Information processing method and device, electronic equipment, storage medium and computer program product

Abstract

The invention discloses an information processing method and device, electronic equipment, a storage medium and a computer program product, and the method comprises the steps: extracting course data of each course video in a plurality of course videos; the course data comprises one or more of the following categories: course introduction information, video image frames and video voices; generating a plurality of keywords based on all extracted course data; the keywords represent assessment knowledge points of a first course related to the plurality of course videos; based on the keywords, multiple categories of question materials are extracted from all the course data, the incidence relation between different categories of question materials is determined, and the multiple categories of question materials are used for generating electronic questions of the first course.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of education and artificial intelligence, and particularly relates to an information processing method and device, an electronic device, a storage medium and a computer program product. BACKGROUND

[0002] In the field of education, with the development of artificial intelligence, the emergence of innovative technologies such as natural language processing (NLP), multi-modal, and knowledge graph effectively promotes the evolution of intelligent question setting from recommendation to generation, making the question setting process more intelligent. In related technologies, natural language processing technology is generally used for test question setting, and the types of questions involved generally include multiple-choice questions, true or false questions, connection questions, fill-in-the-blank questions, word fill-in-the-blank questions, and short answer questions, and the question setting mode is single. SUMMARY

[0003] To solve the problems in related technologies, the embodiments of the present application provide an information processing method, device, electronic device, storage medium and computer program product.

[0004] The technical solution of the embodiments of the present application is implemented as follows:

[0005] The embodiments of the present application provide an information processing method, characterized in that it comprises:

[0006] Extracting course data of each course video in a plurality of course videos; the course data comprises one or more categories of course introduction information, video image frames and video speech;

[0007] Generating a plurality of keywords based on all the extracted course data; the keywords represent the examination knowledge points of a first course related to the plurality of course videos;

[0008] Based on the keywords, extracting a plurality of categories of question materials from all the course data and determining the association relationship between the question materials of different categories; the plurality of categories of question materials are used to generate electronic questions of the first course.

[0009] In the above solution, the method further comprises:

[0010] Obtaining first information; the first information comprises target words and / or target sentences and / or question characteristics for examining the first course;

[0011] From the plurality of categories of question materials, extracting first question materials matched with the first information;

[0012] Extracting second question materials having a set association relationship with the first question materials;

[0013] generate an electronic question of the first course based on the first question material and the second question material.

[0014] In the above scheme, the generating of the plurality of keywords based on the extracted all course data comprises:

[0015] constructing a corresponding information sequence for the course data of each category based on the extracted all course data under each category;

[0016] generating the plurality of keywords based on the analysis result of the constructed all information sequences.

[0017] In the above scheme, the generating of the plurality of keywords based on the analysis result of the constructed all information sequences comprises:

[0018] performing text preprocessing on the constructed all information sequences;

[0019] performing analysis on the text preprocessed all information sequences based on a Term Frequency-Inverse Document Frequency (TF-IDF) algorithm to obtain an analysis result;

[0020] generating the plurality of keywords based on the analysis result.

[0021] In the above scheme, the generating of the plurality of keywords based on the analysis result comprises:

[0022] matching the analysis result with a knowledge point in a set knowledge graph corresponding to the first course, and generating the matching consistent knowledge point as the plurality of keywords.

[0023] In the above scheme, the constructing of the corresponding information sequence for the course data of each category based on the extracted all course data under each category comprises:

[0024] constructing a first information sequence based on the course data of a course introduction information category;

[0025] extracting text information from the video image frames using Optical Character Recognition (OCR) based on the course data of a video image frame category, and constructing a second information sequence based on the extracted text information;

[0026] converting the video speech into text information using Automatic Speech Recognition (ASR) based on the course data of a video speech category, and constructing a third information sequence based on the converted text information.

[0027] In the above scheme, the course data includes video image frames. Correspondingly, extracting the course data for each course video from multiple course videos includes:

[0028] Extract all video image frames from each of the multiple course videos; and / or,

[0029] Video image frames are extracted from each of the plurality of course videos at fixed frame intervals; and / or,

[0030] For each of the multiple course videos, video image frames are extracted based on the waveform features of the video audio.

[0031] This application also provides an information processing apparatus, characterized in that it includes:

[0032] The first extraction unit is used to extract course data from each of the multiple course videos; the course data includes one or more of the following categories: course introduction information, video image frames, and video audio;

[0033] The first generation unit is used to generate multiple keywords based on all extracted course data; the keywords represent the assessment knowledge points of the first course related to the multiple course videos;

[0034] The second extraction unit is used to extract multiple categories of question materials from all course data based on keywords, and to determine the correlation between different categories of question materials; the multiple categories of question materials are used to generate electronic questions for the first course.

[0035] This application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor is used to execute any of the above-described information processing methods.

[0036] This application also provides a readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of any of the above methods.

[0037] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above methods.

[0038] In this embodiment, course data from each course video in multiple course videos related to the first course is extracted to obtain course data of one or more categories. Based on all the extracted course data, multiple keywords representing the assessment knowledge points of the first course are generated. Based on the keywords, multiple categories of question materials are extracted from all course data, and the relationship between different categories of question materials is determined. Here, the generated question materials constitute a question material library related to the first course. Since the course data involves one or more categories of course introduction information, video image frames, and video audio, based on the question material library, questions containing text, image, audio, and video types can be generated, solving the technical problem of a single question generation mode. Attached Figure Description

[0039] Figure 1 This is a schematic diagram illustrating the implementation process of an information processing method according to an embodiment of this application.

[0040] Figure 2 This is a schematic diagram of the structure of an information processing device according to an embodiment of this application;

[0041] Figure 3 This is a schematic diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0042] In the field of education, traditional intelligent question generation is recommendation-based. This method has problems such as long construction time for question banks, low efficiency, poor matching between questions and the knowledge points being tested, and monotonous question patterns.

