CN114650455A - Barrage information processing method and device, electronic equipment and storage medium - Google Patents
Barrage information processing method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114650455A CN114650455A CN202210147409.0A CN202210147409A CN114650455A CN 114650455 A CN114650455 A CN 114650455A CN 202210147409 A CN202210147409 A CN 202210147409A CN 114650455 A CN114650455 A CN 114650455A
- Authority
- CN
- China
- Prior art keywords
- bullet screen
- target
- screen information
- information
- category
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 13
- 238000003672 processing method Methods 0.000 title claims abstract description 12
- 238000013145 classification model Methods 0.000 claims abstract description 73
- 238000012545 processing Methods 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000000034 method Methods 0.000 claims description 32
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 230000006399 behavior Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 description 7
- 238000007726 management method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application
- H04N21/4788—Supplemental services, e.g. displaying phone caller identification, shopping application communicating with other users, e.g. chatting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
- H04N21/4318—Generation of visual interfaces for content selection or interaction; Content or additional data rendering by altering the content in the rendering process, e.g. blanking, blurring or masking an image region
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure relates to a bullet screen information processing method and device, electronic equipment and a storage medium, and relates to the technical field of data processing. The specific scheme comprises the following steps: acquiring target bullet screen information of a target live broadcast room; classifying the target bullet screen information by adopting a target bullet screen classification model corresponding to the target live broadcast room, and determining a target class corresponding to the target bullet screen information, wherein the target bullet screen classification model is obtained by adding bullet screen information training of the target live broadcast room on the basis of a general bullet screen classification model; and when the target category is a first category, deleting the target barrage information from the target live broadcast room, wherein the target barrage information of the first category is barrage information meeting the negative feedback behavior triggering condition of the user. Through the bullet screen classification model to different live broadcast rooms, realized the automatic shielding of bullet screen information of different live broadcast rooms, improved the flexibility that the bullet screen was shielded.
Description
Technical Field
The present disclosure relates to the field of data processing technologies, and in particular, to a method and an apparatus for processing bullet screen information, an electronic device, and a storage medium.
Background
The bullet screen is comment information popped up along with the video content when the video is played, and the bullet screen not only enables a user to express the experience of watching the program, but also enables the user to watch comment content of other users on the program, so that interactivity of the user when watching the video is realized.
Today, many video-like applications provide barrage functionality. In order to prevent some inappropriate barrage information from affecting other users and anchor players watching the video, the electronic device needs to identify and shield the inappropriate barrage information.
In the related art, a mask keyword may be set in advance. When the bullet screen information contains the shielding keyword, the electronic equipment can shield the bullet screen information. However, this method is less flexible and cannot be masked by the electronic device when the user changes Chinese to pinyin or harmonious characters.
Disclosure of Invention
The invention provides a bullet screen information processing method and device, electronic equipment and a storage medium.
The technical scheme of the disclosure is as follows:
according to a first aspect of the present disclosure, there is provided a method for processing bullet screen information, the method including:
acquiring target bullet screen information of a target live broadcast room;
classifying the target bullet screen information by adopting a target bullet screen classification model corresponding to a target live broadcast room, and determining a target category corresponding to the target bullet screen information, wherein the target bullet screen classification model is obtained by adding bullet screen information training meeting preset conditions in the target live broadcast room on the basis of a general bullet screen classification model;
and when the target category is a first category, deleting the target barrage information from the target live broadcast room, wherein the target barrage information of the first category is barrage information meeting the negative feedback behavior triggering condition of the user.
Optionally, the processing method of the bullet screen information further includes:
and when the target bullet screen information is determined to be the bullet screen information deleted by mistake, adding the target bullet screen information into the target bullet screen sample set as a forward sample.
Optionally, the target category is a second category, the target barrage information of the second category is normal barrage information, and the processing method of the barrage information further includes:
and responding to the operation that the account marks the target bullet screen information into the first category, and adding the target bullet screen information into the target bullet screen sample set as a negative sample.
Optionally, the processing method of the bullet screen information further includes:
and when the ratio of the number of the newly added samples in the target bullet screen sample set to the total number of the samples before addition is greater than or equal to a preset percentage, retraining the neural network by adopting the target bullet screen sample set after the samples are added to obtain a new target bullet screen classification model.
