the terminal embodiments described above are merely illustrative. For instance, the division of modules or sub-modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple sub-modules or modules may be combined or integrated into another module, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0243] The modules or submodules described as separate components may or may not be physically separate. The components that constitute a module or submodule may or may not be physical modules or submodules; that is, they may be located in one place or distributed across multiple network modules or submodules. Some or all of the modules or submodules can be selected to achieve the purpose of this embodiment's solution, depending on actual needs.

[0244] Furthermore, the functional modules or sub-modules in the various embodiments of this application can be integrated into one processing module, or each module or sub-module can exist physically separately, or two or more modules or sub-modules can be integrated into one module. The integrated modules or sub-modules described above can be implemented in hardware or in the form of software functional modules or sub-modules.

[0245] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0246] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software unit executed by a processor, or a combination of both. The software unit can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0247] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0248] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An object evaluation method, characterized in that, include: The evaluation data is scored through various evaluation dimensions to obtain the evaluation results for each evaluation dimension. The evaluation data includes the voice data and / or behavioral data of the evaluated object when performing the evaluation task. Each evaluation dimension is a different evaluation dimension used to evaluate a specific aspect of ability. The evaluated object is selected from at least one of the following: living organism, robot, vehicle, moving mechanical parts, and neural network language model. The specific aspect of ability is selected from at least one of the following: language ability, reasoning ability, reading ability, and motor ability. Based on the ability differences of different types of objects in the same evaluation dimension across various evaluation dimensions, and the ability similarity of the same types of objects in different evaluation dimensions across various evaluation dimensions, the correlation between each evaluation dimension and the specific aspect ability is determined; wherein, the specific aspect ability of the same types of objects in the preset multiple types of objects is the same, and the specific aspect ability of different types of objects in the preset multiple types of objects is different; the ability similarity of the same types of objects in different evaluation dimensions across various evaluation dimensions is used to reflect the ability similarity of a group with the same specific aspect ability across various evaluation dimensions; Based on the evaluation results and the correlation between each evaluation dimension and the specific aspect of capability, the evaluated object is classified into one of the preset multiple types of objects.

2. The method according to claim 1, characterized in that, When the different types of objects include a first type of object and a second type of object, the step of determining the correlation between each evaluation dimension and the specific aspect of the ability based on the ability difference degree of different types of objects in the same evaluation dimension in each preset multiple types of objects, and the ability similarity degree of the same types of objects in different evaluation dimensions in each preset multiple types of objects, includes: Based on the difference in ability between the first type of object and the second type of object in the same evaluation dimension in each evaluation dimension, the similarity in ability of the first type of object in each evaluation dimension, and the similarity in ability of the second type of object in each evaluation dimension, the correlation between each evaluation dimension and the specific aspect of ability is calculated respectively. The specific capabilities of the first type of object differ from those of the second type of object.

3. The method according to claim 2, characterized in that, Based on the difference in capabilities of the first type of object and the second type of object in the same evaluation dimensions across each evaluation dimension, the similarity in capabilities of the first type of object across each evaluation dimension, and the similarity in capabilities of the second type of object across each evaluation dimension, the correlation between each evaluation dimension and the specific aspect capability is calculated, including: For each evaluation dimension, the correlation between the evaluation dimension and the specific aspect of the ability is calculated based on the similarity between the ability of the first type of object in this evaluation dimension and the ability of the first type of object in other evaluation dimensions, the similarity between the ability of the second type of object in this evaluation dimension and the ability of the second type of object in other evaluation dimensions, and the difference between the abilities of the first type of object and the second type of object in this evaluation dimension.

4. The method according to claim 3, characterized in that, Based on the similarity between the capabilities of the first type of object in this evaluation dimension and its capabilities in other evaluation dimensions, the similarity between the capabilities of the second type of object in this evaluation dimension and its capabilities in other evaluation dimensions, and the difference between the capabilities of the first type of object and the second type of object in this evaluation dimension, the correlation between this evaluation dimension and the specific aspect capability is calculated, including: Based on the similarity between the ability of the first type of object in this evaluation dimension and the ability of the first type of object in other evaluation dimensions, and the similarity between the ability of the second type of object in this evaluation dimension and the ability of the second type of object in other evaluation dimensions, calculate and determine the average similarity between the ability in this evaluation dimension and the ability in other evaluation dimensions. The correlation between the assessment dimension and the specific aspect ability is determined by multiplying the average similarity between the ability in this assessment dimension and the ability in other assessment dimensions, and the difference between the abilities of the first type of object and the second type of object in this assessment dimension.

5. The method according to claim 1, characterized in that, The evaluation data is scored according to various evaluation dimensions to obtain the evaluation results for each corresponding evaluation dimension, including: Evaluation features corresponding to each evaluation dimension were extracted from the evaluation data. A correlation analysis was conducted on the evaluation features corresponding to each evaluation dimension, and based on the correlation between the evaluation features corresponding to each evaluation dimension, the evaluation features corresponding to each evaluation dimension were divided into evaluation feature classes. By scoring the evaluation feature classes corresponding to each evaluation dimension, the evaluation results for each evaluation dimension are obtained.

6. The method according to claim 5, characterized in that, When the evaluation data is the speech data of the evaluated object when performing the evaluation task, the step of extracting evaluation features corresponding to each evaluation dimension from the evaluation data includes: The evaluation data is input into a pre-trained speech evaluation model to obtain evaluation features corresponding to each evaluation dimension. The speech evaluation model is used to perform phoneme-level error detection and scoring on the speech data of the evaluated object when it performs the evaluation task, and to determine the evaluation features contained in the speech data for evaluating the specific aspects of the ability.