[0043] First, traditional recommendation-based question generation methods typically involve assembling test papers from pre-written questions according to certain algorithmic rules. These questions are then recommended using algorithms such as cosine similarity and Pearson correlation coefficient, taking into account the student's existing knowledge or the exam syllabus. However, a large number of questions in the question bank still require manual editing, resulting in a lengthy construction process and low efficiency. Second, traditional intelligent question generation algorithms generally rely on the subject's knowledge graph or the student's understanding of the subject matter, failing to consider the instructor's understanding of the overall knowledge system. Third, related technologies generally use natural language processing (NLP) for question generation, typically including multiple-choice, true / false, matching, fill-in-the-blank, vocabulary fill-in-the-blank, and short-answer questions—all text-based questions—resulting in a limited question generation model.

[0044] In this embodiment, course data from each course video in multiple course videos related to the first course is extracted to obtain course data of one or more categories. Based on all the extracted course data, multiple keywords representing the assessment knowledge points of the first course are generated. Based on the keywords, multiple categories of question materials are extracted from all course data, and the relationship between different categories of question materials is determined. Here, the generated question materials constitute a question material library related to the first course. Since the course data involves one or more categories of course introduction information, video image frames, and video audio, based on the question material library, questions containing text, image, audio, and video types can be generated, solving the technical problem of a single question generation mode.

[0045] The present application will now be described in further detail with reference to the accompanying drawings and embodiments.

[0046] Figure 1 This is a schematic diagram illustrating the implementation flow of an information processing method provided in an embodiment of this application, with reference to... Figure 1 The method includes:

[0047] Step 101: Extract course data for each course video from multiple course videos.

[0048] The course data includes one or more of the following categories: course introduction information, video image frames, and video audio.

[0049] In this embodiment, course data for each of multiple course videos is extracted, and these multiple course videos belong to the same category of courses. For example, course videos of the same category may be course videos in the same field, course videos of the same type of skill enhancement, or course videos from the same series. In practical applications, course videos can also be categorized according to the dimensions of the questions the question setter expects.

[0050] In this embodiment of the application, course data for each course video in multiple course videos of the same category can be extracted using technologies such as intelligent search engines, intelligent recommendation algorithms, knowledge graphs, and data mining.

[0051] In practical applications, course introduction information includes, but is not limited to: courseware, blackboard writing, or introductory materials. Among these, introductory materials include, but are not limited to, course objectives, course syllabus, target audience, course duration, lecturer introduction, or specific content of learning materials.

[0052] In practical applications, video image frames are extracted from the course video, including but not limited to video image frames at one or more or all frame moments in the course video.

[0053] In practical applications, video audio refers to the audio extracted from the audio files of the course video, including but not limited to the audio of all audio files in the course video and the audio of some audio files.

[0054] Step 102: Generate multiple keywords based on all extracted course data.

[0055] Among them, the keywords represent the assessment knowledge points of the first course related to the multiple course videos.

[0056] In this application embodiment, keywords include, but are not limited to, star rating, importance, type, and source, which are characteristic attributes used to describe the assessment knowledge points. The star rating or importance can be obtained by statistically analyzing the frequency or weight of each assessment knowledge point in the course video, ranking the assessment knowledge points according to frequency or weight, and then classifying or categorizing the ranked assessment knowledge points. For example, the higher the frequency or weight of an assessment knowledge point, the higher its star rating or importance. The type of keyword refers to whether the keyword is a single word or a sentence composed of multiple words. The source of the keyword refers to whether the keyword comes from course introduction information, video image frames, or video audio.

[0057] Step 103: Based on keywords, extract multiple categories of question materials from all course data and determine the relationships between different categories of question materials.

[0058] The aforementioned question materials from various categories are used to generate the electronic questions for this first course.

[0059] In this embodiment of the application, since the course data includes one or more categories of course introduction information, video image frames, and video audio, the topic materials extracted based on the course data may involve one or more types of text, audio, image, and video.

[0060] In this embodiment of the application, the relationship between text-type and voice-type question materials includes: inclusion relationship, contextual relationship, and similar relationship.

[0061] An inclusion relationship refers to the following: audio-based question materials contain the semantic content of text-based question materials, also known as audio-based question materials positively including text-based question materials; or text-based question materials contain the semantic content of audio-based question materials, also known as audio-based question materials negatively including text-based question materials. For example, the text-based question material might be "cloud + terminal," while the audio-based question material's semantic content is "Solving the problem of poor communication between examiners and examinees: a cloud + terminal solution for remote real-time audio and video interaction." That is, the audio-based question material contains the semantic content of "cloud + terminal," so the relationship between the text-based and audio-based question materials is an inclusion relationship, also known as audio-based question materials positively including text-based question materials.

[0062] Contextual relationship refers to the following: Audio-based question materials contain semantic content of text-based question materials, and the semantic content of the text-based question materials plays a transitional role within the semantic content of the audio-based question materials. This is also known as a positive contextual relationship between audio-based and text-based question materials. Alternatively, text-based question materials contain semantic content of audio-based question materials, and the semantic content of the audio-based question materials plays a transitional role within the semantic content of the text-based question materials. This is also known as a negative contextual relationship between audio-based and text-based question materials. For example, the text-type question material is "safety net solution," while the audio-type question material is: "To prevent various abnormal situations caused by network fluctuations, bandwidth limitations, traffic restrictions, etc. on the candidate's or examiner's side, the system has various safety net solutions, including a replay re-upload solution, a safety net solution for objective questions + essay questions, a re-upload solution for audio / video recording questions, and a re-upload solution for photo uploads." The audio-type question material not only contains the term "safety net solution," but also explains the reasons for its existence above it, namely, "to prevent various abnormal situations caused by network fluctuations, bandwidth limitations, traffic restrictions, etc. on the candidate's or examiner's side." Furthermore, it provides a detailed explanation of the "safety net solution," namely, "including a replay re-upload solution, a safety net solution for objective questions + essay questions, a re-upload solution for audio / video recording questions, and a re-upload solution for photo uploads." Therefore, the relationship between the audio-type and text-type question materials is a contextual relationship, specifically a positive contextual relationship.

