A review text generation method, device and electronic device

Abstract

The present invention belongs to the field of education, and provides a method, device and electronic device for generating comment text. The method comprises: establishing a comment corpus, the comment corpus comprises multi-level comment tags, each level of comment tags comprises multiple comment tags and multiple basic corpora corresponding to each comment tag, each basic corpus comprises positive appositive phrases and negative appositive phrases; receiving teacher input, the teacher input comprises selecting multiple comment tags; based on the teacher input, using the comment corpus, performing level label matching; according to the matching result, calling the corresponding generation template, and automatically generating text evaluation content. The present invention can realize automatic or editable generation of text comment content, and at the same time provide teachers with an editable function for real-time modification, which can effectively help teachers quickly correct online homework and reduce working time.

Description

Technical Field

[0001] The present invention belongs to the field of education, and is particularly applicable to the field of online education. More specifically, it relates to a review text generation method, device and electronic equipment. Background Art

[0002] With the development of the Internet, more and more online courses have emerged, and teachers imparting knowledge through online lectures or online classes has become an important way of learning.

[0003] However, in online education, the workload of each teacher increases with the number of students. The workload of each teacher for commenting or grading every day is particularly large, such as commenting on submitted notes, commenting on online classroom performance, and correcting homework to be corrected within a specific time (mainly online homework or homework submitted online), etc., and these tasks all have timeliness issues. To complete the above work, teachers need to spend a lot of time and even a lot of energy. Therefore, teachers engaged in online education usually have problems with heavy workload and long time consumption, and even cannot complete the review or correction work in a timely and effective manner.

[0004] Therefore, it is necessary to provide a more effective review text generation method to solve the above problems. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] The present invention aims to solve the problem that teachers engaged in online education have a heavy online workload and a long time consumption, and even cannot complete online review or correction work in a timely and effective manner.

[0007] (II) Technical solution

[0008] To solve the above technical problems, one aspect of the present invention proposes a comment text generation method, which is used for teacher comments in online education APPs, and the method comprises the following steps: establishing a comment corpus, the comment corpus includes multi-level comment labels, each level of comment labels includes multiple comment labels and multiple basic corpora corresponding to each comment label, each basic corpus includes positive appositive phrases and negative appositive phrases; receiving teacher input, the teacher input includes selecting multiple comment labels; based on the teacher input, using the comment corpus, performing level label matching; according to the matching results, calling the corresponding generation template to automatically generate text evaluation content.

[0009] According to a preferred embodiment of the present invention, it also includes: according to generation rules, presetting generation templates associated with comment labels at various levels; the generation rules include setting generation parameters, automatically or editably generating evaluation content, and the generation parameters include text generation scenarios, course business types, text generation length, writing parameters, timeliness parameters, completion structure, whether the submitted homework content is complete, whether there are pictures, and overall review information.

[0010] According to a preferred embodiment of the present invention, it also includes: the comment labels at each level include automatically generated labels corresponding to writing parameters, timeliness parameters, whether the submitted homework content is complete, whether there are pictures, and overall review information; corresponding evaluation content is automatically generated according to the automatically generated labels in the teacher's input data.

[0011] According to a preferred embodiment of the present invention, the method further includes: generating one or more basic sentences for each basic corpus; parsing and expanding the basic corpus containing multiple appositive phrases using the Cartesian product method, and determining the generation of a corresponding number of basic sentences based on the number of appositive phrases and the number of appositive words.

[0012] According to a preferred embodiment of the present invention, the level label matching includes: detecting whether there is a first-level review label in the current teacher input data, and whether there is a second-level review label belonging to the first-level review label; judging the good or bad tendency of each level of labels, and returning the predetermined generation result.

[0013] According to a preferred embodiment of the present invention, it also includes: level tags are used to indicate the hierarchy of the evaluation content structure, wherein the first-level review tag is used to indicate the top layer of the evaluation content structure, and the first-level review tag corresponds to the text generated by splicing the review texts generated by at least two second-level review tags; the second-level review tag is used to indicate the stratification of the evaluation content structure, and the second-level review tag corresponds to the generation of a basic sentence.

[0014] According to a preferred embodiment of the present invention, it also includes: determining multiple corresponding labels based on the level label matching results; when automatically selecting basic sentences in the corresponding labels, extracting each basic sentence with equal probability, splicing and combining the extracted basic sentences, and combining them to generate a template to generate text evaluation content corresponding to the teacher's input.

