Conversational analysis system and method

Abstract

This specification provides a dialogue analysis system and method. The system includes: an acquisition module for acquiring a text sequence, wherein the text sequence is obtained by real-time text conversion of classroom dialogues; a matching module for searching, in a pre-established rule base, using a large language model, whether a target sequence matching the text sequence exists; wherein the rule base includes classification rule sequences for various dialogue types and / or reference text sequences corresponding to each dialogue type; and an analysis module for analyzing, based on the target sequence matched by the matching module, to obtain the target dialogue type corresponding to the classroom dialogue. This disclosure combines a rule base with a large language model to achieve usability and flexibility in classroom dialogue analysis, ensuring both deep integration of the dialogue analysis system with educational theory and the ability to process complex dialogue sequences through a large language model, achieving accurate dialogue classification and analysis.

Description

Technical Field

[0001] This specification relates to the field of natural language processing, and in particular to a dialogue analysis system and method. Background Technology

[0002] In the field of educational technology, the quality of classroom dialogue is a crucial factor influencing student learning outcomes. High-quality classroom dialogue can foster students' critical thinking, deep learning, and collaborative abilities.

[0003] However, the process of systematically analyzing these dialogues is extremely complex and time-consuming, especially the analysis of classroom dialogue sequences. Traditional analysis methods mainly rely on manual coding, which requires a large amount of human involvement, consumes a lot of time, and is easily affected by subjective factors.

[0004] Furthermore, existing automated analysis tools are typically based on machine learning and natural language processing techniques (such as convolutional neural networks and long short-term memory networks). While these techniques have made progress in dialogue pattern recognition, they often require large amounts of data for training, are difficult to adapt to changes in different educational scenarios, and lack deep integration with teaching theories. Summary of the Invention

[0005] In view of this, this specification provides a dialogue analysis system and method.

[0006] According to a first aspect of the embodiments of this specification, a dialogue analysis system is provided, the system comprising:

[0007] The acquisition module is used to acquire a text sequence; wherein the text sequence is obtained by real-time text conversion of classroom dialogues;

[0008] The matching module is used to search in a pre-established rule base, using a large language model, for whether there is a target sequence that matches the text sequence; wherein, the rule base includes classification rule sequences for multiple dialogue types and / or reference text sequences corresponding to each dialogue type;

[0009] The analysis module is used to analyze the target dialogue type corresponding to the classroom dialogue based on the target sequence matched by the matching module.

[0010] According to a second aspect of the embodiments of this specification, a dialogue analysis method is provided, the method comprising:

[0011] Obtain a text sequence; wherein the text sequence is obtained by real-time text conversion of classroom dialogues;

[0012] In a pre-established rule base, a large language model is used to search for whether a target sequence matching the text sequence exists; wherein, the rule base includes classification rule sequences for multiple dialogue types and / or reference text sequences corresponding to each dialogue type;

[0013] Based on the matched target sequence, the target dialogue type corresponding to the classroom dialogue is analyzed and obtained.

[0014] According to a third aspect of the embodiments of this specification, a dialogue analysis apparatus is provided, the apparatus comprising: one or more processors; wherein the processors are configured to perform steps of the method as described in any of the second aspects.

[0015] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores computer instructions thereon, which, when executed by a processor, implement the steps of the method as described in any of the second aspects.

[0016] According to a fifth aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the steps of the method as described in any of the second aspects.

[0017] The technical solutions provided in the embodiments of this specification may include the following beneficial effects:

[0018] In the embodiments described in this specification, a rule base can be combined with a large language model to achieve the usability and flexibility of classroom dialogue analysis. This ensures both the deep integration of the dialogue analysis system with educational theory and the ability to process complex dialogue sequences through the large language model, thereby achieving accurate dialogue classification and analysis.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this specification. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this specification and, together with the description, serve to explain the principles of this specification.

[0021] Figure 1 This is one of the structural diagrams of a dialogue analysis system provided in an exemplary embodiment.

[0022] Figure 2 This is a second schematic diagram of the structure of a dialogue analysis system provided in an exemplary embodiment.

[0023] Figure 3 This is the third schematic diagram of the structure of a dialogue analysis system provided in an exemplary embodiment.

[0024] Figure 4 This is the fourth schematic diagram of the structure of a dialogue analysis system provided in an exemplary embodiment.

[0025] Figure 5 This is one of the flowcharts illustrating a dialogue analysis method provided in an exemplary embodiment.

[0026] Figure 6 This is a second flowchart illustrating a dialogue analysis method provided in an exemplary embodiment.

[0027] Figure 7 This is a flowchart of a dialogue analysis method provided in an exemplary embodiment.

[0028] Figure 8 This is the fifth schematic diagram of the structure of a dialogue analysis system provided in an exemplary embodiment.

[0029] Figure 9 This is a schematic diagram of the structure of a server provided in an exemplary embodiment. Detailed Implementation

[0030] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this specification as detailed in the appended claims.