[0063] A similarity relationship refers to the semantic similarity between words in audio-based question materials and words in text-based question materials. For example, the semantic content of the text-based question material is "data cockpit," while the semantic content of the audio-based question material is "command center." "Data cockpit" can also be used as "command center." The two words have similar semantic content, so the relationship between the text-based and audio-based question materials is a similarity relationship.

[0064] In practical applications, for ease of description, the relationship between text-based and audio-based question materials can be represented by the following formula:

[0065] T1[relation C1 / relation C2]S1;

[0066] T2[relation X1 / relation X2 / relation X3]S2;

[0067] T2[relation S]S3;

[0068] In this context, T1 and T2 represent different words and phrases in the text-based question materials, while S1, S2, and S3 represent different sounds in the audio-based question materials. The same word or phrase can be related to two different sounds, and the same sound can also be related to two different words or phrases. Square brackets "[]" indicate the relationship between the two sides of the brackets, and " / " inside the brackets indicates "or". Relation C represents inclusion, with C1 being positive inclusion and C2 being negative inclusion; Relation X represents context, with X1 being positive context, X2 being negative context, and X3 being a non-contextual relationship; Relation S represents similarity.

[0069] In this embodiment of the application, the relationship between text-type and image-type title materials includes: inclusion relationship and reference relationship.

[0070] An inclusion relationship refers to a situation where image-type question material contains the semantic content of text-type question material, also known as image-type question material positively including text-type question material; or text-type question material contains the semantic content of image-type question material, also known as image-type question material negatively including text-type question material. For example, if the text-type question material is "Cloud Examination Room," and the image-type question material is "Overall Architecture of the Cloud Examination Room System," meaning the image-type question material contains the semantic content of "Cloud Examination Room," then the relationship between the image-type question material and the text-type question material is an inclusion relationship, also known as image-type question material positively including text-type question material. For example, if the text-type question material is "Cloud Art Examination," and the image-type question material is "Application Scenarios of the Cloud Examination Room," since the Cloud Examination Room product system consists of different sub-products such as Cloud Art Examination, Cloud Postgraduate Examination, and Cloud Vocational Examination, meaning "Cloud Examination Room" contains the semantic content of "Cloud Art Examination, Cloud Postgraduate Examination, and Cloud Vocational Examination," then the relationship between the image-type question material and the text-type question material is an inclusion relationship, also known as text-type question material negatively including image-type question material.

[0071] Referential relationship refers to a situation where, although the semantic content of image-type question materials and text-type question materials do not appear to be identical or inclusive, the abstract conceptual description of one type of question material, such as its function, role, or detailed introduction, constitutes the semantic content of the other type. If the abstract conceptual description of the image-type question material is the semantic content of the text-type question material, then the image-type and text-type question materials have a positive referential relationship, while the text-type and image-type question materials have a negative referential relationship. For example, the text-type question material is "Separation of Assessment and Evaluation," but this text does not appear in the image-type question material. However, semantic analysis reveals that the image-type question material is a detailed illustration of the separation of assessment and evaluation. In other words, the abstract conceptual description of the image-type question material is the semantic content of the text-type question material. Therefore, the relationship between the image-type and text-type question materials is a referential relationship, meaning the image-type and text-type question materials have a positive referential relationship, while the text-type and image-type question materials have a negative referential relationship.

[0072] In practical applications, for ease of description, the following formula can be used to represent the relationship between text-type and image-type question materials:

[0073] T1[relation C1 / relation C2]P1;

[0074] T1[relation R1 / relation R2]P2;

[0075] In this context, T1 and T2 represent different words and phrases in the text-based question materials, and P1 and P2 represent different images in the image-based question materials. The same word or phrase can be related to two different images, and the same image can also be related to two different words or phrases. Square brackets "[]" indicate the relationship between the two sides of the brackets, and " / " inside the brackets indicates "or". Relation C represents inclusion, with C1 being positive inclusion and C2 being negative inclusion; Relation R represents reference, with R1 being positive reference (word or phrase refers to image) and R2 being negative reference (image refers to word or phrase).

[0076] In this embodiment of the application, the association between the image type and the video type of the question material includes: attribution relationship, that is, the image in the question material is a frame of the video.

[0077] In practical applications, for ease of description, the following formula can be used to represent the relationship between image-type and video-type question materials:

[0078] P1[relation B1 / relation B2]V1;

[0079] In this context, square brackets “[]” indicate the relationship between the two sides of the brackets, and “ / ” inside the square brackets means “or”. P1 represents different pictures in the picture-type question materials, V1 represents different videos in the video-type question materials, and relationship B represents the attribution relationship, with B1 being attribution and B2 being non-attribution.

[0080] In this embodiment of the application, the generated question materials constitute a question material library related to the first course. Based on the above question material library, questions including text type, image type, audio type and video type can be generated, which solves the technical problem of single question generation mode.

[0081] In one embodiment, the above method further includes:

[0082] Obtain the first information. This first information includes target words and / or target sentences and / or question features used to assess the first course.

[0083] Extract the first question material that matches the first piece of information from the multiple categories of question materials;

[0084] Extract the second question material that has a defined relationship with the first question material;

[0085] Based on the materials for the first and second questions, generate the electronic questions for this first course.

[0086] In this embodiment, the target words and / or target sentences are words or phrases proposed by the test creator based on the objective they need to assess. For example, if the test creator's objective is "intelligent solutions," then the target words and / or target sentences would be "intelligent cloud examination room," "remote examination," or "assessing the candidate's understanding of intelligent solutions." The test creator may choose not to specify target words and / or target sentences, in which case one or more keywords will be used as the default target words for the questions.

[0087] In this embodiment, the question features refer to the question mode, question type, and number involved in the question creation. The question mode includes one or more of text mode, voice mode, image mode, and video mode; the question type includes, but is not limited to, single-choice, multiple-choice, question-and-answer, and description types. For example, the question features include 10 single-choice questions in text mode, 10 single-choice questions in voice mode, 5 description questions in image mode, and 5 multiple-choice questions in video mode. The question creator may not specify a question mode and may use the default text mode.

[0088] In this embodiment of the application, a first question material matching the first information is extracted, wherein the type of the first question material includes text type, voice type, image type and video type.