[0015] According to a preferred embodiment of the present invention, it also includes: adjusting the generated text evaluation content, and the adjustment processing includes adjusting the extraction weight containing the words according to the number of occurrences or frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

[0016] The second aspect of the present invention proposes a comment text generation device, which is used for teacher comments in online education APPs, and the device includes: an establishment module, which is used to establish a comment corpus, and the comment corpus includes multi-level comment tags, each level of comment tags includes multiple comment tags and multiple basic corpora corresponding to each comment tag, each basic corpus includes positive appositive phrases and negative appositive phrases; a receiving module, which is used to receive teacher input, and the teacher input includes selecting multiple comment tags; a processing module, based on the teacher input, uses the comment corpus to perform level label matching; an automatic generation module, according to the matching results, calls the corresponding generation template to automatically generate text evaluation content.

[0017] According to a preferred embodiment of the present invention, it also includes a setting module, which presets generation templates associated with comment labels at various levels according to generation rules; the generation rules include setting generation parameters, automatically or editably generating evaluation content, and the generation parameters include text generation scenarios, course business types, text generation length, writing parameters, timeliness parameters, completion structure, whether the submitted homework content is complete, whether there are pictures, and overall review information.

[0018] According to a preferred embodiment of the present invention, it also includes: the comment labels at each level include automatically generated labels corresponding to writing parameters, timeliness parameters, whether the submitted homework content is complete, whether there are pictures, and overall review information; corresponding evaluation content is automatically generated according to the automatically generated labels in the teacher's input data.

[0019] According to a preferred embodiment of the present invention, the method further includes: generating one or more basic sentences for each basic corpus; parsing and expanding the basic corpus containing multiple appositive phrases using the Cartesian product method, and determining the generation of a corresponding number of basic sentences based on the number of appositive phrases and the number of appositive words.

[0020] According to a preferred embodiment of the present invention, it also includes a detection module, which is used to detect whether there is a first-level review label in the current teacher input data, whether there is a second-level review label belonging to the first-level review label; judge the good or bad tendency of each level of label, and return the predetermined generation result.

[0021] According to a preferred embodiment of the present invention, it also includes: level tags are used to indicate the hierarchy of the evaluation content structure, wherein the first-level review tag is used to indicate the top layer of the evaluation content structure, and the first-level review tag corresponds to the text generated by splicing the review texts generated by at least two second-level review tags; the second-level review tag is used to indicate the stratification of the evaluation content structure, and the second-level review tag corresponds to the generation of a basic sentence.

[0022] According to a preferred embodiment of the present invention, it also includes: determining multiple corresponding labels based on the level label matching results; when automatically selecting basic sentences in the corresponding labels, extracting each basic sentence with equal probability, splicing and combining the extracted basic sentences, and combining them to generate a template to generate text evaluation content corresponding to the teacher's input.

[0023] According to a preferred embodiment of the present invention, it also includes: adjusting the generated text evaluation content, and the adjustment processing includes adjusting the extraction weight containing the words according to the number of occurrences or frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

[0024] A third aspect of the present invention provides an electronic device, comprising a processor and a memory, wherein the memory is used to store a computer executable program, and when the computer program is executed by the processor, the processor executes the review text generation method.

[0025] The fourth aspect of the present invention further proposes a computer-readable medium storing a computer-executable program, and when the computer-executable program is executed, the review text generation method is implemented.

[0026] (III) Beneficial effects

[0027] Compared with the prior art, the present invention can ensure the diversity, richness and structural integrity of comment content generation through the setting of multi-level labels; by receiving label data input by teachers, it can realize automatic or editable generation of text comment content, and provide teachers with editable functions for real-time modification; it can effectively help teachers quickly correct online homework and reduce working time, thereby improving teachers' work efficiency and optimizing the workflow of teachers correcting online homework. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a flowchart of an example of a method for generating a review text according to Embodiment 1 of the present invention;

[0029] Figure 2 is a flowchart of another example of the method for generating a review text according to Embodiment 1 of the present invention;

[0030] Figure 3 It is a flowchart of another example of the method for generating a review text according to Embodiment 1 of the present invention;

[0031] Figure 4 is a schematic diagram of an example of a review text generating device according to Embodiment 2 of the present invention;

[0032] Figure 5 is a schematic diagram of another example of a review text generating device according to Embodiment 2 of the present invention;

[0033] Figure 6 is a schematic diagram of another example of the review text generating device of embodiment 2 of the present invention;

[0034] Figure 7 is a schematic structural diagram of an electronic device according to an embodiment of the present invention;

[0035] Figure 8 is a schematic diagram of a computer-readable recording medium according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In the process of introducing specific embodiments, the detailed description of the structure, performance, effect or other characteristics is to enable those skilled in the art to fully understand the embodiments. However, it does not exclude that those skilled in the art can implement the present invention with a technical solution that does not contain the above-mentioned structure, performance, effect or other characteristics under certain circumstances.

