TW202042561A - Advertisement putting method, system and device based on online education and storage medium - Google Patents

Abstract

The invention provides an advertisement putting method, system and device, and a storage medium. The advertisement putting method comprises the steps: segmenting at least one online education recorded video of a first type of users into a plurality of video paragraphs according to a preset duration; adding at least one video tag to the video paragraph according to the content information of the video paragraph; performing expression recognition on the first type of users in each video paragraph, and adding at least one learning state tag; screening the video paragraphs according to preset learning state tags, and generating at least one advertisement keyword according to the video tags of the selected video paragraphs; searching at least one piece of advertisement information from the advertisement pool according to the advertisement keywords, and putting advertisement information for the first type of users.

Description

Advertisement placing method, system, equipment and computer readable recording medium based on online learning

The present invention relates to an advertisement placement method, system, equipment and computer readable recording medium, in particular to an advertisement placement method, system, equipment and computer readable recording medium based on online learning.

In today's Internet environment, each system has its own technical strengths and characteristics for advertising, such as Google's keyword search, advertising in the social environment, and intermittent advertising of audio-visual media. However, the existing teaching advertisement delivery method is rigid and rigid, unable to provide corresponding advertisement information according to the learning status of the students, and the conversion rate of the advertisement is very low.

Therefore, one of the objectives of the present invention is to provide an online learning-based advertising method that can improve the prior art.

Therefore, the online learning-based advertisement placement method of the present invention includes: step S110: dividing a recorded video showing a first type of user online learning into a plurality of video segments according to a preset time length; step S120: according to the videos The content information of at least one video segment of the paragraph adds a video tag to the video segment; step S130: perform expression recognition on the first type of user in the video segment, when it is determined that the expression of the first type of user belongs to the same expression When the ratio of the cumulative time length of the category to the total time length of the video segment exceeds a preset threshold, add a learning status tag corresponding to the expression type to the video segment; step S140: According to one of the learning status tags Or a plurality of preset learning status tags filter the video segments, and generate an advertisement keyword according to the video tag of each selected video segment; step S150: according to each advertisement keyword from an advertisement pool An advertisement information is searched out, and the advertisement information is delivered to the first type of user.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, step S120 includes: Step S121: performing graphic recognition on at least one frame of the video paragraph to obtain the first text corresponding to the video paragraph , Perform voice recognition on the movie paragraph to obtain a second text corresponding to the movie paragraph; Step S122: Obtain a word as the movie paragraph according to the first text and the second text corresponding to the movie paragraph Video tags.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, step S122 is to use the word with the highest total number of occurrences in the first text and the second text as the video tag of the video paragraph.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, step S122 is to use the word with the highest number of occurrences among the words that appear in both the first text and the second text as the video paragraph. The video tag.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, in step S130, the learning state label is a positive learning state label corresponding to a positive learning expression type or a negative learning corresponding to a negative learning expression type. Status label.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, step S140 includes: filtering the video segments according to the active learning status tag, and generating the video segment based on the video tag of the filtered video segment Advertising keywords; Step S150 includes: searching out an event advertisement information from the advertisement pool according to the advertisement keywords, and placing the event advertisement information to the first type of users.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, step S140 includes: filtering the video segments according to the negative learning status tag, and generating the video segment based on the video tag of the filtered video segment Advertising keywords; Step S150 includes: searching out a course advertisement information from the advertisement pool according to the advertisement keywords, and delivering the course advertisement information to the first type of users.

In some implementations of the online learning-based advertisement placement method of the present invention, the advertisement placement method further includes step S100 before step S110: preset an associated second type of user for the first type of user; and, Step S150 includes: delivering the advertisement information to one of the first type of user and the second type of user.

In some implementation aspects of the online learning-based advertisement delivery method of the present invention, step S150 further includes: generating an advertisement push information, and sending the advertisement push information to the second type of user, and the advertisement push information includes the advertisement Information, and the video segment used to generate the advertising information.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, the recorded video is a video recorded in real time when a teacher and a student conduct one-to-one online learning, and step S110 is every time a preset period of time has elapsed. The part of the recorded video corresponding to the preset time length is regarded as one of the video segments.

In some implementation aspects of the online learning-based advertisement placement method of the present invention, the preset time length is between 10 seconds and 10 minutes.

Another object of the present invention is to provide an advertisement placement system based on online learning that can implement the advertisement placement method.

The online learning-based advertising system of the present invention is used to implement the online learning-based advertising method as described in any of the foregoing implementation modes, and includes a video paragraph generation module, a video tag adding module, and a video tag Add a module, an advertising keyword generation module, and an advertising information delivery module. The video segment generation module is used to divide a recorded video showing a first-type user for online learning into multiple video segments according to a preset time length. The video tag adding module is used for adding a video tag to the video segment according to the content information of at least one of the video segments. The status tag adding module is used to recognize the expression of the first type of user in the segment of the movie, and when it is determined that the expression of the first type of user belongs to the same type of expression, the cumulative length of time accounts for the total time of the segment of the movie When the ratio of the length exceeds a preset threshold, a learning status tag corresponding to the expression type is added to the video segment. The advertising keyword generation module is used to filter the video segments based on one or more preset learning status tags in the learning status tags, and generate a video tag based on each video segment that is filtered out Advertising keywords. The advertisement information delivery module is used to search out an advertisement information from an advertisement pool according to each advertisement keyword, and deliver the advertisement information to the first type of user.

Another object of the present invention is to provide an advertisement placement device based on online learning that can implement the advertisement placement method.

The online learning-based advertisement placement device of the present invention includes a processor and a memory storing executable instructions that can be executed by the processor, and the processor is configured to execute the executable instructions as described above Any of the online learning-based advertisement placement methods described in any implementation mode.