[0089] In this embodiment of the application, a second question material with a predetermined association relationship with the first question material is extracted based on the association relationship between different categories of question materials. The types of the second question material include text, audio, image, and video.

[0090] In this embodiment, by acquiring first information, which includes target words and / or target sentences and / or question features for assessing the first course, first question materials matching the first information are extracted from multiple categories of question materials, and second question materials with a predetermined association relationship with the first question materials are extracted. Based on the first and second question materials, electronic questions for the first course are generated. Text, audio, images, and video information from the course videos can be extracted as materials and matched with target words and / or target sentences and / or question features to select matching materials, making the questions more targeted.

[0091] In one embodiment, based on all extracted course data, multiple keywords are generated, including:

[0092] Based on all the course data extracted under each category, a corresponding information sequence is constructed for the course data of each category;

[0093] Based on the parsing results of all the constructed information sequences, multiple keywords are generated.

[0094] Specifically, based on all the course data extracted from each category, a corresponding information sequence is constructed for the course data of each category, including:

[0095] Based on the course data categorized by course introduction information, construct the first information sequence;

[0096] Based on course data categorized by video image frame, OCR is used to extract text information from the video image frames, and a second information sequence is constructed based on the extracted text information.

[0097] Based on course data categorized by video and audio, ASR is used to convert video and audio into text information, and a third information sequence is constructed based on the converted text information.

[0098] In this embodiment of the application, course data of the course introduction information category is extracted using natural language processing technology to obtain text information and construct a first information sequence.

[0099] Based on the course data of video image frame categories, correspondingly, the course data of each course video in multiple course videos is extracted, including: extracting all video image frames of each course video in multiple course videos; extracting video image frames in each course video in multiple course videos at fixed frame intervals; and extracting video image frames in each course video in multiple course videos based on the waveform features of video speech.

[0100] In this embodiment of the application, before constructing the second information sequence, it is necessary to extract course data of the video image frame category of each course video from multiple course videos.

[0101] In this embodiment of the application, all video image frames of each course video in multiple course videos can be extracted to obtain all the information of the course video images. When the number of course videos is small or the memory usage is small, the method of extracting video image frames is fast.

[0102] In this embodiment of the application, video image frames are extracted from each of the multiple course videos at fixed frame intervals. For example, when the time for image changes in a video image exceeds a certain threshold, such as when the main content of a course video is to explain a courseware, and the courseware is turned over on average once every 5 seconds in the course video, video image frames are extracted from each of the multiple course videos at fixed frame intervals. For example, one video image is extracted at 10 frame intervals. If the course video contains 50 frames per second, then extracting one video image at 10 frame intervals can obtain all the information of the courseware, and is faster than the method of extracting all video image frames.

[0103] For each of multiple course videos, video image frames are extracted based on the waveform features of the video audio. For example, when the amplitude, standard deviation, or variance of the video audio waveform exceeds a certain threshold within a certain time period, a video image frame is extracted. For instance, if a teacher in a course video is explaining knowledge presented in a video image and makes a sound for 10 seconds, the amplitude, standard deviation, or variance of the video audio waveform within any 2 seconds of that 10 seconds will exceed the set threshold. This effectively identifies whether the teacher is speaking, allowing for the extraction of video image frames showing the teacher speaking.

[0104] In this embodiment of the application, after extracting the video image frames of each course video from multiple course videos, and before using OCR to extract text information from the video image frames, the following steps are included:

[0105] Identify fixed regions within video image frames of a course video. For example, these fixed regions are subtitles or courseware that typically appear at fixed locations in the video. Since the information contained in subtitles and courseware is generally important to the course video, identifying these fixed regions can improve the efficiency of extracting text information from video image frames using OCR.

[0106] In this embodiment of the application, identifying the fixed region, for example, includes: using an object detection algorithm to distinguish the border of the courseware region, and identifying the features of the courseware region by training a model. The object detection algorithm may be the YOLO (You Only Look Once) algorithm or the Faster R-CNN (Region Convolutional Neural Network) algorithm; and using background modeling techniques to distinguish between dynamic and static regions in the course video. The background modeling techniques include Gaussian Mixture Model (GMM) techniques.

[0107] In this embodiment, the "frame comparison method" is used to identify the fixed region. For example, after each video frame is identified, multiple text boxes are identified. The height and width of the identified text boxes are shown in Table 1. A total of 12 text boxes are identified across 5 frames. Most video frames have text boxes with a height of approximately 70 (frames 1, 2, and 3), with significant width variations. Based on this, text boxes with this height are determined to be subtitle text boxes. The more frames there are, the easier it is to distinguish subtitle text boxes using this method.

[0108] Table 1

[0109]

[0110]

[0111] In this embodiment of the application, after identifying fixed regions of video image frames based on course data of video image frame categories, text information is extracted from the video image frames using OCR, and a second information sequence is constructed based on the extracted text information.

[0112] In this embodiment, when using ASR to convert video speech into text information and constructing a third information sequence based on the converted text information, the text information converted by ASR is not accurate enough when the speech content in the video speech is in dialect or contains specialized terms. Therefore, when using speech recognition ASR to convert video speech into text information, an error correction dictionary is used to replace incorrect words with correct words. For example, the error correction dictionary can correct text errors through methods such as dictionary matching and edit distance calculation. For example, the error correction dictionary can correct text errors through a language model and based on contextual information. For example, the error correction dictionary can be a dictionary formed by statistically analyzing words with high error frequencies and their corresponding correct words. For example, the error correction dictionary can be an error correction dictionary for a specific subject area, formed by statistically analyzing vocabulary in that subject area.

[0113] In this embodiment, since course introduction information is generally structured text information collected by trainers or managers, and is the most direct interpretive text of the course video, the information sequence constructed from the course introduction information can be used to correct the information sequences obtained from video image frames and video audio, respectively. The first information sequence is used to correct the second information sequence to obtain a corrected second information sequence; the first information sequence is used to correct the third information sequence to obtain a corrected third information sequence. Correction methods include, but are not limited to: correction using natural language processing techniques, correction using error-correcting dictionaries, correction after matching using string matching and / or regular expression matching and / or word vector model methods, correction by replacing incorrect terms, and correction by adding missing words.