[0037] The flowchart in the accompanying drawings is only an exemplary process demonstration, and does not mean that the solution of the present invention must include all the contents, operations and steps in the flowchart, nor does it mean that it must be executed in the order shown in the figure. For example, some operations / steps in the flowchart can be decomposed, some operations / steps can be combined or partially combined, etc. The execution order shown in the flowchart can be changed according to the actual situation without departing from the main purpose of the present invention.

[0038] Frames in the attached figure Figure 1 Generally, it refers to functional entities, which do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processing unit devices and / or microcontroller devices.

[0039] The same reference numerals in the drawings represent the same or similar elements, components or parts, and thus the repeated description of the same or similar elements, components or parts may be omitted below. It should also be understood that although the first, second, third and other attributives representing the numbers may be used herein to describe various devices, elements, components or parts, these devices, elements, components or parts should not be limited by these attributives. In other words, these attributives are only used to distinguish one from another. For example, the first device may also be called the second device, but it does not deviate from the essential technical solution of the present invention. In addition, the terms "and / or" and "and / or" refer to all combinations including any one or more of the listed items.

[0040] In view of the above problems, the present invention proposes a method for generating comment text, which is particularly suitable for teacher comments in online education APPs. The method receives teacher input through an online platform, uses a comment corpus, performs level label matching, and calls the corresponding generation template according to the matching result to automatically generate text evaluation content. As a result, the method can effectively help teachers quickly correct online homework and reduce working time, thereby improving teachers' work efficiency and optimizing the workflow of teachers correcting online homework.

[0041] Figure 1 This is a flowchart of an example of the review text generation method according to the first embodiment of the present invention.

[0042] like Figure 1 As shown, the review text generation method includes the following steps.

[0043] Step S101, establishing a review corpus, the review corpus includes multi-level review tags, each level of review tags includes multiple review tags and multiple basic corpora corresponding to each review tag, each basic corpus includes positive appositive phrases and negative appositive phrases.

[0044] Step S102, receiving teacher input, wherein the teacher input includes selecting multiple review tags.

[0045] Step S103, based on the teacher input, using the review corpus, level label matching is performed.

[0046] Step S104, based on the matching result, calling the corresponding generation template to automatically generate text evaluation content.

[0047] Preferably, level label matching is performed based on teacher input data and using a review corpus to identify automatically generated labels.

[0048] Furthermore, according to the identified automatically generated tags, the corresponding generation template is called to automatically generate text evaluation content, thereby reducing the teacher's workload of marking and effectively improving the teacher's work efficiency.

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0050] In the process of introducing specific embodiments, the detailed description of the structure, performance, effect or other characteristics is to enable those skilled in the art to fully understand the embodiments. However, it does not exclude that those skilled in the art can implement the present invention with a technical solution that does not contain the above-mentioned structure, performance, effect or other characteristics under certain circumstances.

[0051] The flowchart in the accompanying drawings is only an exemplary process demonstration, and does not mean that the solution of the present invention must include all the contents, operations and steps in the flowchart, nor does it mean that it must be executed in the order shown in the figure. For example, some operations / steps in the flowchart can be decomposed, some operations / steps can be combined or partially combined, etc. The execution order shown in the flowchart can be changed according to the actual situation without departing from the main purpose of the present invention.

[0052] Frames in the attached figure Figure 1 Generally, it refers to functional entities, which do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processing unit devices and / or microcontroller devices.

[0053] The same reference numerals in the drawings represent the same or similar elements, components or parts, and thus the repeated description of the same or similar elements, components or parts may be omitted below. It should also be understood that although the first, second, third and other attributives representing the numbers may be used herein to describe various devices, elements, components or parts, these devices, elements, components or parts should not be limited by these attributives. In other words, these attributives are only used to distinguish one from another. For example, the first device may also be called the second device, but it does not deviate from the essential technical solution of the present invention. In addition, the terms "and / or" and "and / or" refer to all combinations including any one or more of the listed items.