Another object of the present invention is to provide a computer-readable recording medium that enables an electronic device to implement the advertisement placement method.

The computer of the present invention can read the recording medium and store an application program. When the application program is loaded and executed by an electronic device, the electronic device can implement the online learning-based advertisement placement as described in any of the foregoing implementation modes. method.

The effect of the present invention is that the online learning-based advertising method, system, equipment and computer-readable recording medium can confirm the learning content and learning status of each user based on the recorded video of online learning, and produce targeted Moreover, the flexible and changeable advertising information can be accurately applied to various types of users to increase the accuracy and conversion rate of advertising information, and improve the humanized experience.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are represented by the same numbers. In addition, the "electrical connection" mentioned in this patent specification generally refers to a wired electrical connection between multiple electronic devices/devices/components connected through conductive materials, and a radio connection through wireless communication technology for wireless signal transmission . Moreover, the “electrical connection” mentioned in this patent specification also refers to the “direct electrical connection” formed by the direct connection between two electronic equipment/devices/components, and the “direct electrical connection” between two electronic equipment/devices/components "Indirect electrical connection" formed by connecting other electronic equipment/devices/components.

It should be understood that the present invention can be implemented in a variety of different modes, and the following embodiments are used to enable the present invention to be described more comprehensively and completely, so that those with ordinary knowledge in the technical field can follow this The patent specification is used to implement the technical means proposed by the present invention, and is not used to limit the scope of implementation of the present invention.

An embodiment of the online learning-based advertisement placement method of the present invention is implemented by, for example, an online learning-based advertisement placement system, and FIG. 1 is a flowchart of the online learning-based advertisement placement method of this embodiment. As shown in FIG. 1, the online learning-based advertisement placement method of this embodiment includes the following steps.

Step S110: Divide at least one recorded video of online learning for the first type of user into multiple video segments according to a preset time length. More specifically, the advertisement delivery system divides the recorded video showing the online learning of the first-type user into the video segments according to the preset time length.

Step S120: Add at least one video tag to the video segment according to the content information of at least one of the video segments. More specifically, in this embodiment, the advertisement delivery system adds one or more video tags to each video segment based on the content information of each video segment.

Step S130: Perform facial expression recognition for the first-type user in each movie segment and add at least one learning status tag. More specifically, the advertisement delivery system performs facial expression recognition on the first type of user in the video segment, and when the advertisement delivery system determines that the first type of user’s facial expressions belong to the same type of facial expression, the cumulative length of time accounts for When the proportion of the total time length of the video segment exceeds a preset threshold, a predetermined learning status tag corresponding to the expression type is added to the video segment.

Step S140: Filter video segments according to preset learning status tags, and generate at least one advertisement keyword based on the video tags of the selected video segments. More specifically, the advertisement delivery system in this embodiment, for example, screens the video segments according to one or more preset learning status tags (here as target learning status tags) among the learning status tags. , And filter out all the video segments that have the learning status tag that meets the target learning status tag(s) from the video segments, and then generate an advertisement keyword based on the video tag of each video segment that is filtered out.

Step S150: Search for at least one piece of advertisement information based on advertisement keywords from an advertisement pool, and deliver advertisement information to the first type of users. More specifically, the advertisement delivery system in this embodiment, for example, searches for advertisement information corresponding to the advertisement keyword from the advertisement pool according to each advertisement keyword, and delivers the searched advertisement information to the first type of user All advertising information.

It is supplemented that, in this embodiment, the first type of users refers to, for example, students who learn through online courses. In addition, the recorded video may be, for example, a video recorded in real time during a one-to-one online course between a teacher and a student, and the recorded video includes, for example, a part showing the upper body of the teacher and a showing The upper body part of the student, and a part presenting the teaching materials related to the online course. In this embodiment, the teaching materials may be reference materials used by the teacher to assist in teaching, or It is the blackboard written by the teacher in the course of teaching, but it is not limited to this.

In this embodiment, the online learning-based advertising method divides the entire recorded video into the video segments and adds tags (that is, the video tags and/or learning status tags), so that each video segment All have at least one label. In addition, in this embodiment, facial expression recognition is performed on the first-type user (that is, the student) in each video segment to add learning status tags representing various facial expression types, so as to specifically distinguish the first-type user in each film tag And accurately determine the relevant advertising keywords to deliver the most appropriate advertising information to users. Regarding the foregoing advertising information delivery methods, in addition to the delivery methods described in this embodiment, in other embodiments, the advertising delivery system's delivery methods for different types of users and advertising information goals may be different from this embodiment. For example, in this embodiment, the environment in which the advertising delivery method is applied is an online virtual classroom, which includes the interaction between teachers and students in audio and video, and the advertising information delivery targets include students and teachers. And the content of the advertisements are all related to teaching and learning, for example, and the purpose of advertising is not only to attract the target to purchase goods and participate in activities, but also includes the release and prompts of various information.

The present invention applies the concept of advertising information delivery to online virtual classrooms, and its purpose is different from general Internet (Internet) advertising. In addition to promoting the company’s products and activities to customers, the present invention also provides a platform for information delivery. Communicate and convey information to customers. In view of the shortcomings of the existing technology in the online virtual classroom, the unique operation mode and habits in the class are added to produce a display effect different from general advertising.

In a preferred embodiment, step S120 includes the following steps.

Step S121: Perform image and text recognition on at least one frame in the movie paragraph to obtain the first text of the movie paragraph. And perform voice recognition on the movie paragraph to obtain the second text of the movie paragraph. More specifically, the advertisement delivery system in this embodiment, for example, performs image and text recognition on each frame in the movie paragraph to obtain the first text corresponding to the movie paragraph, and the advertisement delivery system also Perform voice recognition on each video segment to obtain a second text corresponding to the video segment. However, in other embodiments, the advertisement delivery system may also perform voice recognition on only one of the video segments, and perform image and text recognition on one of the frames of the video segment. , But not limited to this embodiment.