[0114] In this embodiment of the application, multiple keywords are generated based on the parsing results of the constructed first information sequence, second information sequence, and third information sequence.

[0115] In this embodiment, a first information sequence is constructed based on course data categorized by course introduction information. Based on course data categorized by video image frames, text information is extracted from the video image frames using OCR, and a second information sequence is constructed based on the extracted text information. Based on course data categorized by video audio, video audio is converted into text information using ASR, and a third information sequence is constructed based on the converted text information. Specifically, when acquiring video image frames, all video image frames of each course video from multiple course videos can be extracted; and / or, video image frames can be extracted from each course video from multiple course videos at fixed frame intervals; and / or, video image frames can be extracted from each course video from multiple course videos based on the waveform features of the video audio. The above methods can improve the efficiency and accuracy of constructing the first, second, and third information sequences, thereby facilitating the accurate and efficient generation of multiple keywords.

[0116] In one embodiment, based on the parsing results of all constructed information sequences, multiple keywords are generated, including:

[0117] Perform text preprocessing on all the constructed information sequences;

[0118] The text preprocessing includes: removing special symbols and / or numbers and / or stop words from the information sequence, and / or converting the letters in the information sequence to uppercase and lowercase;

[0119] Based on the TF-IDF algorithm, all information sequences after text preprocessing are parsed to obtain the parsing results;

[0120] Based on the analysis results, multiple keywords are generated.

[0121] In this embodiment of the application, before using the TF-IDF algorithm, all information sequences after text preprocessing are segmented to obtain multiple segmented fragments.

[0122] In this embodiment of the application, when using the TF-IDF algorithm, it is necessary to calculate the TF-IDF value. The TF-IDF value of the i-th word can be expressed by the following formula: TF-IDF=(T i / N)×log(M / D i ), where T i Let N be the number of times the i-th word appears in all information sequences, N be the total number of word segments in all information sequences, M be the total number of courses in multiple course videos, and D be the total number of courses in multiple course videos. i Let represent the number of courses containing the i-th word in the course videos.

[0123] In this embodiment of the application, when using the TF-IDF algorithm, the TF-IDF value of the information sequence corresponding to the course data of each category is calculated respectively.

[0124] In this embodiment of the application, by way of example, the words in the text are sorted from largest to smallest according to the statistical TF-IDF value, and the words with the highest ranking, i.e. the larger TF-IDF value, are selected to generate multiple keywords.

[0125] In this embodiment of the application, in order to extract keywords from all information sequences, the TextRank algorithm can be used instead of the TF-IDF algorithm to extract keywords.

[0126] In this embodiment, text preprocessing is performed on all constructed information sequences; the preprocessed information sequences are parsed based on the term frequency-inverse document frequency (TF-IDF) algorithm to obtain the parsing results; based on the parsing results, the multiple keywords are generated, which can comprehensively consider the term frequency of keyword fragments in the document and the importance of the entire course set, and can better express the key information of the course.

[0127] In one embodiment, based on the parsing results, multiple keywords are generated, including:

[0128] The analysis results are matched with the knowledge points in the set knowledge graph corresponding to the first course, and the matching knowledge points are generated as multiple keywords.

[0129] In this embodiment of the application, the parsing results are matched with the knowledge points in the set knowledge graph corresponding to the first course to find the same or similar knowledge points and generate multiple keywords. The similar knowledge points include: knowledge points with the same meaning but expressed in different words and phrases, and knowledge points with hierarchical relationships.

[0130] In this embodiment of the application, by matching the parsing results with the knowledge points in the set knowledge graph corresponding to the first course, and generating multiple keywords from the matching knowledge points, the correlation between keywords can be better discovered. For example, synonyms and words with hierarchical relationships can be better extracted, and a correlation relationship can be established between multiple keywords to form a more complete and concise question.

[0131] The following application examples further illustrate the solution of this application.

[0132] Step 1: Extract course data for each course video from multiple course videos in the "5G Smart Cloud Examination Room" category using technologies such as intelligent search engines, intelligent recommendation algorithms, knowledge graphs, and data mining.

[0133] Specifically, if intelligent recommendation algorithms are used, course videos of the same category are extracted by finding words with higher similarity to "5G Smart Cloud Examination Room," such as course videos related to smart universities or education and learning. If knowledge graph technology is used, course videos of the same category are extracted by finding words that are not highly similar to "5G Smart Cloud Examination Room" but have a related connection. Course videos of the same category are those containing words such as "cloud postgraduate entrance examination," "cloud art examination," and "cloud vocational examination," which are subordinate concepts of "cloud examination room." If data mining technology is used, course videos of the same category are extracted by finding words that represent different stages of development of "5G Smart Cloud Examination Room." Course videos of the same category are those containing words such as "standardized examination room," "cloud examination room," and "smart examination room."

[0134] The course data includes one or more of the following categories: course introduction information, video image frames, and video audio. Course introduction information includes, but is not limited to, courseware, whiteboard notes, or introductory materials. Introductory materials include, but are not limited to, course objectives, course outline, target audience, course duration, instructor introduction, or specific content of learning materials. Video image frames are video-related images extracted from the course video, including, but not limited to, full or partial screen images from one or more frames in the course video. Video audio is video-related audio extracted from the course video, including, but not limited to, the audio from all or part of the audio in the course video.

[0135] Step 2: Construct the first information sequence based on the course data of the course introduction information category.

[0136] Here, based on course data categorized by course introduction information, natural language processing techniques are used to extract textual information from this category, constructing the first information sequence. Course introduction information is typically structured text collected by trainers or managers and is the most direct interpretive text for the course videos.

[0137] Step 3: Based on the course data of video image frame categories, use OCR to extract text information from the video image frames, and construct a second information sequence based on the extracted text information.