[0031] The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “the,” and “the” as used in this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0032] It should be understood that although the terms first, second, third, etc., may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this specification, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0033] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0034] Currently, the following methods can be used for language comprehension and dialogue analysis:

[0035] Method 1: Manual coding.

[0036] Early dialogue analysis relied on manual coding analysis by human experts. While this method offered some flexibility and accuracy, it became increasingly impractical as classroom size and dialogue complexity increased.

[0037] Method 2: Rule-based Artificial Intelligence (AI).

[0038] Early AI systems relied on predefined rules to operate. These systems could work effectively in relatively closed environments, but their performance was limited in dynamic and complex classroom environments due to their lack of adaptability to new situations.

[0039] Method 3: The introduction of machine learning and large language models.

[0040] With the rise of machine learning and deep learning, the education field has also begun to introduce data-driven analysis models, such as Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM).

[0041] The specific implementation method may include the following steps:

[0042] Dataset creation: First, classroom dialogues are recorded and transcribed to generate a dialogue dataset for training the model.

[0043] Model training: These datasets are used to train CNNs or LSTMs, which then learn patterns from the dialogue to perform classification.

[0044] Automatic encoding: After training, the model can automatically encode and classify new input classroom dialogues, thereby generating an analysis report on the structure of the classroom dialogues.

[0045] While these models can handle large amounts of data, they have the following main drawbacks:

[0046] Insufficient integration with educational theory: Most automated dialogue analysis technologies rely on data-driven learning and lack deep integration with educational theory. Therefore, while these systems can provide dialogue pattern analysis under normal circumstances, they cannot conduct in-depth analysis based on specific teaching objectives and classroom needs, thus limiting their application in educational scenarios.

[0047] Insufficient dialogue sequence recognition capability: There is a greater focus on classifying single dialogue fragments, making it difficult to effectively analyze complex dialogue sequences. Dialogue sequences are crucial in education because students' cognitive development often manifests in continuous dialogue processes, and the inability to identify and classify these continuous dialogue patterns is a significant weakness.

[0048] The issue of model dependency and flexibility: Because automated analysis techniques heavily rely on pre-trained models, these models typically require large amounts of labeled data for effective training. However, in the education field, acquiring labeled data is costly, and this data is often context-dependent; a model may perform well in one scenario but adapt poorly to different classroom settings.

[0049] Lack of real-time feedback: Machine learning models typically run in offline environments, and the analysis results are often generated after class, which cannot provide teachers with real-time classroom feedback and limits their application value in actual teaching.

[0050] To automatically analyze classroom dialogue sequences and provide real-time theoretical support feedback, thereby effectively supporting the improvement of classroom teaching quality, this specification proposes a dialogue analysis system and method. This system, based on a rule base and a large language model, achieves usability and flexibility in classroom dialogue analysis. It ensures both deep integration of the dialogue analysis system with educational theory and the ability to handle complex dialogue sequences through the large language model, enabling accurate dialogue classification and analysis.

[0051] The system and methods provided in this manual are applicable to various educational scenarios, especially in primary and secondary school classrooms, for analyzing dialogues between students and teachers, as well as discussions among students. Furthermore, they can be widely used in teacher professional development projects to help teachers obtain real-time feedback on their classroom dialogues, thereby improving teaching effectiveness.

[0052] It is understood that the systems and methods provided in this specification can also be applied to other scenarios, including but not limited to corporate training, psychological counseling, medical consultation, social media analysis, and customer service. Any solution that combines rule bases with large language models for dialogue analysis should fall within the scope of this disclosure.

[0053] The following is a brief introduction to the dialogue analysis system provided in this publication.

[0054] Figure 1 This is a schematic diagram of the structure of a dialogue analysis system provided in an exemplary embodiment. For example... Figure 1 As shown, the system may include:

[0055] Get module 101;

[0056] Matching module 102;

[0057] Analysis module 103.

[0058] In some embodiments, the acquisition module 101 can be used to acquire a text sequence. This text sequence is obtained by real-time text conversion of classroom dialogues.

[0059] In one example, the acquisition module 101 can be connected to an audio acquisition device, such as a microphone, to acquire audio signals of classroom conversations and convert the acquired audio signals into text sequences in real time.

[0060] In some embodiments, the matching module 102 can be used to search in a pre-established rule base for a target sequence that matches the text sequence using a large language model.

[0061] In one example, a Large Language Model (LLM) refers to a deep learning model trained on a large amount of text data, capable of generating natural language text or understanding the meaning of language text. LLMs can handle various natural language tasks, such as text classification, question answering, and dialogue. In this specification, the LLM can be used to perform text analysis on a text sequence and search for matching target sequences in a rule base. Subsequently, the LLM can also be used to generate and output a dialogue analysis report.

[0062] In one example, the dialogue analysis system can collect a large amount of relevant literature on educational theories, analyze and summarize this literature, and thus create the rule base.

[0063] For example, the rule base may include a sequence of classification rules for various dialogue types and / or a sequence of reference texts corresponding to each dialogue type.