Step S122: Obtain at least one word with the highest number of occurrences as the movie tag of the movie paragraph according to the first text and the second text corresponding to the movie paragraph. More specifically, in this embodiment, the advertisement delivery system obtains one or more words as the video(s) of the video segment, for example, according to the first text and the second text corresponding to each video segment label.

In the present invention, the tag setting of the video paragraphs of any teaching video is performed completely through the program method. In addition, to creatively extract a certain number of picture frames from the movie paragraph, the prior art image and text recognition method can be used to obtain the first text content about the blackboard writing content from the blackboard writing part of the picture as the first text, and also use the voice of the prior art Recognition means, the text of the explanation content obtained from the teacher’s explanation on the blackboard part as the second text. In the present invention, at least one word that appears most frequently in the first text and the second text is used as a movie tag of the main content of the paragraph of the movie. It is worth noting that, in this embodiment, the part of the recorded video showing the teacher (that is, the teaching video) is a special video format. More specifically, the teacher does not follow the teaching material in the teaching video. Materials (such as writing on the blackboard) are read word by word, but will be combined with the content of the textbook materials for extended explanations or case introductions. The content of textbook materials is usually relatively concise. The speech recognition of the teaching content cannot determine which part of the textbook data the teacher explains in each video paragraph. Therefore, determining the video tag based on the textbook data alone will not only make the accuracy of the video tag low, but also cannot reflect it. Teacher's teaching habits. On the other hand, if only the speech of the teacher's teaching content is recognized to determine the video tag, the process of determining the video tag will be interfered by the teacher's personal teaching habits, etc., which is also detrimental to the accuracy of the video tag. Therefore, the present invention combines the different texts obtained in two ways (namely, the first text and the second text). In this way, each video tag can be closer to the main content of the corresponding video paragraph, and greatly increase The accuracy of the definition of the movie tag is thus, the present invention can provide an optimized way of adding movie tags for the unique attributes of the teaching movie while taking into account the calculation cost.

In a preferred embodiment, step S122 includes using at least one word with the highest total number of occurrences in the first text and the second text as the movie tag of the movie paragraph, but it is not limited to this. More specifically, in the described embodiment, the advertisement delivery system uses the word with the highest total number of occurrences in the first text and the second text as the video tag of the video paragraph. For example: in the first text, "engine" appeared 4 times, "maintenance" appeared 3 times, and the remaining words appeared up to 2 times, while in the second text, "engine" appeared 8 times, "maintenance" Appears 6 times, and the remaining words appear 2 times at most. The two words with the highest total number of occurrences in the first text and the second text are "engine" and "maintenance". In this example, the advertising system will "Engine" and "Maintenance" are used as two video tags in the video segment.

In a preferred embodiment, step S122 includes using at least one word that appears in both the first text and the second text and has the highest number of occurrences as the movie tag of the movie paragraph. More specifically, in the described preferred embodiment, the advertisement delivery system uses, for example, the word with the highest number of occurrences among all words that appear in both the first text and the second text as the video The video tag of the paragraph. For example: In the first text, "Li Bai" appears 3 times, "Du Fu" appears once, and the remaining words appear up to 2 times and do not contain "Ancient Poetry", while in the second text, "Li Bai" appears Two times, "Ancient Poem" appeared 6 times, "Du Fu" appeared once, and the remaining words appeared up to 2 times. In this example, although the total number of occurrences of "ancient poems" is the largest (that is, 6 times in the second text), in the described embodiment, in order to more comprehensively consider the comprehensive effect of the teaching materials and the teacher's dictation , The ad delivery system will use the "Li Bai" that appears in both the first text and the second text and has the highest number of appearances (appears 5 times in the first and second text in total) as the video tag of the video paragraph.

In a preferred embodiment, in step S130, when the ratio of the user expression obtained by the expression recognition to the duration of the video segment exceeds a preset threshold, a preset learning status tag corresponding to the user expression is added to the video segment . More specifically, when the advertisement delivery system determines that the proportion of the cumulative time length of the first type of user’s expressions belonging to the same expression category to the total time length of the video segment exceeds the preset threshold, it adds to the video segment The learning status label corresponding to the expression type. For example: when the cumulative length of time that the first type of user in a video segment shows the same expression type of expression exceeds 40% of the total time length of the video segment, the advertising system will correspond to the expression type The learning status label of is used as the learning status label of the video segment. In other words, in this example, the preset threshold is implemented as 40%, but it is not limited to this.

In a preferred embodiment, in step S130, the learning state label includes a positive learning state label mapped to a preset positive learning expression and a negative learning state label mapped to a preset negative learning expression. More specifically, in the preferred embodiment, the expression types include a positive learning expression type and a negative learning expression type, and each learning state label can be, for example, a corresponding to the positive learning expression type The positive learning state label, or the negative learning state label corresponding to the negative learning expression type. Furthermore, each active learning state tag may indicate one or more of "keep your head in the video", "keep your eyes open for a long time", and "keep your head in the middle of the video", but it is not limited to this. On the other hand, each negative learning status label may indicate, for example, "head is not in the video", "eyes closed", "yawning", "head is not in the middle of the video", "head tilted", "frowning", "smiling" , "Black screen", "Look at the video capture device", "Mouth closed", "Squint" one or more, but not limited to this.