[0054] Figure 2 It is a flowchart of another example of the review text generating method according to the first embodiment of the present invention.

[0055] Next, we will refer to Figure 2 The review text generation method of the present invention is specifically described.

[0056] First, in step S101, a review corpus is established, which includes multi-level review tags. Each level of review tags includes multiple review tags and multiple basic corpora corresponding to each review tag. Each basic corpus includes positive appositive phrases and negative appositive phrases.

[0057] Specifically, multi-level labels are used to represent the hierarchy of the evaluation content structure, thereby ensuring the diversity, richness and structural integrity of the review content generation.

[0058] In this example, the multi-level tags include a first-level review tag and a second-level review tag, but are not limited thereto. In other examples, three or more levels of tags may also be included. The above are only described as preferred examples and should not be construed as limitations on the present invention.

[0059] Specifically, the first-level review tag is used to represent the top level of the evaluation content structure, and the first-level review tag corresponds to the text generated by the concatenation of the review texts generated by at least two second-level review tags. For example, the first-level review tag is represented by comment, and is composed of one or more units. Comment concatenates the review texts generated by all units in order into a text and returns it as the final result.

[0060] Furthermore, the secondary review tag is used to represent the hierarchical structure of the review content, and the secondary review tag generates a basic sentence. For example, the secondary review tag is the basic component unit of the generated template, and each unit may be a Const_String or a Sequence or a Random. The generated result returned by the unit is the generated result returned by the Const_String or the Sequence or the Random.

[0061] It should be noted that Const_String represents a constant string, which is used to configure the fixed text part of the review or the line break symbol of the paragraph quality inspection. The string itself constitutes the generation result returned by Const_String. Sequence and Random are both composed of one or more BaseUnits. Sequence concatenates the text generated by each BaseUnit into a whole in sequence and returns it as its own generation result; while Random concatenates the text generated by each BaseUnit into one in random order and returns it as its own generation result.

[0062] In this example, each level tag needs to be configured with multiple basic corpora, where the number of configured basic corpora is greater than 10.

[0063] Specifically, each basic corpus generates one or more basic sentences.

[0064] Preferably, the Cartesian product method is used to parse and expand the basic corpus containing multiple appositive phrases, and the generation of a corresponding number of basic sentences is determined according to the number of appositive phrases and the number of appositive words.

[0065] Specifically, M appositive phrases are configured in a basic corpus, and each appositive phrase contains N1, N2...NM words respectively. When parsing the corpus, a method similar to the Cartesian product method will be used to generate N1*N2*...*NM basic sentences.

[0066] For example, you can use a variety of marking methods, specifically "our (child|baby)'s (classroom|class) performance is very good". In this example, there are two appositive phrases, and all the words in the appositive phrases can be replaced at will. In the above case, 4 basic sentences can be generated: a. "our child's classroom performance is very good"; b. "our baby's classroom performance is very good"; c. "our child's class performance is very good"; d. "our baby's classroom performance is very good".

[0067] It should be noted that the basic corpus is configured by the business configuration personnel and is a plain text without any special symbols. In addition, the rationality of the appositive phrases is reviewed or screened by the business configuration personnel.

[0068] Next, in step S102, teacher input is received, where the teacher input includes selecting a plurality of review tags.

[0069] For example, in the scenario of correcting students' homework, teacher A's input data is received, and the input data includes a first-level comment label and multiple second-level comment labels, where the first-level comment label is a general comment information label, and the multiple second-level comment labels include a timeliness label corresponding to whether the homework is completed on time, a completion status label corresponding to the completion structure, a writing label corresponding to handwriting, etc.

[0070] It should be noted that the above description is only given as an example and should not be construed as a limitation to the present invention.

[0071] Next, in step S103, based on the teacher input, the review corpus is used to perform level label matching.

[0072] like Figure 2 As shown, it also includes step S201 of real-time monitoring of the teacher's input data.

[0073] In step S201, the teacher input data is monitored in real time to more quickly and effectively identify the level labels in the teacher input data.

[0074] Specifically, it is detected whether there is a first-level review tag in the current teacher input data, and whether there is a second-level review tag belonging to the first-level review tag.

[0075] For example, it is detected that there are first-level comment tags and second-level comment tags in the current teacher input data, and the number of first-level comment tags and their corresponding generated text content, the number of second-level comment tags and their corresponding generated text content, etc. are further identified.