[0064] For example, the types of dialogues involved in this specification include, but are not limited to, at least one of the following: critical inquiry dialogue; collaborative knowledge construction dialogue; guiding dialogue; supportive dialogue; reflective dialogue; metacognitive dialogue.

[0065] Critical inquiry dialogue refers to a dialogue in which participants constructively challenge and counter-challenge each other's viewpoints during discussions or learning, thereby promoting a deeper understanding and resolution of the problem. It emphasizes that participants question existing viewpoints or assumptions and, through rational dialogue and reflection, critically examine and evaluate various arguments, thus promoting the collaborative construction and innovation of knowledge.

[0066] Among them, knowledge collaborative dialogue can refer to a dialogue in which participants solve the same problem through the collaborative use of knowledge from multiple disciplines during a discussion or learning process.

[0067] Among them, guided dialogue can refer to guided dialogue that can help participants think of ways to solve problems and even reflect on them.

[0068] Supportive dialogue can refer to a dialogue in which, when at least one participant proposes a theory, other participants support that theory by summarizing conclusions, providing arguments, and so on.

[0069] Reflective dialogue can refer to a dialogue in which participants, while solving a problem, reflect on the impact of that problem on others or other areas.

[0070] Metacognitive dialogue can refer to dialogue about "metacognition," where metacognition is cognition of cognition. Specifically, in the context of learning, metacognition refers to the understanding and control of thinking and learning activities, including metacognitive knowledge and metacognitive control.

[0071] The above is merely an illustrative example; other dialogue types may also be included, and this specification does not limit this.

[0072] For example, the classification rule sequence in this specification may refer to the rules that a dialogue conforms to when classifying a dialogue. For instance, when a keyword matching a certain dialogue type A appears in the dialogue, it can be determined that the dialogue belongs to dialogue type A.

[0073] For example, when guiding keywords appear in a conversation, the conversation can be classified as a guiding conversation. Guiding keywords include phrases like "Could you give an example?" or "Is there anything else you'd like to add?"

[0074] For example, when supportive keywords appear in a conversation, the conversation can be classified as a supportive conversation. Supportive keywords include words such as "agree," "promise," and "that's absolutely right."

[0075] For example, when critical keywords appear in a dialogue, the dialogue can be classified as a critical inquiry dialogue. Critical keywords include phrases such as "this approach is wrong," "this line of thinking is flawed," and "this has negative consequences."

[0076] For example, the classification rule sequence can be used to indicate at least one of the following: the minimum number of dialogue rounds corresponding to each dialogue type; and the dialogue structure corresponding to each dialogue type.

[0077] For example, the minimum number of dialogue rounds corresponding to a critical inquiry dialogue is 3, and the corresponding dialogue structure includes: (Reasoning Invitation (REI), Elaboration Invitation (ELI)) and argumentation (Reasoning (RE), Elaboration (EL)), and also includes the querying (Q) behavior.

[0078] For example, the minimum number of dialogue rounds for reflective dialogue is 2, and the corresponding dialogue structure includes: (Reasoning (RE), Agreement (A)), and may include querying (Q) behavior.

[0079] The above is merely an illustrative example, and this disclosure does not limit the classification rules for dialogue types.

[0080] For example, each dialogue type can correspond to at least one reference text sequence, which can also be called the "ideal text sequence". It is the most appropriate and reasonable text sequence obtained by sequence prediction based on the connotation of the dialogue type classification and the encoding of the dialogue function.

[0081] For example,

[0082] The reference text sequence corresponding to a certain dialogue type can be at least one of the following:

[0083] REI→RE→Q: Invitation to reason, followed by reasoning, and then questioning the reasoning.

[0084] Q→RE→REI: Question a statement, then reason, and then invite further reasoning.

[0085] Coordination Invitation (CI) → Q → RE: Coordination Invitation, followed by questioning, and then reasoning.

[0086] The above is merely an illustrative example, and this disclosure does not limit the specific content included in the rule base.

[0087] In one example, the matching module 102 can first use a large language model to annotate the text sequence to obtain the annotation result. Further, in the rule base, the large language model can be used to search for whether there is a target sequence that matches the annotation result.

[0088] For example, the text sequence obtained from the classroom dialogue may include at least one text subsequence, each text subsequence corresponding to one round of dialogue in the classroom dialogue.

[0089] For example, the large language model annotates the text content for each text subsequence (i.e., each round of dialogue). Specifically, the large language model can encode the text content of each text subsequence, converting the text subsequence into codewords that the large language model can understand. These codewords constitute the annotation results.

[0090] The large language model can obtain the above annotation results based on the predefined dialogue encoding table (a mapping table between text and codewords).

[0091] For example, if the annotation result matches the minimum number of dialogue rounds corresponding to at least one dialogue type, it can be determined that the target sequence exists.

[0092] For example, if the annotation result matches the dialogue structure corresponding to at least one dialogue type, it can be determined that the target sequence exists.