[0138] After extracting video image frames from each of multiple course videos, and before extracting text information from the video image frames using OCR, the process includes: identifying fixed regions within the video image frames of the course videos. For example, these fixed regions are subtitles or courseware that typically appear at fixed locations in the video. Since the information contained in subtitles and courseware is generally important to the course videos, identifying these fixed regions can improve the efficiency of extracting text information from the video image frames using OCR.

[0139] Based on the course data of video image frame categories, after identifying fixed regions of video image frames, OCR is used to extract text information from the video image frames, and a second information sequence is constructed based on the extracted text information.

[0140] Step 4: Based on the course data of video and audio categories, use ASR to convert the video and audio into text information.

[0141] Here, when using ASR to convert video speech into text and constructing a third information sequence based on the converted text, the text converted by ASR is not accurate enough when the speech content in the video speech contains dialects or technical terms. Therefore, when using ASR to convert video speech into text, an error correction dictionary is used to replace incorrect words with correct words. A third information sequence is then constructed based on the converted text.

[0142] Step 5: Use the first information sequence to correct the second information sequence to obtain the corrected second information sequence; use the first information sequence to correct the third information sequence to obtain the corrected third information sequence.

[0143] The correction methods include, but are not limited to: using natural language processing techniques for correction, using error correction dictionaries for correction, using string matching and / or regular expression matching and / or word vector model methods for correction after matching, replacing erroneous terms for correction, and adding missing words for correction.

[0144] Step 6: Based on the parsing results of all constructed information sequences, generate multiple keywords.

[0145] First, all the constructed information sequences are preprocessed. This preprocessing includes removing special symbols and / or numbers and / or stop words from the information sequences, and / or converting the capitalization of letters in the information sequences. Then, the preprocessed information sequences are segmented into multiple word segments.

[0146] Based on the TF-IDF or TextRank algorithm, word frequency statistics are performed on all information sequences after text preprocessing. Multiple words with high frequency rankings are selected to obtain the parsing results. Based on these parsing results, they are matched with knowledge points in the designated knowledge graph corresponding to the first course to identify identical or similar knowledge points and generate multiple keywords.

[0147] Step 7: Obtain first information including target words and / or target sentences and / or question features for assessing the first course.

[0148] In this embodiment, the target word and / or target sentence are words or sentences proposed by the question setter according to the target to be tested. The question characteristics are the question pattern, question type, and number involved in this question setting. Among them, the question pattern includes one or more of the following: text pattern, voice pattern, image pattern, and video pattern; the question type includes, but is not limited to, single-choice, multiple-choice, question-and-answer, and descriptive types.

[0149] Step 8: From the multiple categories of question materials, extract the first question material that matches the first information, and extract the second question material that has a set association relationship with the first question material;

[0150] Step 9: Based on the first and second question materials, generate the electronic questions for the first course.

[0151] In this embodiment of the application, exemplarily, the question is a text-based single-choice question. The first question material is an image-type question material, the semantic content of which is "The examiner in the cloud examination room waiting area can mark the candidate's device status and communicate with the candidate via video and text." The second question material is a text-type question material with an inclusion relationship with the first question material, namely "examiner in the waiting area." This inclusion relationship means that the image in the first question material contains the semantic content of the second question material; that is, the image in the first question material contains the semantic content of "examiner in the waiting area." Then, based on the second question material, similar words such as "chief examiner," "invigilator," and "proctor" are generated as confusing answers. Alternatively, if the question is a text-based single-choice question, then the question stem is: "The cloud examination room () can mark the candidate's device status and communicate with the candidate via video and text." The options for this electronic question are: A. Invigilator; B. Proctor; C. Chief examiner; D. Waiting area examiner. The answer to this electronic question is: D.

[0152] In this embodiment of the application, exemplarily, the question is a text-based multiple-choice question. The first question material is an image-based question material containing the text "5G Smart Cloud Examination Room, based on 5G+cloud+large video capabilities, creates an integrated online+offline remote examination solution to solve problems such as normalized epidemic prevention and control, the time-consuming and labor-intensive nature of offline examinations, and the high cost of on-site examinations." The text-based question materials "5G," "cloud," and "large video," which have an inclusion relationship with the first question material, are selected as the second question material. This inclusion relationship means that the image in the first question material contains the semantic content of the second question material; that is, the image in the first question material contains the semantic content of "5G," "cloud," and "large video." Then, based on the second question material, a confusing answer, "dedicated line," is generated. Additionally, the question is a text-based multiple-choice question. The question stem for this electronic exam is: The 5G Smart Cloud Exam Room, based on the capabilities of ()+()+(), creates an integrated online and offline remote examination solution to address issues such as normalized epidemic prevention and control, the time and effort required for organizing offline exams, and the high cost of on-site exams. The options for this electronic exam are A. 5G; B. Cloud; C. Large video; D. Dedicated line. The answer to this electronic exam is ABC.

[0153] In this embodiment of the application, exemplarily, the question features are single-choice questions in both voice and text modes. The first question material is a voice-based question material whose semantic content is "The command center can provide information to educational management units, facilitating their intuitive understanding of the current examination situation, such as the total number of examination rooms, the number of examination rooms already opened, the number of examinees, the number of examination subjects, etc." The second question material is a text-based question material with a similar relationship to the first question material, namely, "data cockpit." This similarity means that the semantic content of the first question material is similar to the semantic content of the second question material; that is, "data cockpit" is similar to the semantic content of "command center" in "The command center can provide information to educational management units, facilitating their intuitive understanding of the current examination situation, such as the total number of examination rooms, the number of examination rooms already opened, the number of examinees, the number of examination subjects, etc." Then, based on the second question material, semantically similar words and phrases are selected to generate confusing answers such as "big data technology," "emergency command center," and "examination organization." The question stem for this electronic question is: Please find the content being introduced based on the audio (audio playback included); the options for this electronic question are: A. Big Data Technology; B. Data Dashboard; C. Emergency Command Center; D. Examination Organization. The answer to this electronic question is B.