[0076] Furthermore, a preliminary judgment is made on the good or bad tendency of the detected labels at each level for level label matching.

[0077] Specifically, based on the teacher input and according to the above test results, the level labels in the test results are matched with the level labels in the review corpus using the review corpus.

[0078] It should be noted that the above description is only given as an example and should not be construed as a limitation to the present invention.

[0079] Next, in step S104, according to the matching result, the corresponding generation template is called to automatically generate text evaluation content.

[0080] Specifically, after the level labels are matched, a plurality of corresponding review labels matching the teacher input are determined from the review corpus.

[0081] like Figure 3 As shown, the process also includes step S301 of setting generation rules.

[0082] In step S301, generation rules are set to preset generation templates associated with review tags at various levels according to the generation rules.

[0083] Specifically, the generation rules include setting generation parameters, wherein the generation parameters include text generation scenario, course business type, text generation length, writing parameters, timeliness parameters, completion structure, whether the submitted homework content is complete, whether there are illustrations, and overall evaluation information.

[0084] Furthermore, the generation rules also include automatic or editable generation of evaluation content. In other words, by receiving label data input by the teacher, automatic or editable generation of text evaluation content (i.e. text comment content) is achieved, while providing the teacher with an editable function for real-time modification.

[0085] Preferably, the comment labels at each level include automatically generated labels corresponding to writing parameters, timeliness parameters, whether the submitted homework content is complete, whether there are pictures, and overall comment information; the corresponding evaluation content is automatically generated according to the automatically generated labels in the teacher's input data. Therefore, by receiving the label data input by the teacher, it is possible to automatically or editably generate text comment content, and at the same time provide the teacher with an editable function for real-time modification, which can effectively help teachers quickly correct online homework and reduce working time, thereby improving the teacher's work efficiency and optimizing the teacher's workflow for correcting online homework.

[0086] Furthermore, based on the determined multiple corresponding tags, when automatically selecting basic sentences within the corresponding tags, each basic sentence is extracted with equal probability, the extracted basic sentences are spliced ​​and combined, and combined to generate a template to generate text evaluation content corresponding to the teacher's input.

[0087] In this example, the generated template includes the structural hierarchy of the review text content, the positional relationship, and the arrangement or combination of the review text content generated by the review tags at each level.

[0088] Furthermore, when the level labels in the teacher's input data are different, different generation templates will be called in the same generation scenario.

[0089] In another example, the generated text evaluation content is adjusted, and the adjustment process includes adjusting the extraction weight of the words according to the number of occurrences or the frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

[0090] For example, when processing the corpus of each review tag, some repeated words are likely to appear between different sentences, such as "very good", "great", "awesome" and other words that are likely to appear repeatedly in positive reviews. In order to avoid the above situation, the extraction probability of each basic sentence is adjusted.

[0091] Specifically, when the review text content is generated according to the generation template, the number of occurrences of each word in the word set is recorded in real time.

[0092] Furthermore, when randomly selecting basic sentences, the weight of each sentence is adjusted to perform weighted sampling. In this way, the method of automatically generating text review content can be further optimized, and a more intelligent review method can be achieved, which can also effectively help teachers quickly correct online homework and reduce working time.

[0093] For example, based on the initial extraction probability p0, sentence s, and the number of repeated words m, the extraction weight of sentence s is p0 / 10^cnt_m.

[0094] In this way, the automatic generation method can be further optimized, which can effectively help teachers to quickly correct online homework and reduce working time, thereby improving teachers' work efficiency and optimizing the workflow of teachers correcting online homework.

[0095] It should be noted that, in Example 2, descriptions of the same parts as in Example 1 are omitted.

[0096] Those skilled in the art will appreciate that the modules in the above device embodiments may be distributed in the device as described, or may be changed accordingly and distributed in one or more devices different from the above embodiments. The modules in the above embodiments may be combined into one module, or may be further split into multiple submodules.

[0097] Compared with the prior art, the present invention can ensure the diversity, richness and structural integrity of comment content generation through the setting of multi-level labels; by receiving label data input by teachers, it can realize automatic or editable generation of text comment content, and provide teachers with editable functions for real-time modification; it can effectively help teachers quickly correct online homework and reduce working time, thereby improving teachers' work efficiency and optimizing the workflow of teachers correcting online homework.