[0093] For example, if the annotation result matches the minimum number of dialogue rounds and / or dialogue structure corresponding to at least one dialogue type, it can be determined that the target sequence exists.

[0094] For example, if the annotation result matches a reference text sequence corresponding to at least one dialogue type, it can be determined that the target sequence exists.

[0095] For example, if the annotation result matches the minimum number of dialogue rounds corresponding to at least one dialogue type and the reference text sequence, it can be determined that the target sequence exists.

[0096] For example, if the annotation result matches the dialogue structure corresponding to at least one dialogue type and the reference text sequence, it can be determined that the target sequence exists.

[0097] For example, if the annotation result matches the minimum number of dialogue rounds, dialogue structure, and reference text sequence corresponding to at least one dialogue type, it can be determined that the target sequence exists. In some embodiments, if the matching module 102 matches at least one target sequence, the analysis module 103 can analyze and obtain the target dialogue type to which the classroom dialogue belongs based on the target sequence.

[0098] For example, if a classroom dialogue consists of three rounds, and the target sequence matched by the matching module 102 is REI→RE→Q, then the analysis module 103 can determine that the classroom dialogue belongs to a critical inquiry dialogue.

[0099] Understandably, in this specification, the large language model no longer needs to be trained on a sample dataset to classify classroom dialogues; that is, manual annotation of the sample dialogue text set is unnecessary, saving labor costs. In this specification, the large language model can directly match the standard results with sequences in the rule base to determine the target dialogue type corresponding to the classroom dialogue, achieving accurate dialogue classification and analysis.

[0100] In the above embodiments, the rule base can be combined with the large language model to achieve the usability and flexibility of classroom dialogue analysis. This ensures the deep integration of the dialogue analysis system with educational theory and enables the processing of complex dialogue sequences through the large language model, achieving accurate dialogue classification and analysis.

[0101] In some embodiments, Figure 2 It is based on Figure 1 The illustrated embodiment provides a schematic diagram of the structure of another dialogue analysis system. For example... Figure 2 As shown, the system may also include:

[0102] Processing module 104.

[0103] In one example, the processing module 104 can be used to preprocess the converted text sequence, the preprocessing including at least text proofreading, which includes but is not limited to proofreading the text content, text structure, and text format.

[0104] In the above embodiments, the system may also include a processing module to preprocess the text sequence, thereby improving the reliability of classroom dialogue analysis.

[0105] In some embodiments, Figure 3 It is based on Figure 1 The illustrated embodiment provides a schematic diagram of the structure of another dialogue analysis system. For example... Figure 3 As shown, the system may also include:

[0106] Output module 105.

[0107] In one example, output module 105 can be used to generate and output a dialogue analysis report using a large language model based on the target dialogue type.

[0108] For example, a dialogue analysis report includes, but is not limited to, at least one of the following:

[0109] The target dialogue type to which each round of dialogue in the classroom dialogue belongs;

[0110] The proportion of each of the aforementioned target dialogue types;

[0111] The content assessment results of the classroom dialogue.

[0112] The evaluation results of classroom dialogues may include, but are not limited to, the strengths of the classroom dialogues and / or areas that need improvement.

[0113] The above embodiments can provide real-time analysis reports of classroom dialogues, which helps to understand classroom content, promote interaction between teachers and students, and have high usability.

[0114] In some embodiments, Figure 4 It is based on Figure 1 The illustrated embodiment provides a schematic diagram of the structure of another dialogue analysis system. For example... Figure 4 As shown, the system may also include:

[0115] Processing module 104;

[0116] Output module 105.

[0117] In the above embodiments, the dialogue analysis system can process and analyze dialogue data in real time during class and generate dialogue analysis reports to help teachers adjust their teaching strategies in a timely manner. This real-time feedback function greatly improves the system's practicality and enhances the interactivity and flexibility of classroom teaching.

[0118] Figure 5 This is a flowchart illustrating a dialogue analysis method provided in an exemplary embodiment. Figure 5 As shown, the method may include the following steps:

[0119] Step S501: Obtain the text sequence.

[0120] In some embodiments, the text sequence is obtained by real-time text conversion of classroom dialogue. The process of acquiring the text sequence is similar to the action performed by the acquisition module 101, and will not be described in detail here.

[0121] Step S503: In the pre-established rule base, search for whether there is a target sequence that matches the text sequence using a large language model.

[0122] In some embodiments, the rule base includes a sequence of classification rules for various dialogue types and / or a sequence of reference text corresponding to each dialogue type. The process of matching the target sequence is similar to the action performed by the matching module 102, and will not be described in detail here.

[0123] Step S504: Based on the matched target sequence, analyze and obtain the target dialogue type corresponding to the classroom dialogue.

[0124] In some embodiments, the process of analyzing the target dialogue type is similar to the action performed by the analysis module 103, and will not be described again here.

[0125] In the above embodiments, the rule base can be combined with the large language model to achieve the usability and flexibility of classroom dialogue analysis. This ensures the deep integration of the dialogue analysis system with educational theory and enables the processing of complex dialogue sequences through the large language model, achieving accurate dialogue classification and analysis.