In a preferred embodiment, step S140 includes: filtering video segments according to active learning status tags, and generating at least one advertisement keyword based on the video tags of the selected video segments. Step S150 includes: searching for at least one event advertisement information from the advertisement pool according to the advertisement keywords, and delivering the event advertisement information to the first type of users, but not limited to this. More specifically, in step S140 of the preferred embodiment, the advertisement delivery system screens the video segments based on the active learning status tag, and generates the video segment based on the video tag of the selected video segment The keyword of the ad. Moreover, in step S150, the advertisement delivery system searches for the event advertisement information from the advertisement pool according to the advertisement keywords, and delivers the event advertisement information to the first type of user. Generally, the active learning status label indicates that the first type of user is interested in the learning content, so the event advertisement information is first pushed to the first type of user.

In a preferred embodiment, step S140 includes: filtering video segments based on negative learning status tags, and generating at least one advertisement keyword based on the video tags of the selected video segments. More specifically, in step S140 of the preferred embodiment, the advertisement delivery system screens the video segments based on the negative learning status tag, and generates the video segment based on the video tag of the filtered video segment The keyword of the ad. Furthermore, in step S150, the advertisement delivery system searches for the course advertisement information from the advertisement pool according to the advertisement keywords, and delivers the course advertisement information to the first type of user. Generally, the negative learning status label indicates that the first type of user is not interested in or does not understand the learning content, so the course advertisement information is preferentially pushed to the first type of user.

In a preferred embodiment, the following steps are included before step S110: S100, preset at least one associated second-type user for each first-type user. Step S150 includes placing advertisement information to at least one of the first type of user and the second type of user associated with the first type of user. More specifically, in the described preferred embodiment, the advertisement placement method further includes a step S100 before step S110: the advertisement placement system presets an associated second type of user for the first type of user. And, step S150 includes: the advertisement delivery system delivers the advertisement information to one of the first type of user and the second type of user. In the described preferred embodiment, the first type of user is, for example, a student, and the second type of user is, for example, the student’s parent. The described preferred embodiment sends advertisement information most suitable for the student to Parents of students, to expand the channels for parents to learn about the students' learning, and to help parents find out the most suitable advertising information for students' current status, greatly improving the conversion rate of advertising information.

In a preferred embodiment, in step S150, at least one video segment used to generate advertisement information for the first type of user and the advertisement information are used to generate advertisement push information, and sent to the second type associated with the first type of user. Class users. more specifically. In step S150 of the preferred embodiment, the advertisement delivery system generates the advertisement push information and sends the advertisement push information to the second type of user, and the advertisement push information includes the advertisement information, for example, And the video segment used to generate the advertising information. In order to let the second type of users (parents) better understand the students’ learning situation, the preferred embodiment described will be used to obtain video segments of advertising information (for example: students close their eyes, yawn, (Or expressing doubts) The video segment and advertising information are sent to the parent associated with the student, so that parents can update the student’s learning situation and the reason for pushing the advertising information to the parent, and further improve the conversion rate of advertising information.

In a preferred embodiment, the recorded video is a real-time recorded video during one-to-one online learning between the teacher and the student, and the recorded video within the preset time length is generated as a video segment every time a preset time length has passed . more specifically. In the described preferred embodiment, the recorded video is a video recorded in real time during one-to-one online learning between a teacher and a student, and the advertisement placement system in step S110 is performed every time a preset period of time has elapsed. The part of the recorded video corresponding to the preset time length is regarded as one of the video segments. Through the described preferred embodiment, in the recorded video recorded in real time, every time a video segment is recorded, a video tag is added to the video segment, which can improve the timeliness of adding video tags and the convenience of overall program promotion .

In a preferred embodiment, the selection range of the preset time length is between 10 seconds and 10 minutes, in other words, the preset time length is between 10 seconds and 10 minutes. For example, the preset time length can be implemented as 30 seconds or 1 minute, but is not limited to this.

In the online virtual classroom of each of the above embodiments, the concept of advertising is applied to the promotion of information in the virtual classroom, combining the customer’s learning field, learning level, interests and hobbies, favorite courses, occupation, age... etc. Data, in accordance with the operating environment and habits in the virtual classroom, such as the type of accustomed class device, browser type, attention and closing time of advertising messages, select the most suitable message content and placement location for customers, so as not to affect customer teaching , Provide the most useful information on the premise of learning.

2 to 9 are schematic diagrams of embodiments of the method for placing advertisements based on online learning of the present invention. Fig. 2 is a schematic diagram of segmenting a recorded movie in an embodiment of the online learning-based advertising placement method of the present invention. As shown in FIG. 2, in another preferred embodiment of the present invention, the advertisement delivery system divides a recorded video into multiple videos of 1 minute according to the preset time length (for example, 1 minute) Paragraphs A1, A2, A3, A4, A5, A6, A7, A8, A9... (hereinafter omitted).

FIG. 3 is an exemplary frame in the film section A1 in FIG. 2. As shown in FIG. 3, the advertisement delivery system extracts a frame from the video paragraph A1 for image and text recognition to obtain the first text about the content of the teaching material. In the video paragraph A1, the teacher 2 is, for example, introducing the usage of the preposition "like", and the first text obtained by the advertisement delivery system from the video paragraph A1 is, for example, as shown in the following table. Preposition 1. like is the same as... Because like can be placed before nouns as verbs and prepositions, pay attention to the distinction. What is she like? What is she like? (Preposition) 2. as: He was interested in playing chess as a child. He was interested in playing chess as a child. 3. by by/by/riding/being...: The bridge was built by prisoners. This bridge was built by prisoners.

The advertisement delivery system performs voice detection on the video segment A1 to obtain the second text. In this embodiment, the two words "preposition" and "like" that appear in both the first text and the second text with the highest number of occurrences are used as the video tags of the video paragraph A1.