Optionally, the processing method of the bullet screen information further includes:
acquiring barrage samples of a first category from a plurality of live broadcast rooms, and acquiring barrage samples of a second category from the plurality of live broadcast rooms to obtain a barrage sample set;
training the neural network by adopting a bullet screen sample set to obtain a general bullet screen classification model;
and when the ratio of the number of the newly added samples in the bullet screen sample set to the total number of the samples before the addition is greater than or equal to a preset percentage, retraining the neural network by adopting the bullet screen sample set after the samples are added to obtain a target bullet screen classification model.
According to a second aspect of the present disclosure, there is provided a processing apparatus of bullet screen information, comprising:
the acquisition module is configured to acquire target barrage information of a target live broadcast room;
the determining module is configured to perform classification on the target bullet screen information acquired by the acquiring module by adopting a target bullet screen classification model corresponding to a target live broadcast room, and determine a target category corresponding to the target bullet screen information, wherein the target bullet screen classification model is obtained by adding bullet screen information which meets preset conditions in the target live broadcast room on the basis of a general bullet screen classification model;
and the deleting module is configured to delete the target barrage information from the target live broadcast room when the target category determined by the determining module is the first category, wherein the target barrage information of the first category is barrage information meeting the negative feedback behavior triggering condition of the user.
Optionally, the apparatus for processing barrage information further includes: an add module configured to perform:
and when the target bullet screen information is determined to be the bullet screen information deleted by mistake, adding the target bullet screen information into the target bullet screen sample set as a forward sample.
Optionally, the target category is a second category, the target barrage information of the second category is normal barrage information, and the processing device for the barrage information further includes: an add module configured to perform:
and responding to the operation that the target bullet screen information is marked as a class by the account, and adding the target bullet screen information into the target bullet screen sample set as a negative sample.
Optionally, the apparatus for processing bullet screen information further includes: a training module;
and the training module is configured to execute that when the ratio of the number of the newly added samples in the target bullet screen sample set to the total number of the samples before addition is greater than or equal to a preset percentage, the neural network is retrained by using the target bullet screen sample set after the samples are added, so that a new target bullet screen classification model is obtained.
Optionally, the apparatus for processing bullet screen information further includes: a training module;
the acquisition module is further configured to acquire barrage samples of a first category from the plurality of live broadcast rooms, acquire barrage samples of a second category from the plurality of live broadcast rooms, and obtain a barrage sample set;
the training module is configured to train the neural network by adopting the bullet screen sample set acquired by the acquisition module to obtain a general bullet screen classification model; and when the ratio of the number of the newly added samples in the bullet screen sample set to the total number of the samples before the addition is greater than or equal to a preset percentage, retraining the neural network by adopting the bullet screen sample set after the samples are added to obtain a target bullet screen classification model.
According to a third aspect of the present disclosure, there is provided an electronic apparatus comprising:
a processor;
a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement any one of the above-mentioned methods for processing optionally bullet screen information in the first aspect.
According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium having instructions stored thereon, which, when executed by a processor of an electronic device, enable the electronic device to perform any one of the above-mentioned methods for processing optionally barrage information of the first aspect.
According to a fifth aspect of the present disclosure, there is provided a computer program product containing instructions which, when run on an electronic device, cause the electronic device to perform the method of processing optionally barrage information as in any of the first aspects.
The technical scheme provided by the disclosure at least brings the following beneficial effects: the method comprises the steps that target bullet screen information of a target live broadcast room is obtained by a bullet screen information processing device, a target bullet screen classification model corresponding to the target live broadcast room is adopted to classify the target bullet screen information, a target category corresponding to the target bullet screen information is determined, when the target category is a first category, the target bullet screen information is deleted from the target live broadcast room, and the target bullet screen information of the first category is bullet screen information meeting negative feedback behavior trigger conditions of a user. Because the bullet screen information training that target bullet screen classification model satisfies the preset condition in increasing the target live broadcast room on general bullet screen classification model basis obtains, the bullet screen information of different live broadcast rooms is different for the target bullet screen classification model that different live broadcast rooms correspond is different. The bullet screen information is classified by utilizing the bullet screen classification model aiming at each live broadcast room, if the bullet screen information is in a first category, the bullet screen information is deleted, intelligent shielding of the bullet screen information of each live broadcast room is realized, and flexibility of bullet screen shielding is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
Fig. 1 is one of flowcharts illustrating a method of processing bullet screen information according to an exemplary embodiment.
Fig. 2 is a second flowchart illustrating a method for processing bullet screen information according to an exemplary embodiment.
Fig. 3 is one of the logical structure block diagrams of a bullet screen information processing apparatus according to an exemplary embodiment.