[0126] Figure 6 It is based on Figure 5 The illustrated embodiment provides a flowchart of another dialogue analysis method. For example... Figure 6 As shown, the method may include the following steps:

[0127] Step S502: Preprocess the text sequence.

[0128] In some embodiments, preprocessing includes at least text proofreading. The preprocessing process is similar to the actions performed by processing module 104, and will not be described in detail here.

[0129] In the above embodiments, the text sequence can be preprocessed, which improves the reliability of classroom dialogue analysis.

[0130] Figure 7 It is based on Figure 6 The illustrated embodiment provides a flowchart of another dialogue analysis method. For example... Figure 7 As shown, the method may include the following steps:

[0131] Step S505: Based on the target dialogue type, generate and output a dialogue analysis report using a large language model.

[0132] In some embodiments, the process of outputting a dialogue analysis report is similar to the action performed by the output module 105, and will not be described again here.

[0133] In the above embodiments, dialogue data can be processed and analyzed in real time during class, and dialogue analysis reports can be generated to help teachers adjust their teaching strategies in a timely manner. This real-time feedback function greatly improves the system's practicality and enhances the interactivity and flexibility of classroom teaching.

[0134] The above process is further illustrated with examples below.

[0135] This specification provides a classroom dialogue analysis system based on an (expert) rule base and a large language model. By combining the knowledge of educational experts with the natural language processing capabilities of the large language model, this system can automatically identify and analyze dialogue sequences in the classroom and provide real-time feedback. The specific technical implementation scheme is as follows:

[0136] Creation of the expert rule base:

[0137] The system first creates an expert rule base based on a large amount of educational theory and existing literature. This rule base contains classification criteria for different dialogue types and ideal patterns for dialogue sequences. These rules are defined according to educational objectives, such as inquiry-based dialogue, collaborative knowledge construction, and reflective dialogue, ensuring that the system's analysis conforms to educational theory.

[0138] Introduction of large language models:

[0139] Large language models (such as GPT) are used to understand the natural language in classroom dialogues and are combined with expert knowledge from a rule base. This combination enables the system to automatically classify classroom dialogues by simultaneously utilizing symbolic reasoning and flexible language understanding.

[0140] The dialogue sequence analysis process includes:

[0141] Step 1, Data Collection and Preprocessing: Obtain classroom dialogue data through classroom recordings or classroom dialogue transcription, and perform text preprocessing to ensure the data is structured.

[0142] Step 2, Dialogue Type Identification: Use a large language model to process dialogue data, identify the type of dialogue (such as exploratory dialogue, collaborative dialogue, etc.), and match it with rules in the expert rule base.

[0143] Step 3, Sequence Pattern Analysis: By defining dialogue sequences in the rule base, the system classifies continuous dialogue segments according to sequence and identifies complex dialogue patterns, such as teacher-student Q&A, student discussions, and deep reasoning.

[0144] Step 4, Real-time Feedback: Based on the analysis results of the dialogue, the system provides teachers with a real-time dialogue analysis report, pointing out the strengths and areas for improvement in the classroom dialogue.

[0145] System architecture and implementation:

[0146] This system, for example Figure 8 As shown, it mainly includes a data processing module 801, an expert rule base module 802, a natural language processing module 803, and a feedback generation module 804. Each module works together to realize the entire process of data processing from collection and analysis to feedback.

[0147] The data processing module 801 can convert the audio signal of the classroom dialogue into a text sequence and preprocess the text sequence.

[0148] Among them, the expert rule base module 802 can store classification rule sequences for multiple dialogue types and / or the ideal dialogue sequence corresponding to each dialogue type.

[0149] Among them, the natural language processing module 803 can be a large language model, which can determine the target sequence that matches the text sequence from the sequence provided by the expert rule base module 802, and determine the target dialogue type corresponding to the classroom dialogue based on the target sequence, thereby realizing the classification of complex dialogues.

[0150] The feedback generation module 804 can generate and output a dialogue analysis report.

[0151] Algorithm implementation:

[0152] The rule base module utilizes a symbolic reasoning mechanism based on expert knowledge, combined with a large language model, to perform semantic analysis on the dialogue content. The system employs a hierarchical rule matching algorithm to progressively filter and classify dialogue sequences, ensuring the accuracy of classification and the rationality of sequence recognition.

[0153] In the above embodiments, the accuracy and flexibility of dialogue analysis are improved by combining a symbolic expert rule base with a large language model. This allows the system to accurately identify complex sequences in classroom dialogues and classify and analyze them according to specific educational objectives. Compared to existing purely data-driven models, this system has better adaptability and generalizability in different classroom contexts. Furthermore, it reduces reliance on labeled data: existing machine learning models typically rely on large amounts of labeled data, while this system, by introducing an expert rule base, reduces its dependence on large-scale training data, enabling effective dialogue analysis even with limited data. Real-time feedback is also provided; the system can process classroom dialogue data in real time and generate instant dialogue analysis reports, helping teachers adjust their teaching strategies based on students' dialogue performance and improving the quality of classroom teaching. In addition, it supports deep integration with educational theory. By combining an expert rule base based on educational theory, this invention not only performs dialogue analysis but also provides theoretical support based on educational objectives, providing a basis for teachers' teaching decisions.