In addition, the advertisement delivery system performs facial expression recognition on the facial area 31 of the student 3 who is learning the content, and obtains a positive learning status label (holding the head in the film, keeping the eyes open for a long time, keeping the head in the middle of the film, etc.), indicating The student 3 was very attentive and devoted when learning this video passage, and was able to accept the content of the video course well.

Specifically, in this embodiment, the expression recognition function of the advertisement delivery system can use existing or future expression recognition methods, and can obtain information in the video data through the expression recognition algorithm based on the facial feature information of the person. Facial features. This can speed up data processing efficiency and reduce system load. Specifically, the advertisement delivery system, for example, extracts corresponding film frames from the film segment A1 every first predetermined period for expression recognition, thereby speeding up data processing efficiency and reducing system load. The first predetermined period, for example, can be implemented for any time from 0.1 second to 10 seconds, but is not limited to this.

FIG. 4 is an exemplary frame in the movie segment A2 in FIG. 2. As shown in FIG. 4, the advertisement delivery system extracts a frame from the video paragraph A2 for image and text recognition to obtain the first text about the content of the teaching material. In the video paragraph A2, the teacher 2 is, for example, introducing the usage of the preposition "as", and the first text obtained by the advertising system from the video paragraph A2 is the same as the first text in the video paragraph A1, so it will not be repeated.

The advertisement delivery system performs voice detection on the video paragraph A2 to obtain the second text. In this embodiment, the two words "preposition" and "as" with the highest frequency appearing in both the first text and the second text are used as The movie tag of the movie paragraph A2.

The advertising system extracts a frame from the video paragraph A3 (teacher 2 is still introducing the usage of the preposition "as") for graphic recognition to obtain the first text about the content of the teaching material (and the first text from the video paragraph A1) The first text is the same, so I won’t repeat it). The advertisement delivery system performs voice detection on the video segment A3 to obtain the second text. In this embodiment, the two words "preposition" and "as" that appear in both the first text and the second text with the highest frequency are used as the video tags of the video paragraph A3.

In addition, the advertisement delivery system performs facial expression recognition on the facial area 31 of the student 3 who is learning the content, and obtains a positive learning status label (holding the head in the film, keeping the eyes open for a long time, keeping the head in the middle of the film, etc.), indicating The student 3 was very attentive and devoted when learning this video passage, and was able to accept the content of the video course well.

FIG. 5 is an exemplary frame in the movie segment A4 in FIG. 2. As shown in Figure 5, the advertising system extracts a frame from the video paragraph A4 for graphic recognition to obtain the first text about the content of the teaching material (the same as the first text obtained from the video paragraph A1, so it is not Repeat). In the video paragraph A4, teacher 2 is introducing the usage of the preposition "by", for example. The advertisement delivery system performs voice detection on the video segment A4 to obtain the second text. In this embodiment, the two words "preposition" and "by" that appear in both the first text and the second text with the highest frequency are used as the video tags of the video paragraph A4.

In addition, the advertisement delivery system performs facial expression recognition on the facial area 31 of the student 3 who is learning the content, and obtains a positive learning status label (holding the head in the film, keeping the eyes open for a long time, keeping the head in the middle of the film, etc.), indicating The student 3 was very attentive and devoted when learning this video passage, and was able to accept the content of the video course well.

FIG. 6 is an exemplary frame in the movie segment A5 in FIG. 2. As shown in Fig. 6, the advertisement delivery system extracts a frame from the video paragraph A5 for image and text recognition to obtain the first text about the content of the teaching material. In the video paragraph A5, the teacher 2 is, for example, introducing the usage of the preposition "in", and the first text obtained by the advertisement delivery system from the video paragraph A1 is, for example, as shown in the following table. Preposition 4. in (language): What's this in Chinese? How do you say this in Chinese? 5. on ride (car)/walk (step), through (radio/TV): Do you go there on foot? Do you go there on foot? 6. over: They keep in touch over the radio while working. They keep in touch over the radio while working.

The advertisement delivery system performs voice detection on the video segment A5 to obtain the second text. In this embodiment, the two words "preposition" and "in" that appear in both the first text and the second text and appear the most frequently are used as the video tags of the video paragraph A5.

The advertising system extracts a frame from the video paragraph A6 (teacher 2 is still introducing the usage of the preposition "in") for graphic recognition to obtain the first text about the content of the teaching material (and the first text obtained from the video paragraph A5) The text is the same, so I won’t repeat it). The advertisement delivery system performs voice detection on the video segment A6 to obtain the second text. In this embodiment, the two words "preposition" and "in" that appear in both the first text and the second text with the highest frequency are used as the video tags of the video paragraph A6.

In addition, the advertisement placement system performs facial expression recognition on the facial area 31 of the student 3 who is learning the content, and obtains a negative learning status label (head is not in the middle of the video, tilted head, etc.), indicating that the student 3 is learning this video segment Time is not attentive enough, difficult to invest, unable to accept the content of the video course.

FIG. 7 is an exemplary frame in the movie segment A7 in FIG. 2. As shown in FIG. 7, the advertisement delivery system extracts a frame from the video paragraph A7 for image and text recognition to obtain the first text about the content of the teaching material. In the video paragraph A7, the teacher 2 is, for example, introducing the usage of the preposition "on", and the first text obtained by the advertising system from the video paragraph A7 is the same as the first text in the video paragraph A5, so it will not be repeated.

The advertisement delivery system performs voice detection on the video segment A7 to obtain the second text. In this embodiment, the two words "preposition" and "on" that appear in both the first text and the second text with the highest frequency are used as the video tags of the video paragraph A7.

The advertisement placement system extracts a frame from the video paragraph A8 (teacher 2 is still introducing the usage of the preposition "on") for graphic recognition to obtain the first text about the content of the teaching material (and from the video paragraph A5) The first text is the same, so I won’t repeat it). The advertisement delivery system performs voice detection on the video segment A8 to obtain the second text. In this embodiment, the two words "preposition" and "on" that appear in both the first text and the second text with the highest frequency are used as the video tags of the video paragraph A8.