[0154] In this embodiment of the application, exemplarily, the question's characteristics are image-based descriptive questions. The first question material is an image-type question material, specifically a customer case study of a cloud exam room product. The second question material is a text-type question material that has a referential relationship with the first question material, namely, "cloud exam room". This referential relationship means that although the semantic content of the second question material does not appear in the first question material, the abstract conceptual description of the first question material, that is, the abstract conceptual description of the image-type question material, refers to the semantic content of the second question material. In other words, semantic analysis shows that "customer case study of cloud exam room product" in the image refers to "cloud exam room". Therefore, the question stem of the electronic question is: Please describe the information related to "cloud exam room" shown in the image in 100 words (with image attached).

[0155] In this embodiment of the application, exemplarily, the question features a video-type fill-in-the-blank question. The first question material is a video-type question material, and the information in this video-type question material is: an introduction to the core advantages of the cloud examination room. The second question material is an image-type question material that has a relationship with the first question material, and this relationship means that the image in the second question material is a frame of the video of the first question material. Among them, the text information in the image-type second question material is: "The core advantages of the cloud examination room product are system security and stability, and security and no leakage of examination information. Specifically, it includes four aspects: no loss of information security, no leakage of examination papers security, system security and stability, and backup of file information." Optionally, text-type question materials that have an inclusion relationship with the second question material are extracted, namely "information security", "examination papers", "system", and "file". In addition, the question features are: The question stem of the electronic question is: Please fill in the blanks according to the video information (with video). The content of the electronic question is: The core advantages of the cloud examination room product are system security and stability, and security and no leakage of examination information, specifically including four aspects: no loss of _____, no leakage of _____, security and stability of _____, and backup of _____ information.

[0156] To implement the information processing method of this application embodiment, this application embodiment also provides an information processing apparatus, such as... Figure 2 As shown, the device includes:

[0157] The first extraction unit 201 is used to extract course data from each of the multiple course videos; the course data includes one or more of the following categories: course introduction information, video image frames, and video audio;

[0158] The first generation unit 202 is used to generate multiple keywords based on all extracted course data, wherein the keywords represent the assessment knowledge points of the first course related to the multiple course videos.

[0159] The second extraction unit 203 is used to extract multiple categories of question materials from all course data based on keywords, and to determine the correlation between different categories of question materials. The multiple categories of question materials are used to generate the electronic questions for the first course.

[0160] In one embodiment, the device further includes:

[0161] The first acquisition unit is used to acquire first information; wherein, the first information includes target words and / or target sentences and / or question features for assessing the first course;

[0162] The third extraction unit is used to extract the first question material that matches the first information from the multiple categories of question materials;

[0163] The fourth extraction unit is used to extract second question materials that have a predetermined relationship with the first question material;

[0164] The second generation unit is used to generate electronic questions for the first course based on the first question material and the second question material.

[0165] In one embodiment, the first generating unit 202 is further configured to:

[0166] Based on all the course data extracted under each category, a corresponding information sequence is constructed for each category of course data; based on the parsing results of all the constructed information sequences, multiple keywords are generated.

[0167] In one embodiment, the first generation unit 202 generates multiple keywords based on the parsing results of all constructed information sequences, including: performing text preprocessing on all constructed information sequences. This text preprocessing includes: removing special symbols and / or numbers and / or stop words from the information sequences, and / or converting the capitalization of letters in the information sequences; parsing all preprocessed information sequences based on the TF-IDF algorithm to obtain parsing results; and generating multiple keywords based on the parsing results.

[0168] In one embodiment, the first generation unit 202 generates multiple keywords based on the parsing results, including: matching the parsing results with knowledge points in the set knowledge graph corresponding to the first course, and generating multiple keywords from the matching knowledge points.

[0169] In one embodiment, the first generation unit 202 constructs a corresponding information sequence for each category of course data based on all course data extracted under each category, including: constructing a first information sequence based on course data of the course introduction information category; extracting text information from video image frames using OCR based on course data of the video image frame category, and constructing a second information sequence based on the extracted text information; and converting video speech into text information using ASR based on course data of the video speech category, and constructing a third information sequence based on the converted text information.

[0170] In one embodiment, the course data includes video image frames, and correspondingly, the first extraction unit 201 is used to:

[0171] Extract all video image frames from each of the multiple course videos; extract video image frames from each of the multiple course videos at fixed frame intervals; and extract video image frames from each of the multiple course videos based on the waveform features of the video audio.

[0172] It should be noted that the information processing apparatus provided in the above embodiments is only illustrated by the division of the above program modules. In practical applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the apparatus can be divided into different program modules to complete all or part of the processing described above. In addition, the information processing apparatus and the information processing method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0173] Based on the hardware implementation of the above program modules, embodiments of this application also provide an electronic device, such as... Figure 3 As shown, the electronic device 300 includes:

[0174] Communication interface 301 enables information exchange with other network nodes;

[0175] The processor 302 is connected to the communication interface 301 to enable information interaction with other network nodes and to execute the methods provided by one or more of the above-mentioned technical solutions when running a computer program. The computer program is stored in the memory 303.

[0176] When processor 302 runs a computer program to execute one or more of the above information processing methods, specifically:

[0177] Processor 302 is used for:

[0178] Extract course data from each of multiple course videos; the course data includes one or more of the following categories: course introduction information, video image frames, and video audio; generate multiple keywords based on all extracted course data; the keywords represent the assessment knowledge points of the first course related to the multiple course videos; based on the keywords, extract multiple categories of question materials from all course data and determine the correlation between different categories of question materials; the above multiple categories of question materials are used to generate electronic questions for the first course.

[0179] In one embodiment, the processor 302 is further configured to:

[0180] Obtain first information; wherein, the first information includes target words and / or target sentences and / or question features used to assess the first course;

[0181] Extract the first question material that matches the first piece of information from the multiple categories of question materials;

[0182] Extract the second question material that has a defined relationship with the first question material;

[0183] Based on the materials for the first and second questions, generate the electronic questions for this first course.

[0184] In one embodiment, the processor 302 is used to:

[0185] Based on all the course data extracted under each category, a corresponding information sequence is constructed for each category of course data; based on the parsing results of all the constructed information sequences, multiple keywords are generated.