[0154] In some embodiments, this specification also provides a dialogue analysis system based on full machine learning.

[0155] In this approach, the system relies entirely on machine learning models, rather than incorporating expert rule bases. While this reduces human intervention, its lack of reliance on educational experts results in weaker accuracy and theoretical support for dialogue analysis. Furthermore, the model requires significant training time and a large amount of labeled data, making it difficult to adapt to diverse educational contexts.

[0156] In some embodiments, this specification also provides a dialogue analysis system based on a predefined rule set. In this approach, the system relies entirely on the predefined rule set, does not use a large language model, and the dialogue analysis report can be generated manually. While this approach can provide relatively accurate analysis in specific scenarios, it lacks flexibility, struggles to handle complex and varied classroom dialogues, and is insufficiently adaptable to new educational contexts.

[0157] In this instruction manual, it should be noted that:

[0158] (1) The combination of expert rule base and large language model.

[0159] The core innovation of this invention lies in combining a symbolic expert rule base with a large language model, enabling flexible and theoretically supported classroom dialogue analysis. This innovation ensures both the system's deep integration with educational theory and the ability to handle complex dialogue sequences through a large language model, achieving accurate classification and analysis.

[0160] (2) Real-time feedback mechanism.

[0161] The system can process and analyze dialogue data in real time during class and generate dialogue analysis reports to help teachers adjust their teaching strategies promptly. This real-time feedback function greatly improves the system's usability and enhances the interactivity and flexibility of classroom teaching.

[0162] (3) Automated classification and recognition of dialogue sequences.

[0163] Through an expert rule base, the system can effectively identify and classify continuous dialogue sequences, capturing subtle changes in students' thinking development and classroom interactions. This function overcomes the shortcomings of insufficient dialogue sequence recognition capabilities.

[0164] Example 1, Scenario: Elementary school math class, topic: "Introduction to fractions".

[0165] Teacher: Today we're going to learn a new mathematical concept—fractions. Do you all know what fractions are?

[0166] Student A: A fraction is a number used to divide things. For example, if you divide an apple in half, it's one-half.

[0167] Student B: I know that fractions can also represent how many pieces a pie is cut into, and what fraction each piece represents of the whole pie.

[0168] Teacher: Very good, students have a good understanding of fractions. So, what are the characteristics of fractions?

[0169] Student C: A fraction consists of a numerator and a denominator. The numerator represents the part to be divided, and the denominator represents the number of parts into which the whole is divided.

[0170] Student D: I think fractions can be compared in size, for example, one-half is larger than one-third.

[0171] Teacher: You all did a great job. Next, let's practice how to write fractions and how to compare their sizes.

[0172] System analysis results:

[0173] Dialogue type: Instructional dialogue

[0174] Discourse roles: Teacher (facilitator), Student A (narrator), Student B (narrator), Student C (narrator), Student D (narrator).

[0175] Discourse functions: Teacher: Asking questions, guiding students to think, and organizing exercises; Students A, B, C, and D: stating opinions, expressing ideas, and participating in exercises.

[0176] Sequence mode: teacher-student Q&A, student discussion, teacher-organized practice

[0177] The dialogue analysis report includes the following:

[0178] 1. Classroom Dialogue Analysis

[0179] The classroom dialogues are mainly inquiry-based, with teachers guiding students to think and express themselves through questioning. Students actively participate in the discussion, creating a lively classroom atmosphere.

[0180] Students' ability to understand the concept of fractions and to provide examples of their applications demonstrates their mastery of the new knowledge.

[0181] Students were able to describe the characteristics of fractions, such as that fractions consist of a numerator and a denominator, and that fractions can be compared in size, which demonstrated their logical thinking ability.

[0182] 2. Student Participation Analysis

[0183] Students showed high participation, actively participated in discussions, and expressed themselves fluently, using a rich vocabulary and a wide range of sentences.

[0184] Students are able to actively participate in exercises and are able to correctly write fractions and compare their sizes.

[0185] 3. Analysis of Students' Thinking Development

[0186] Students are able to think about problems from different perspectives, such as understanding the concept of fractions from the perspective of dividing objects and dividing shapes.

[0187] Students are able to perform simple reasoning and judgment, such as comparing the size of fractions.

[0188] 4. Analysis of Teachers' Teaching Behaviors

[0189] Teachers can effectively guide students to think and express themselves, provide appropriate feedback, and encourage students to actively participate in discussions.

[0190] Teachers can use a variety of teaching methods, such as questioning, discussion, inspiration, and practice, to stimulate students' interest in learning.