7. The method according to claim 1, characterized in that, The number of evaluation tasks is multiple, and the multiple evaluation tasks perform capability evaluation on each evaluation dimension.

8. The method according to claim 1, characterized in that, The step of classifying the evaluated object into one of the preset multiple object types based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of ability includes: Based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of the ability, a comprehensive evaluation result for the specific aspect of the evaluated object is determined. Based on the comprehensive evaluation results of the specific aspects of the evaluated object, the evaluated object is classified into one of the preset multiple types of objects.

9. The method according to any one of claims 1 to 8, characterized in that, Each of the assessment dimensions includes at least two of the following: perception dimension, attention dimension, and memory dimension; The specific abilities mentioned include reading ability.

10. The method according to any one of claims 2 to 8, characterized in that, The capability differences between the first type of object and the second type of object in the same evaluation dimension across all evaluation dimensions are determined through the following process: Obtain the evaluation scores for each evaluation dimension for a set number of first-type objects and a set number of second-type objects, respectively; For each evaluation dimension, the difference between the evaluation score distribution of the set number of first-type objects and the evaluation score distribution of the set number of second-type objects is calculated to obtain the capability difference degree of the first-type objects and the second-type objects in the same evaluation dimension in each evaluation dimension.

11. The method according to any one of claims 2 to 8, characterized in that, The ability similarity of the first type of object across each evaluation dimension, and the ability similarity of the second type of object across each evaluation dimension, are determined through the following processing: Obtain the evaluation scores for each evaluation dimension for a set number of first-type objects and a set number of second-type objects, respectively; The ability similarity of the first type of objects in each evaluation dimension is determined by calculating the similarity of their evaluation scores for each evaluation dimension. Furthermore, by calculating the similarity of the evaluation scores of each second type of object to each evaluation dimension, the ability similarity of the second type of object in each evaluation dimension is determined.

12. An object evaluation device, characterized in that, include: The data acquisition unit is used to score the evaluation data through various evaluation dimensions to obtain the evaluation results for each evaluation dimension. The evaluation data includes voice data and / or behavioral data of the evaluated object when performing the evaluation task. Each evaluation dimension is a dimension used to evaluate specific aspects of ability. The evaluated object is selected from at least one of the following: living organism, robot, vehicle, moving mechanical parts, and neural network language model. The specific aspect of ability is selected from at least one of language ability, reasoning ability, reading ability, and motor ability. The calculation and processing unit is used to determine the correlation between each evaluation dimension and the specific aspect ability based on the ability difference degree of different types of objects in the same evaluation dimension in each preset multiple types of objects, and the ability similarity of the same types of objects in different evaluation dimensions in each preset multiple types of objects; wherein, the specific aspect ability of the same types of objects in the preset multiple types of objects is the same, and the specific aspect ability of different types of objects in the preset multiple types of objects is different; the ability similarity of the same types of objects in different evaluation dimensions in each preset multiple types of objects is used to reflect the ability similarity of a group with the same specific aspect ability in each different evaluation dimension; The classification processing unit is used to classify the evaluated object into one of the preset multiple types of objects based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect of the capability.

13. An electronic device, characterized in that, include: Memory and processor; The memory is connected to the processor and is used to store computer programs; The processor is configured to implement the object evaluation method as described in any one of claims 1 to 11 by running a program in the memory.

14. A capability assessment device, characterized in that, include: Input / output devices, and a processor connected to said input / output devices; The input / output device is used to output the graphical interface of the evaluation task and the audio data of the evaluation task, and to collect evaluation data when the evaluated object performs the evaluation task. The evaluation data includes the speech data and / or behavioral data of the evaluated object when performing the evaluation task. The evaluated object is selected from at least one of the following: organism, robot, vehicle, moving mechanical parts, and neural network language model. The processor is configured to score the evaluation data through various evaluation dimensions to obtain evaluation results for each evaluation dimension; each evaluation dimension is a dimension used to evaluate specific aspects of ability; the specific aspects of ability are selected from at least one of language ability, reasoning ability, reading ability, and motor ability; based on the ability difference of different types of objects in the same evaluation dimension in each of the preset multiple types of objects, and the ability similarity of the same types of objects in different evaluation dimensions in each of the preset multiple types of objects, the processor determines the relationship between each evaluation dimension and the specific aspects of ability. The correlation between them; wherein, objects of the same type among the preset multiple types of objects have the same specific aspect ability, and objects of different types among the preset multiple types of objects have different specific aspect abilities; the ability similarity of objects of the same type among the preset multiple types of objects in different evaluation dimensions is used to reflect the ability similarity of groups with the same specific aspect ability in different evaluation dimensions; based on the evaluation results and the correlation between the evaluation dimensions and the specific aspect ability, the evaluated object is classified into one type of object among the preset multiple types of objects.

15. The device according to claim 14, characterized in that, The input / output devices include a touch screen, a microphone, and a speaker; The touch screen is used to display the graphical interface of the evaluation task and to collect behavioral data of the evaluated object on the touch screen when performing the evaluation task. The speaker is used to output the audio data of the evaluation task; The microphone is used to collect voice data of the evaluated object when it performs the evaluation task.

16. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the object evaluation method as described in any one of claims 1 to 11.

Images