In addition, the advertisement delivery system performs facial expression recognition on the facial area 31 of the student 3 who is learning the content, and obtains the active learning status label (the head is kept in the film, the eyes are open for a long time, the head is kept in the middle of the film, etc.), It means that the student 3 is very attentive and engaged when learning this video passage, and can accept the content of the video course well.

FIG. 8 is an exemplary frame in the movie segment A9 in FIG. 2. As shown in FIG. 8, the advertisement delivery system extracts a frame from the video paragraph A9 for image and text recognition to obtain the first text about the content of the teaching material. In the video paragraph A9, the teacher 2 is, for example, introducing the usage of the preposition "over", and the first text obtained by the advertising system from the video paragraph A9 is the same as the first text in the video paragraph A5, so it will not be repeated.

The advertisement delivery system performs voice detection on the video paragraph A9 to obtain the second text. In this embodiment, the two words "preposition" and "over" that appear in both the first text and the second text and appear the most frequently As the movie tag of the movie paragraph A9.

In addition, the advertisement delivery system performs facial expression recognition on the facial area 31 of the student 3 who is learning the content, and obtains a positive learning status label (holding the head in the film, keeping the eyes open for a long time, keeping the head in the middle of the film, etc.), indicating The student 3 was very attentive and devoted when learning this video passage, and was able to accept the content of the video course well.

Then, in this example, the video paragraphs A1~A9 are filtered according to the negative learning status tags to obtain video paragraphs A5 and A6, and the video tags "preposition" and "in" of the video paragraphs A5 and A6 are used as advertising keywords (" "Preposition" and "in"). FIG. 9 is a schematic diagram of the advertisement delivery system searching for advertisement information according to the learning status tag and advertisement keywords and delivering it to users. As shown in FIG. 9, the server 10 prestores multiple pieces of educational advertisement information 11, 12, 13, 14, 15, 16, 17 ... (hereinafter omitted) with advertisement keywords. In this embodiment, the server 10 searches for corresponding teaching advertisement information based on the two advertising keywords of "preposition" and "in", so as to obtain one of the teaching advertisement information related to "preposition" and "in" The grammar textbook advertisement 11. In addition, the advertisement delivery system generates advertisement push information 19 from the video paragraph A5 or A6 and the grammar textbook advertisement 11 and sends them to a mobile device 30 held by the student 3 and a second type associated with the student 3, respectively. The mobile device 40 of the user 4 (for example, the father of the student 3). Through the advertisement push information 19, the second type of user 4 can better know the learning situation of the student 3. This embodiment will be used to obtain the video segment of the advertisement information (for example: the student closes his eyes, yawns, (Or expressing doubts) The video segment and advertising information are sent to the parent associated with the student, so that the parent can understand the student’s learning situation and the reason for pushing the advertising information to the parent, and further improve the conversion rate of the advertising information.

After using the online learning-based advertising method of this embodiment, even in the same online learning video class, the advertising information is accurately selected according to the different needs of each student, which can better match the different learning progress and knowledge background. Different students can accurately apply different types of users, increase the accuracy and conversion rate of advertising information, and improve the humanized experience.

Fig. 10 is a schematic diagram of a first embodiment of an advertisement placement system based on online learning of the present invention. The online learning-based advertisement placement system 5 in FIG. 10 is used to implement the online learning-based advertisement placement method of each of the above embodiments. In this embodiment, the advertisement delivery system 5 includes, for example, a video paragraph generation module 51, a video tag addition module 52, a status tag addition module 53, an advertisement keyword generation module 54, and an advertisement information. Placement module 55.

The video segment generating module 51 is used for dividing at least one recorded video of online learning of the first type of users into multiple video segments according to a preset time length. In other words, the video segment generation module 51 is used to divide a recorded video showing a first-type user for online learning into a plurality of video segments according to a preset time length.

The video tag adding module 52 is used for adding at least one video tag to the video segment according to the content information of the video segment. In other words, the video tag adding module 52 is used to add a video tag to at least one of the video segments according to the content information of the video segment.

The status label adding module 53 is used to perform facial expression recognition and add at least one learning status label for the first-type user in each movie segment. In other words, the status tag adding module 53 is used to recognize the first type of user expression in the segment of the movie, and when it is determined that the first type of user’s expression belongs to the same expression type, the accumulated time length accounts for When the proportion of the total time length of the video segment exceeds a preset threshold, a learning status tag corresponding to the expression type is added to the video segment.

The advertisement keyword generating module 54 is used to filter video segments according to preset learning status tags, and generate at least one advertisement keyword according to the video tags of the selected video segments. In other words, the advertising keyword generation module 54 is used to filter the video segments according to one or more preset learning status tags among the learning status tags, and according to each video that is filtered out The video tag of the paragraph generates an advertisement keyword.

The advertisement information delivery module 55 is used to search for at least one piece of advertisement information according to advertisement keywords from the advertisement pool, and deliver advertisement information to the first type of users. In other words, the advertisement information placement module 55 is used to search out an advertisement information from the advertisement pool according to each advertisement keyword, and deliver the advertisement information to the first type of user.

The present invention also provides an embodiment of an advertisement placement device based on online learning. The advertisement placement device includes a processor and a storage electrically connected to the processor. The memory stores executable instructions that can be executed by the processor. In addition, the processor is configured to implement the steps of the aforementioned online learning-based advertisement placement method by executing the executable instructions.

Each of the foregoing embodiments can confirm the learning content and learning status of each user's online learning recording video, and generate targeted and flexible advertising information, so that it can be accurately applied to various types of users. Increase the accuracy and conversion rate of advertising information, and improve the user-friendly experience.