[0191] 5. Suggestions for classroom improvement

[0192] This can further guide students to engage in in-depth thinking and discussion, such as considering the application of fractions in real life and how to solve practical problems related to fractions.

[0193] Students can be encouraged to practice writing, such as writing a math journal entry about fractions.

[0194] Multimedia technologies, such as animation and games, can be used to enrich classroom teaching content and enhance students' learning interest.

[0195] Example 2, Scenario: Junior high school English class, topic: "Environmental Protection"

[0196] Teacher: Today we learned about environmental protection. What are your thoughts on environmental protection?

[0197] Student A: I think environmental protection is very important because environmental problems affect our lives.

[0198] Student B: I agree. We should start with ourselves, reduce waste, and conserve resources.

[0199] Student C: I think environmental protection is not just an individual act, but also requires the joint efforts of the government and all sectors of society.

[0200] Teacher: So, what do you all think we should do to better protect the environment?

[0201] Student D: We can participate in environmental volunteer activities, such as planting trees and cleaning up garbage.

[0202] Student E: We can also advocate for green travel, such as cycling and taking public transportation.

[0203] Teacher: Very good, everyone has offered many excellent suggestions. Now, have you encountered any environmental issues in your daily lives?

[0204] Student F: I noticed that some people litter in public places, which is very uncivilized.

[0205] Student G: I think some people overspend and waste a lot of resources.

[0206] Teacher: So, what do you think we should do to solve these problems?

[0207] Student H: We can raise environmental awareness and let more people understand the importance of environmental protection.

[0208] Student I: We can formulate relevant laws and regulations to punish acts that damage the environment.

[0209] Teacher: Very good, everyone has offered many excellent suggestions. We hope that through our joint efforts, we can protect our environment.

[0210] System analysis results:

[0211] Dialogue type: Reflective dialogue

[0212] Discourse roles: Teacher (facilitator), Student A (narrator), Student B (narrator), Student C (narrator), Student D (narrator), Student E (narrator), Student F (narrator), Student G (narrator), Student H (narrator), Student I (narrator)

[0213] Discourse functions: Teachers: ask questions and guide students to think; Student AI: state opinions, express ideas, and reflect on their own behavior.

[0214] Sequence Patterns: Teacher-Student Q&A, Student Discussions, Reflective Dialogue

[0215] The dialogue analysis report includes the following:

[0216] 1. Classroom Dialogue Analysis

[0217] The classroom dialogues are mainly reflective, with teachers guiding students to reflect on their own behavior and think about how to better protect the environment.

[0218] Students were able to actively participate in the discussion and put forward some practical environmental protection suggestions, which reflects their concern and thinking about environmental issues.

[0219] Students are able to reflect on their own behavior, such as littering and overconsumption, and propose ways to improve it.

[0220] 2. Student Participation Analysis

[0221] Students showed high participation, actively participated in discussions, and expressed themselves fluently, using a rich vocabulary and a wide range of sentences.

[0222] Students are able to actively participate in reflective dialogues and offer valuable insights.

[0223] 3. Analysis of Students' Thinking Development

[0224] Students are able to think about problems from different perspectives, such as from the perspectives of personal behavior, social phenomena, laws and regulations, etc.

[0225] Students are able to engage in in-depth reflection, such as reflecting on the impact of their own behavior on the environment, and are able to propose improvement measures.

[0226] 4. Analysis of Teachers' Teaching Behaviors

[0227] Teachers can effectively guide students to think and express themselves, provide appropriate feedback, and encourage students to actively participate in discussions.

[0228] Teachers can use a variety of teaching methods, such as questioning, discussion, and inspiration, to stimulate students' interest in learning and guide them to reflect.

[0229] 5. Suggestions for classroom improvement

[0230] This can further guide students to engage in in-depth reflection, such as considering the relationship between environmental protection and economic development, and how to balance environmental protection with human needs.

[0231] Students can be encouraged to practice writing, such as writing a reflective essay about environmental protection.

[0232] Students can be organized to participate in environmental protection practice activities, such as visiting environmental protection bases and participating in environmental volunteer activities.

[0233] In some embodiments, this specification also provides a dialogue analysis apparatus that can be deployed on a server of a dialogue analysis system. The apparatus may include one or more processors, wherein the processors are configured to perform the steps of the method as described in any of the preceding embodiments.

[0234] Figure 9 This is a schematic structural diagram of a server for a dialogue analysis system provided in an exemplary embodiment. Please refer to... Figure 9 At the hardware level, the device includes a processor 902, an internal bus 904, a network interface 906, memory 808, and non-volatile memory 910, and may also include other hardware required for its functions. One or more embodiments of this specification can be implemented in software, for example, a large model can be deployed on the server, with the processor 902 reading the corresponding computer program from the non-volatile memory 910 into memory 908 and then running it. Of course, besides software implementation, one or more embodiments of this specification do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution entity of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0235] Based on the same concept as the methods described above, this specification also provides a computer-readable storage medium having computer instructions stored thereon that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0236] Based on the same concept as the methods described above, this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0237] In addition, this disclosure also provides a database operation and maintenance system, which may include a database; a large language model; wherein the model can perform the steps of the method described in any of the above embodiments.