Fig. 4 is a second logic structure block diagram of a bullet screen information processing device according to an exemplary embodiment.
Fig. 5 is a block diagram illustrating a structure of an electronic device according to an example embodiment.
Detailed Description
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
It should be noted that the embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
It should be noted that the user information (including but not limited to user device information, user personal information, etc.) related to the embodiments of the present disclosure is information authorized by the user or sufficiently authorized by each party.
At present, the barrage becomes an important element for watching videos and live broadcasts for a plurality of audiences, and the barrage not only enables users to express the experience of watching programs, but also enables the users to watch the comment contents of other users to the programs, thereby realizing the interactivity of the users when watching the videos.
The method for processing the bullet screen information provided by the embodiment of the disclosure can be applied to a scene of shielding the bullet screen information by the electronic equipment. According to the related technology, the bullet screen information containing the shielding keywords is shielded by setting the shielding keywords. Or shielding the bullet screen information by using a general bullet screen classification model. However, the flexibility of the way of setting the shielding keywords is low, and the general barrage classification model is not suitable for all live broadcast rooms.
In order to realize automatic shielding of bullet screen information in different live broadcast rooms and improve flexibility of bullet screen shielding, the embodiment of the disclosure provides a bullet screen information processing method. The bullet screen information is classified by utilizing the bullet screen classification model aiming at each live broadcast room, if the bullet screen information is in a first category, the bullet screen information is deleted, intelligent shielding of the bullet screen information of each live broadcast room is realized, and flexibility of bullet screen shielding is improved.
The execution main body of the method for processing bullet screen information provided by the embodiment of the present disclosure is a processing device of bullet screen information. The processing device of the bullet screen information can be electronic equipment, and the electronic equipment can be electronic equipment used by a host user or electronic equipment used by a house management user. Specifically, the anchor user or the house management can log in the live webcast platform to perform live webcast by using the electronic device. Audience users can log in a live webcast platform to watch live webcasts by using own electronic equipment, and can release barrage information to a live webcast room. The bullet screen information may be normal bullet screen information or illegal bullet screen information.
Fig. 1 is a flowchart illustrating a method for processing bullet screen information according to an exemplary embodiment, and as shown in fig. 1, the method may include the following steps 101 to 103.
101. The electronic equipment acquires target bullet screen information of a target live broadcast room.
When a main broadcasting user or a house management uses the electronic equipment to log in a network live broadcasting platform for network live broadcasting, the electronic equipment can acquire target barrage information published in a target live broadcasting room in the live broadcasting process.
Generally, the barrage is displayed in a text form. In some cases, the barrage may also be presented in the form of barrage rich media (such as simple emoticon pictures, character icons, flash animation effects).
102. The electronic equipment classifies the target bullet screen information by adopting a target bullet screen classification model corresponding to the target live broadcast room, and determines a target category corresponding to the target bullet screen information.
After the electronic equipment acquires the target bullet screen information, a target bullet screen classification model corresponding to the target live broadcast room can be acquired, the target bullet screen information is classified by the aid of the target bullet screen classification model, and a target category corresponding to the target bullet screen information is determined. The target bullet screen classification model is obtained by adding bullet screen information training meeting preset conditions in a target live broadcast room on the basis of the general bullet screen classification model. The bullet screen information in different live broadcast rooms is different, so that the target bullet screen classification models in different live broadcast rooms are different.
It is to be understood that the target category may be a preset category. For example, the target category may be a first category or a second category. The first category refers to the category of the offending barrage, and the second category refers to the category of the normal barrage.
In addition, the general barrage classification model is obtained by training barrage information from a plurality of live broadcast rooms and is suitable for barrage classification models of all live broadcast rooms. The bullet screen information meeting the preset condition in the target live broadcast room can be normal bullet screen information deleted by mistake, and can also be information marked as an illegal bullet screen by an anchor user or a house management.
Optionally, in the embodiment of the present disclosure, after the electronic device classifies the target bullet screen information by using the target bullet screen classification model, the probability that the target bullet screen information belongs to different categories is obtained. The electronic equipment can belong to different categories according to the probability of the target bullet screen information
And determining the target type corresponding to the target bullet screen information according to the probability of different types.
For example, after the electronic device classifies the target bullet screen information by using the target bullet screen classification model, a first probability that the target bullet screen information belongs to a first class and a second probability that the target bullet screen information belongs to a second class are obtained. Then, the electronic device may determine whether the target category corresponding to the target bullet screen information is the first category or the second category according to the first probability, the second probability, and the preset probability value. For example, the electronic device may determine that the target category is the first category when it is determined that the first probability is greater than a first preset probability value and the second probability is less than a second preset probability value.