Those with ordinary knowledge in the technical field to which the present invention belongs should understand that various aspects of the present invention can be implemented as methods, systems, equipment, or computer program products. Therefore, various aspects of the present invention can be implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, micro-commands, etc.), or a combination of hardware and software, here Can be collectively referred to as "circuit", "module" or "platform".

Fig. 11 is a schematic structural diagram of an embodiment of an advertisement placement device based on online learning of the present invention. The advertisement placement device 600 according to this embodiment of the present invention will be described below with reference to FIG. 11. It is supplemented that the advertisement placement device 600 shown in FIG. 11 is only one embodiment of the present invention, and is not used to limit the implementation, function, and protection scope of the present invention.

As shown in FIG. 11, the advertisement placement device 600 is, for example, in the form of a computer device. The components of the advertisement delivery device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus bar 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, and the like.

The storage unit 620 stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 executes the steps of the advertisement placement method of each of the foregoing embodiments. For example, the processing unit 610 may perform the steps shown in FIG. 1.

The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory (RAM) 6201 and/or a high-speed temporary memory 6202, and the storage unit 620 may further include a unique memory (ROM) ) 6203.

The storage unit 620 may also include a program/utility tool 6204 having one or more program modules 6205. Such program modules 6205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. , Each of these examples or some combination may include the implementation of a network environment.

The bus 630 can represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or any bus structure using a variety of bus structures Local bus.

The advertisement placement device 600 may also communicate with one or more external devices 700 (such as keyboards, pointing devices, Bluetooth devices, etc.), and may also communicate with one or more devices that enable a user to interact with the advertisement placement device 600, and/ Or communicate with any device (such as a router, modem, etc.) that enables the advertisement placement device 600 to communicate with one or more other computing devices. This communication can be done through the I/O port 650. In addition, the advertisement placement device 600 may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 660. The network adapter 660 can communicate with other modules of the advertisement placement device 600 through the bus 630. It should be understood that although not shown in the figure, other hardware and/or software modules can be used in conjunction with the advertisement placement device 600, including but not limited to: micro-commands, device drivers, redundant processing units, external disk drive arrays, RAID systems , Tape drives and data backup storage platforms.

The present invention also provides an embodiment of a computer-readable recording medium. The computer-readable recording medium stores an application program. When the application program is loaded and executed by an electronic device, the electronic device can implement the aforementioned The steps of the online learning-based advertisement placement method of each embodiment. In some possible implementation manners, various aspects of the present invention can also be implemented in the form of a computer program product, and the computer program product includes a program code, and when the program code of the computer program product runs on an electronic device, The program code can enable the electronic device to implement the steps of the online learning-based advertisement placement method of each of the foregoing embodiments.

FIG. 12 is a schematic structural diagram of an embodiment of a computer-readable recording medium of the present invention. Referring to FIG. 12, the computer-readable recording medium of this embodiment is implemented as a computer program product 800 that enables an electronic device to implement the aforementioned advertising placement method, for example, it may be a portable compact disc (CD-ROM). And include code, and its code can be run on an electronic device (such as a personal computer). However, the computer-readable recording medium of the present invention is not limited to this. In this specification, the computer-readable recording medium can be any tangible medium that contains or stores a program that can be used by an instruction execution system, device, or device Or use it in combination.

The computer program product can use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or a combination of any of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, pen drives, hard drives, random access memory (RAM), read-only memory (ROM) , Programmable read-only memory (EPROM or flash memory), optical fiber, optical disc (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.

The computer-readable recording medium may include a data signal propagated in baseband or as a part of a carrier wave, which carries readable program codes. This propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The computer-readable recording medium may also be any readable medium other than a readable storage medium, and the readable medium can send, propagate, or transmit a program for use by or in combination with the instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

The programming code for performing the operations of the present invention can be written in any combination of one or more programming languages. Programming languages include object-oriented programming languages—such as Java, C++, etc., as well as conventional procedural programming languages— Such as "C" language or similar programming languages. The code can be executed entirely on the user's computer device, partly on the user's computer device, executed as an independent software package, partly on the user's computer device and partly executed on the remote computer device, or entirely on the remote computer device Execute on the device or server. In the case of remote computer equipment, the remote computer equipment can be connected to the user’s computer equipment through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer equipment ( For example, use an Internet service provider to connect through the Internet).

In summary, this patent specification provides online learning-based advertising methods, systems, equipment, and computer-readable recording media, which can confirm the learning content and learning status of each user’s online learning recording video. Targeted and flexible advertising information can be accurately applied to different types of users to increase the accuracy and conversion rate of advertising information, and improve the humanized experience, so it can indeed achieve the purpose of the invention.

However, the above are only examples of the present invention. When the scope of implementation of the present invention cannot be limited by this, all simple equivalent changes and modifications made in accordance with the scope of the patent application of the present invention and the content of the patent specification still belong to This invention patent covers the scope.