[0238] Other embodiments of this specification will readily occur to those skilled in the art upon consideration of the specification and practice of the invention claimed herein. This specification is intended to cover any variations, uses, or adaptations that follow the general principles of this specification and include common knowledge or customary techniques in the art not claimed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this specification are indicated by the following claims.

[0239] It should be understood that this specification is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this specification is limited only by the appended claims.

[0240] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A dialogue analysis system, characterized in that, The system includes: The acquisition module is used to acquire a text sequence; wherein the text sequence is obtained by real-time text conversion of classroom dialogues; A matching module is used to search, in a pre-established rule base, using a large language model, for the existence of a target sequence that matches the text sequence. The rule base includes classification rule sequences for various dialogue types and / or reference text sequences corresponding to each dialogue type. The dialogue types include at least one of the following: critical inquiry dialogue, where participants constructively challenge and counter-challenge each other's viewpoints during discussion or learning to promote a deeper understanding and solution to a problem; and knowledge collaborative construction dialogue, where participants collaboratively solve the same problem using knowledge from multiple disciplines during discussion or learning. The dialogue can be categorized into several types: problem-solving dialogue, guiding dialogue (which helps participants think of solutions and reflect on their approach), supportive dialogue (where other participants support a theory by summarizing key points and providing evidence), reflective dialogue (where participants consider the impact of the problem on others or other areas while solving it), and metacognitive dialogue (where participants discuss metacognition, the awareness and control of thinking and learning activities, including metacognitive knowledge and metacognitive control). The analysis module is used to analyze the target sequence matched by the matching module to obtain the target dialogue type corresponding to the classroom dialogue; The classification rule sequence is used to indicate at least one of the following: The minimum number of dialogue rounds corresponding to each dialogue type; The dialogue structure corresponding to each dialogue type; The matching module is also used for: The text sequence is annotated using a large language model; In the rule base, the large language model is used to search for whether the target sequence matches the annotation result. If the annotation result matches the minimum number of dialogue rounds and / or dialogue structure corresponding to at least one dialogue type, the existence of the target sequence is determined, and / or, if the annotation result matches the reference text sequence corresponding to at least one dialogue type, the existence of the target sequence is determined.

2. The system according to claim 1, characterized in that, The system also includes: A processing module is used to preprocess the text sequence, the preprocessing including at least text proofreading.

3. The system according to claim 1 or 2, characterized in that, The system also includes: The output module is used to generate and output a dialogue analysis report based on the target dialogue type using a large language model.

4. The system according to claim 3, characterized in that, The dialogue analysis report includes at least one of the following: The target dialogue type corresponding to each round of dialogue in the classroom dialogue; The proportion of each of the aforementioned target dialogue types; The content assessment results of the classroom dialogue.

5. A dialogue analysis method, characterized in that, The method includes: Obtain a text sequence; wherein the text sequence is obtained by real-time text conversion of classroom dialogues; In a pre-established rule base, a large language model is used to search for a target sequence that matches the text sequence. The rule base includes classification rule sequences for various dialogue types and / or reference text sequences corresponding to each dialogue type. The dialogue types include at least one of the following: critical inquiry dialogue, where participants constructively challenge and counter-challenge each other's viewpoints during discussion or learning to promote a deeper understanding and solution to the problem; and knowledge collaborative construction dialogue, where participants collaboratively solve the same problem through knowledge from multiple disciplines during discussion or learning. Dialogue; guided dialogue, which refers to a guiding dialogue that helps participants come up with solutions to problems and reflect on them; supportive dialogue, which refers to a dialogue in which, when at least one participant proposes a theory, other participants support that theory by summarizing conclusions and providing arguments; reflective dialogue, which refers to a dialogue in which participants, while solving a problem, reflect on the impact of the problem on others or other fields; metacognitive dialogue, which refers to dialogue on metacognition, which refers to the understanding and control of thinking and learning activities, and includes metacognitive knowledge and metacognitive control. Based on the matched target sequence, the target dialogue type corresponding to the classroom dialogue is analyzed and obtained; The classification rule sequence is used to indicate at least one of the following: The minimum number of dialogue rounds corresponding to each dialogue type; The dialogue structure corresponding to each dialogue type; The step of searching for a target sequence that matches the text sequence in a pre-established rule base using a large language model includes: The text sequence is annotated using a large language model; In the rule base, the large language model is used to search for whether the target sequence matches the annotation result. If the annotation result matches the minimum number of dialogue rounds and / or dialogue structure corresponding to at least one dialogue type, the existence of the target sequence is determined, and / or, if the annotation result matches the reference text sequence corresponding to at least one dialogue type, the existence of the target sequence is determined.

6. The method according to claim 5, characterized in that, The method further includes: Based on the target dialogue type, a dialogue analysis report is generated and output using a large language model.

Images