[0186] In one embodiment, the processor 302 is configured to: perform text preprocessing on all the constructed information sequences. The text preprocessing includes: removing special symbols and / or numbers and / or stop words from the information sequences, and / or converting the capitalization of letters in the information sequences; parsing all the preprocessed information sequences based on the TF-IDF algorithm to obtain parsing results; and generating multiple keywords based on the parsing results.

[0187] In one embodiment, the processor 302 is configured to: match the parsing results with the knowledge points in the set knowledge graph corresponding to the first course, and generate multiple keywords from the matching knowledge points.

[0188] In one embodiment, the processor 302 is configured to: construct a first information sequence based on course data of course introduction information category; extract text information from video image frames using OCR based on course data of video image frame category, and construct a second information sequence based on the extracted text information; and convert video speech into text information using ASR based on course data of video speech category, and construct a third information sequence based on the converted text information.

[0189] In one embodiment, the processor 302 is used to:

[0190] Extract all video image frames from each of the multiple course videos; extract video image frames from each of the multiple course videos at fixed frame intervals; and extract video image frames from each of the multiple course videos based on the waveform features of the video audio.

[0191] It should be noted that the specific processing procedures of processor 302 and communication interface 301 can be understood by referring to the above method.

[0192] Of course, in practical applications, the various components in electronic device 300 are coupled together through bus system 304. It can be understood that bus system 304 is used to realize the connection and communication between these components. In addition to a data bus, bus system 304 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, in... Figure 3 The general designated all buses as Bus System 304.

[0193] The memory 303 in this embodiment is used to store various types of data to support the operation of the electronic device 300. Examples of such data include any computer program used to operate on the electronic device 300.

[0194] The methods disclosed in the embodiments of this application can be applied to or implemented by the processor 302. The processor 302 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in the processor 302 or by instructions in the form of software. The processor 302 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 302 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in the memory 303. The processor 302 reads the information in the memory 303 and combines its hardware to complete the steps of the aforementioned method.

[0195] In an exemplary embodiment, the electronic device 300 may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the aforementioned method.

[0196] It is understood that the memory 303 in this embodiment can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memories described in the embodiments of this application are intended to include, but are not limited to, these and any other suitable types of memories.

[0197] In an exemplary embodiment, this application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, such as a memory 303 storing a computer program, which can be executed by the processor 302 of the electronic device 300 to complete the steps described in the aforementioned first network node-side method. The computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM.

[0198] For example, embodiments of this application also provide a computer program product, including a computer program that can be executed by a processor 302 of an electronic device 300 to complete the steps described in the aforementioned first network node-side method.

[0199] It should be noted that terms such as "first" and "second" are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0200] Furthermore, the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0201] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application.

Claims

1. An information processing method characterized by comprising: The method comprises: extracting course data of each course video in the plurality of course videos; the course data comprises one or more of the following categories: course introduction information, video image frames, and video speech; based on all the extracted course data, generating a plurality of keywords; the keywords represent the examination knowledge points of a first course related to the plurality of course videos; based on the keywords, extracting a plurality of categories of question materials from all the course data, and determining the association relationship between the question materials of different categories; the plurality of categories of question materials are used to generate electronic questions of the first course.

2. The method of claim 1, wherein, The method further comprises: obtaining first information; the first information comprises target words and / or target sentences and / or question characteristics used to examine the first course; from the plurality of categories of question materials, extracting first question materials matched with the first information; extracting second question materials having a set association relationship with the first question materials; based on the first question materials and the second question materials, generating electronic questions of the first course.

3. The method of claim 1, wherein, The method further comprises: based on all the extracted course data of each category, constructing a corresponding information sequence for the course data of each category; based on the analysis results of all the constructed information sequences, generating a plurality of keywords.

4. The method of claim 3, wherein, The method further comprises: text preprocessing is performed on all the constructed information sequences; based on the term frequency-inverse document frequency (TF-IDF) algorithm, analyzing all the information sequences after text preprocessing to obtain analysis results; based on the analysis results, generating the plurality of keywords.

5. The method of claim 4, wherein, The method further comprises: matching the analysis results with knowledge points in a set knowledge graph corresponding to the first course, and generating the matching consistent knowledge points as the plurality of keywords.

6. The method of claim 3, wherein, The method further comprises: based on the course data of the course introduction information category, constructing a first information sequence; based on the course data of the video image frame category, using optical character recognition (OCR) to extract text information from the video image frames, and constructing a second information sequence based on the extracted text information; based on the course data of the video speech category, using automatic speech recognition (ASR) to convert the video speech into text information, and constructing a third information sequence based on the converted text information.

7. The method of claim 1, wherein, The course data comprises video image frames, and correspondingly, the method of extracting course data of each course video in the plurality of course videos comprises: extracting all the video image frames of each course video in the plurality of course videos; and / or, extracting video image frames at a fixed frame interval in each course video of the plurality of course videos; and / or, based on the waveform characteristics of the video speech, extracting video image frames in each course video of the plurality of course videos.

8. An information processing apparatus, characterized by comprising: The method comprises: a first extraction unit configured to extract course data of each course video in the plurality of course videos; The course data comprises one or more of the following categories: course introduction information, video image frames and video speech; The first generating unit is configured to generate a plurality of keywords based on the extracted course data; The keywords represent examination knowledge points of a first course related to the plurality of course videos; The second extracting unit is configured to extract a plurality of categories of question materials from the course data based on the keywords, and determine an association relationship between the different categories of question materials; the plurality of categories of question materials are used to generate electronic questions of the first course.

9. An electronic device, comprising: The computer program is stored in the memory and executable on the processor, and the processor is configured to execute the information processing method in any one of claims 1 to 7.

10. A storage medium having stored thereon a computer program, characterized in that The computer program is stored in the memory and executable on the processor, and the processor is configured to execute the information processing method in any one of claims 1 to 7.

11. A computer program product comprising a computer program, characterized in that, The computer program is stored in the memory and executable on the processor, and the processor is configured to execute the information processing method in any one of claims 1 to 7.

Images