[0098] Example 2

[0099] The following describes an apparatus embodiment of the present invention, which can be used to execute a method embodiment of the present invention. The details described in the apparatus embodiment of the present invention should be regarded as supplementary to the above method embodiment; details not disclosed in the apparatus embodiment of the present invention can be implemented with reference to the above method embodiment.

[0100] Reference Figures 4 to 6 , a comment text generating device 400 of Example 2 of the present invention will be described, which is used for teacher comments in online education APP, and the comment text generating device 400 includes: an establishment module 401, used to establish a comment corpus, the comment corpus includes multi-level comment tags, each level of comment tags includes multiple comment tags and multiple basic corpora corresponding to each comment tag, each basic corpus includes positive appositive phrases and negative appositive phrases; a receiving module 402, used to receive teacher input, the teacher input includes selecting multiple comment tags; a processing module 403, used to use the comment corpus to perform level label matching based on the teacher input; an automatic generation module 404, according to the matching result, calls the corresponding generation template to automatically generate text evaluation content.

[0101] like Figure 5 As shown, it also includes a setting module 501, and the setting module 501 presets a generation template associated with comment labels at various levels according to the generation rules; the generation rules include setting generation parameters, automatically or editably generating evaluation content, and the generation parameters include text generation scenarios, course business types, text generation length, writing parameters, timeliness parameters, completion structure, whether the submitted homework content is complete, whether there are illustrations, and overall evaluation information.

[0102] Preferably, it also includes: the comment labels at each level include automatically generated labels corresponding to writing parameters, timeliness parameters, whether the submitted homework content is complete, whether there are illustrations, and overall review information; and corresponding evaluation content is automatically generated according to the automatically generated labels in the teacher's input data.

[0103] Preferably, the method further comprises: generating one or more basic sentences for each basic corpus; parsing and expanding the basic corpus containing a plurality of appositive phrases using the Cartesian product method, and determining to generate a corresponding number of basic sentences according to the number of appositive phrases and the number of appositive words.

[0104] like Figure 6 As shown, it also includes a detection module 601, which is used to detect whether there is a first-level review label in the current teacher input data, whether there is a second-level review label belonging to the first-level review label; judge the good or bad tendency of each level of label, and return the predetermined generation result.

[0105] Preferably, it also includes: level tags are used to indicate the hierarchy of the evaluation content structure, wherein the first-level review tag is used to indicate the top level of the evaluation content structure, and the first-level review tag corresponds to the text generated by splicing the review texts generated by at least two second-level review tags; the second-level review tag is used to indicate the stratification of the evaluation content structure, and the second-level review tag corresponds to the generation of a basic sentence.

[0106] Preferably, it also includes: determining multiple corresponding labels according to the level label matching results; when automatically selecting basic sentences in the corresponding labels, extracting each basic sentence with equal probability, splicing and combining the extracted basic sentences, and combining them to generate a template to generate text evaluation content corresponding to the teacher's input.

[0107] Preferably, it also includes: adjusting the generated text evaluation content, wherein the adjustment process includes adjusting the extraction weight of the words according to the number of occurrences or frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

[0108] Compared with the prior art, the present invention can ensure the diversity, richness and structural integrity of comment content generation through the setting of multi-level labels; by receiving label data input by teachers, it can realize automatic or editable generation of text comment content, and provide teachers with editable functions for real-time modification; it can effectively help teachers quickly correct online homework and reduce working time, thereby improving teachers' work efficiency and optimizing the workflow of teachers correcting online homework.

[0109] Example 3

[0110] The following describes an electronic device embodiment of the present invention, which can be regarded as a specific physical implementation of the method and device embodiments of the present invention. The details described in the electronic device embodiment of the present invention should be regarded as a supplement to the above method or device embodiments; details not disclosed in the electronic device embodiment of the present invention can be implemented with reference to the above method or device embodiments.

[0111] Figure 7 is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention, the electronic device includes a processor and a memory, the memory is used to store a computer executable program, when the computer program is executed by the processor, the processor executes Figure 1 method.

[0112] like Figure 7 As shown, the electronic device is presented in the form of a general computing device. The processor may be one or more and work in coordination. The present invention does not exclude distributed processing, that is, the processor may be dispersed in different physical devices. The electronic device of the present invention is not limited to a single entity, but may also be the sum of multiple physical devices.

[0113] The memory stores a computer executable program, which is usually a machine-readable code. The computer-readable program can be executed by the processor to enable the electronic device to perform the method of the present invention, or at least part of the steps in the method.