S110~S150: steps A1~A9: Video paragraph 2: Teacher 3: Trainees 31: Face area 11~17: Teaching advertising information 4: The second type of user 10: Server 19: Advertising push information 30: mobile device 40: mobile device 5: Advertising system 51: Video paragraph generation module 52: video tag adding module 53: Status tag add module 54: Advertising keyword generation module 55: Advertising information delivery module 600: Advertising equipment 610: Processing Unit 620: storage unit 6201: random access memory 6202: High-speed temporary memory 6203: Only memory 6204: Programs/Utilities 6205: program module 630: Bus 640: display unit 650: I/O port 660: network adapter 700: External device 800: computer program products

Other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, in which: FIG. 1 is a flowchart of an embodiment of an advertisement placement method based on online learning of the present invention; FIG. 2 is a schematic diagram exemplarily showing that a recorded movie is divided into multiple movie segments in the advertisement placement method of this embodiment; Fig. 3 is a schematic diagram exemplarily showing a frame in one of the film segments; Fig. 4 is a schematic diagram exemplarily showing a frame in one of the film segments; FIG. 5 is a schematic diagram exemplarily showing a frame in one of the film segments; FIG. 6 is a schematic diagram exemplarily showing a frame in one of the film segments; FIG. 7 is a schematic diagram exemplarily showing a frame in one of the film segments; FIG. 8 is a schematic diagram exemplarily showing a frame in one of the film segments; Figure 9 is a schematic diagram illustrating how this embodiment searches for advertising information based on learning status tags and advertising keywords and delivers it to users; Fig. 10 is a schematic diagram exemplarily showing an embodiment of the online learning-based advertisement placement system of the present invention; Figure 11 is a schematic diagram exemplarily showing an embodiment of the online learning-based advertisement placement device of the present invention; and FIG. 12 is a schematic diagram exemplarily showing an embodiment of the computer-readable recording medium of the present invention.

S110~S150: steps

Claims (14)

An advertising delivery method based on online learning, including: Step S110: Divide a recorded video showing a first type of user's online learning into multiple video segments according to a preset time length; Step S120: Add a video tag to the video segment according to the content information of at least one of the video segments; Step S130: Perform facial expression recognition on the first-type user in the video segment, and when it is determined that the first-type user’s facial expressions belong to the same type of facial expression, the ratio of the cumulative time length to the total time length of the video segment exceeds a preset When setting the threshold, add a pair of learning status tags corresponding to the expression type to the video segment; Step S140: Filter the video segments according to one or more preset learning status tags among the learning status tags, and generate an advertisement keyword according to the video tag of each video segment that is filtered out; and Step S150: Search for an advertisement information according to each advertisement keyword from an advertisement pool, and deliver the advertisement information to the first type of user. The online learning-based advertisement placement method according to claim 1, wherein step S120 includes: Step S121: Perform graphic recognition on at least one frame of the video paragraph to obtain a first text corresponding to the film paragraph, and perform voice recognition on the film paragraph to obtain a second text corresponding to the film paragraph; Step S122: Obtain a word as the movie tag of the movie paragraph according to the first text and the second text corresponding to the movie paragraph. The online learning-based advertisement placement method according to claim 2, wherein, in step S122, the word with the highest total number of occurrences in the first text and the second text is used as the video tag of the video paragraph. The online learning-based advertisement placement method according to claim 2, wherein, in step S122, the word with the highest number of occurrences among the words appearing in both the first text and the second text is used as the video paragraph The video tag. The online learning-based advertisement placement method according to claim 1, wherein, in step S130, the learning state label is a positive learning state label corresponding to a positive learning expression type or a negative learning corresponding to a negative learning expression type Status label. The advertisement delivery method based on online learning as described in claim 5, wherein: Step S140 includes: screening the video segments according to the active learning status tag, and generating the advertising keyword according to the video tag of the selected video segment; and Step S150 includes: searching out an event advertisement information from the advertisement pool according to the advertisement keyword, and delivering the event advertisement information to the first type of user. The advertisement delivery method based on online learning as described in claim 5, wherein: Step S140 includes: filtering the video segments according to the negative learning status tag, and generating the advertising keyword according to the video tag of the video segment that is filtered out; and Step S150 includes: searching for a course advertisement information from the advertisement pool according to the advertisement keyword, and delivering the course advertisement information to the first type of user. As described in claim 1, the online learning-based advertisement placement method further includes step S100 before step S110: preset an associated second type of user for the first type of user; and step S150 includes: One of the first type of user and the second type of user places the advertisement information. The online learning-based advertisement delivery method according to request item 8, wherein step S150 further includes: generating an advertisement push information, and sending the advertisement push information to the second type of user, and the advertisement push information includes the advertisement Information, and the video segment used to generate the advertising information. The online learning-based advertisement placement method according to claim 1, wherein the recorded video is a video recorded in real time when a teacher and a student conduct one-to-one online learning, and step S110 is every time a preset period of time has elapsed The part of the recorded video corresponding to the preset time length is regarded as one of the video segments. According to claim 10, the online learning-based advertisement placement method, wherein the preset time length is between 10 seconds and 10 minutes. An advertisement placement system based on online learning, which is used to implement the online learning-based advertisement placement method according to any one of request items 1 to 11, and includes: A video segment generation module for dividing a recorded video showing a first-type user for online learning into multiple video segments according to a preset time length; A video tag adding module for adding a video tag to the video segment based on the content information of at least one of the video segments; A status tag adding module for recognizing the expression of the first type of user in the segment of the video, and when it is determined that the expression of the first type of user belongs to the same type of expression, the cumulative length of time accounts for the total length of the segment of the video When the ratio of the time length exceeds a preset threshold, add a pair of learning status tags corresponding to the expression type to the video segment; An advertisement keyword generation module for filtering the video segments according to one or more preset learning status tags in the learning status tags, and generating according to the video tag of each video segment that is filtered out An advertising keyword; and An advertisement information delivery module is used to search out an advertisement information from an advertisement pool according to each advertisement keyword, and deliver the advertisement information to the first type of user. An advertisement placement device based on online learning, including: A processor; and A memory storing executable instructions that can be executed by the processor; Wherein, the processor is configured to implement the online learning-based advertisement placement method according to any one of claim items 1 to 11 by executing the executable instructions. A computer-readable recording medium stores an application program. When the application program is loaded and executed by an electronic device, the electronic device can implement the online learning-based as described in any one of claims 1 to 11 Ad delivery method.

Images