[0114] The memory includes a volatile memory, such as a random access memory unit (RAM) and / or a cache memory unit, and may also be a non-volatile memory, such as a read-only memory unit (ROM).

[0115] Optionally, in this embodiment, the electronic device further includes an I / O interface, which is used for the electronic device to exchange data with an external device. The I / O interface can represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of a variety of bus structures.

[0116] It should be understood that Figure 7 The electronic device shown is only an example of the present invention, and the electronic device of the present invention may also include elements or components not shown in the above examples. For example, some electronic devices also include display units such as display screens, and some electronic devices also include human-computer interaction elements such as buttons, keyboards, etc. As long as the electronic device can execute the computer-readable program in the memory to implement the method of the present invention or at least part of the steps of the method, it can be considered as an electronic device covered by the present invention.

[0117] Figure 8 Schematic diagram of a computer readable recording medium according to an embodiment of the present invention. Figure 8As shown, a computer executable program is stored in a computer-readable recording medium, and when the computer executable program is executed, the above method of the present invention is implemented. The computer-readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, device, or device. The program code contained on the readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0118] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0119] Through the above description of the implementation mode, it is easy for those skilled in the art to understand that the present invention can be implemented by hardware capable of executing a specific computer program, such as the system of the present invention, and the electronic processing unit, server, client, mobile phone, control unit, processor, etc. contained in the system, but it should be noted that the computer software for executing the method of the present invention is not limited to being executed by one or a specific hardware entity, and it can also be implemented in a distributed manner by unspecified specific hardware. For computer software, the software product can be stored in a computer-readable storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), or it can be distributed and stored on the network, as long as it can enable the electronic device to execute the method according to the present invention.

[0120] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the present invention is not inherently related to any specific computer, virtual device or electronic device, and various general devices can also implement the present invention. The above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating a review text, which is used for teacher reviews of online education APPs, characterized in that: The method comprises the following steps: Establishing a review corpus, the review corpus includes multi-level review tags, each level of review tags includes multiple review tags and multiple basic corpora corresponding to each review tag, each level of review tags is configured with multiple basic corpora, and each basic corpus includes positive appositive phrases and negative appositive phrases; Among them, the multi-level review tags represent the levels of the review content structure; Among them, each basic corpus generates one or more basic sentences, and the basic corpus containing multiple appositive phrases is parsed and expanded using Cartesian product, and the corresponding number of basic sentences generated is determined according to the number of appositive phrases and the number of appositive words; Receiving teacher input, the teacher input comprising selecting a plurality of review tags; Based on the teacher input, the review corpus is used to perform level label matching and identify automatically generated labels, including: Monitor the data input by the teacher in real time, detect whether there is a first-level review tag among the multiple review tags currently input by the teacher, and whether there is a second-level review tag belonging to the first-level review tag, and if so, further identify the number of first-level review tags and their corresponding generated text content, and identify the number of second-level review tags and their corresponding generated text content; Among them, the first-level review tag represents the top layer of the evaluation content structure, and the first-level review tag corresponds to the text generated by the review text generated by at least two second-level review tags in sequence or randomly spliced; the second-level review tag represents the layering of the evaluation content structure, and the second-level review tag corresponds to the generation of a basic sentence; Judge the good or bad tendencies of labels at all levels and return the predetermined generated results; According to the generation rules, generation templates associated with review tags at all levels are preset; Based on the identified automatically generated tags, the corresponding generation template is called to automatically generate the corresponding text evaluation content.

2. The method for generating a review text according to claim 1, characterized in that: Also includes: The generation rules include setting generation parameters, automatically or editably generating evaluation content, and the generation parameters include text generation scenario, course business type, text generation length, writing parameters, timeliness parameters, completion structure, whether the submitted homework content is complete, whether there are pictures, and overall evaluation information; and / or, The review labels at each level include automatically generated labels corresponding to writing parameters, timeliness parameters, whether the submitted homework content is complete, whether there are accompanying pictures, and overall review information.

3. The review text generation method according to claim 1 or 2, characterized in that: Also includes: Determine multiple corresponding labels according to the level label matching result; When automatically selecting basic sentences in the corresponding tags, each basic sentence is extracted with equal probability, the extracted basic sentences are spliced ​​and combined, and combined with the generated template to generate text evaluation content corresponding to the teacher's input.

4. The review text generation method according to claim 1 or 2, characterized in that: Also includes: The generated text evaluation content is adjusted, and the adjustment process includes adjusting the extraction weight of each word according to the number of occurrences or the frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

5. The review text generation method according to claim 1 or 2, characterized in that: Also includes: Determine multiple corresponding labels according to the level label matching result; When automatically selecting the basic sentences in the corresponding tags, extracting each basic sentence with equal probability, splicing and combining the extracted basic sentences, and combining them with the generated template to generate text evaluation content corresponding to the teacher's input; as well as, The generated text evaluation content is adjusted, and the adjustment process includes adjusting the extraction weight of each word according to the number of occurrences or the frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

6. A review text generation device, which is used for teacher reviews of online education APP, characterized in that: The device comprises: Establishing a module, for establishing a review corpus, the review corpus includes multi-level review tags, each level of review tags includes multiple review tags and multiple basic corpora corresponding to each review tag, each level of review tags is configured with multiple basic corpora, each basic corpus includes positive appositive phrases and negative appositive phrases; wherein the multi-level review tags represent the hierarchy of the evaluation content structure; wherein each basic corpus generates one or more basic sentences, uses Cartesian product to parse and expand the basic corpus containing multiple appositive phrases, and determines the generation of a corresponding number of basic sentences according to the number of appositive phrases and the number of appositive words; A receiving module, used to receive teacher input, wherein the teacher input includes selecting multiple review tags; A setting module, used to preset generation templates associated with review tags at various levels according to generation rules; The processing module, based on the teacher input, uses the review corpus to perform level label matching and identify automatically generated labels, including: The detection module is used to monitor the data input by the teacher in real time, detect whether there is a first-level comment tag among the multiple comment tags input by the current teacher, and whether there is a second-level comment tag belonging to the first-level comment tag. If there is, further identify the number of first-level comment tags and their corresponding generated text content, and identify the number of second-level comment tags and their corresponding generated text content; wherein the first-level comment tag represents the top layer of the evaluation content structure, and the first-level comment tag corresponds to the text generated by the comment text generated by at least two second-level comment tags in sequence or randomly spliced; the second-level comment tag represents the layering of the evaluation content structure, and the second-level comment tag corresponds to the generation of a basic sentence; judge the good or bad tendency of the tags at each level, and return the predetermined generation result; The automatic generation module calls the corresponding generation template based on the matching results and automatically generates text evaluation content.

7. The review text generating device according to claim 6, characterized in that: Also includes The generation rules include setting generation parameters, automatically or editably generating evaluation content, and the generation parameters include text generation scenario, course business type, text generation length, writing parameters, timeliness parameters, completion structure, whether the submitted homework content is complete, whether there are pictures, and overall evaluation information; and / or, The review labels at each level include automatically generated labels corresponding to writing parameters, timeliness parameters, whether the submitted homework content is complete, whether there are accompanying pictures, and overall review information.

8. The review text generating device according to claim 6 or 7, characterized in that: Also includes: According to the level label matching result, multiple corresponding labels are determined; When automatically selecting basic sentences in the corresponding tags, each basic sentence is extracted with equal probability, the extracted basic sentences are spliced ​​and combined, and combined with the generated template to generate text evaluation content corresponding to the teacher's input.

9. The review text generating device according to claim 6 or 7, characterized in that: Also includes: The generated text evaluation content is adjusted, and the adjustment process includes adjusting the extraction weight of the words according to the number of occurrences or the frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

10. The review text generating device according to claim 6 or 7, characterized in that: Also includes: According to the level label matching result, multiple corresponding labels are determined; When automatically selecting the basic sentences in the corresponding tags, extracting each basic sentence with equal probability, splicing and combining the extracted basic sentences, and combining them with the generated template to generate text evaluation content corresponding to the teacher's input; as well as, The generated text evaluation content is adjusted, and the adjustment process includes adjusting the extraction weight of the words according to the number of occurrences or the frequency of occurrence of each word in the historical text evaluation content to update the extraction probability of each basic sentence.

11. An electronic device, comprising a processor and a memory, wherein the memory is used to store a computer executable program, wherein: When the computer executable program is executed by the processor, the processor executes the review text generating method according to any one of claims 1 to 5.

12. A computer-readable medium storing a computer-executable program, characterized in that: When the computer executable program is executed, the review text generation method according to any one of claims 1 to 5 is implemented.

Images