103. And when the target category is the first category, the electronic equipment deletes the target bullet screen information from the target live broadcast room.
The target bullet screen information of the first category is bullet screen information meeting negative feedback behavior triggering conditions of the user, namely the target bullet screen information of the first category is illegal bullet screen information.
Optionally, in this embodiment of the present disclosure, when the target category is the first category, the electronic device may prohibit a user who publishes the target bullet screen information from speaking in the target live broadcast room, in addition to deleting the target bullet screen information from the target live broadcast room.
Optionally, when the target category is a second category and the normal barrage category of the second category, the electronic device does not perform the operation.
The technical scheme provided by the embodiment at least has the following beneficial effects: the processing device of the bullet screen information acquires target bullet screen information of a target live broadcast room, classifies the target bullet screen information by adopting a target bullet screen classification model corresponding to the target live broadcast room, determines a target category corresponding to the target bullet screen information, deletes the target bullet screen information from the target live broadcast room when the target category is a first category, and the target bullet screen information of the first category is bullet screen information meeting negative feedback behavior trigger conditions of a user. Because the target barrage classification model is obtained by adding barrage information training meeting preset conditions in the target live broadcast room on the basis of the general barrage classification model, the barrage information of different live broadcast rooms is different, and the target barrage classification models corresponding to the different live broadcast rooms are different. The bullet screen information is classified by utilizing the bullet screen classification model aiming at each live broadcast room, if the bullet screen information is in a first category, the bullet screen information is deleted, intelligent shielding of the bullet screen information of each live broadcast room is realized, and flexibility of bullet screen shielding is improved.
Optionally, in this embodiment of the present disclosure, after the electronic device deletes the target bullet screen information from the target live broadcast room and prohibits the user who publishes the target bullet screen information from speaking, if the feedback information from the user is received and the feedback information is used to indicate that the normal bullet screen is incorrectly shielded, the electronic device may determine that the target bullet screen information is the mistakenly deleted bullet screen information in response to a confirmation operation of the anchor user or the room management user on the feedback information, and add the target bullet screen information as a forward sample into the target bullet screen sample set.
The bullet screen information deleted by mistake shows that the accuracy of the classification result of the target bullet screen classification model needs to be improved, and the bullet screen information deleted by mistake is added into the target bullet screen sample set as a forward sample, so that the target bullet screen sample set retrains the bullet screen classification model after meeting certain conditions, real-time updating of the bullet screen classification model is achieved, and the effectiveness of the bullet screen classification model is guaranteed.
Optionally, in this embodiment of the present disclosure, when the target category is the second category and the target barrage information of the second category is the normal barrage information, the electronic device may respond to an operation of marking the target barrage information as the first category by the anchor account or the house management account, and add the target barrage information as a negative sample to the target barrage sample set.
When the target bullet screen information is an illegal bullet screen and is not recognized by the model, the information can be added into the target bullet screen sample set as a negative sample in an artificial marking mode, so that the bullet screen classification model is retrained after the target bullet screen sample set meets certain conditions, real-time updating of the bullet screen classification model is achieved, and effectiveness of the bullet screen classification model is guaranteed.
It can be understood that the operation of marking the target bullet screen information as the first category may specifically be an operation of pulling an account corresponding to the target bullet screen information into a blacklist, or may be an operation of deleting the target bullet screen information.
Optionally, in this embodiment of the present disclosure, when the ratio of the number of the samples newly added to the target barrage sample set to the total number of the samples before addition is greater than or equal to the preset percentage, the electronic device may retrain the neural network by using the target barrage sample set after adding the samples, to obtain a new target barrage classification model. Therefore, the electronic equipment can classify the bullet screen information by adopting the new target bullet screen classification model so as to determine whether to shield the bullet screen information.
The target bullet screen classification model is retrained under the condition that the target bullet screen sample set meets a certain condition, so that continuous iteration of the target bullet screen classification model is realized, and the real-time performance and the accuracy of the target bullet screen classification model are guaranteed.
Optionally, in this embodiment of the present disclosure, with reference to fig. 1, as shown in fig. 2, the method for processing bullet-screen information may further include the following steps 104 to 106.
104. The electronic equipment acquires the barrage sample of the first category from the plurality of live rooms, acquires the barrage sample of the second category from the plurality of live rooms, and obtains a barrage sample set.
The total number of the samples in the bullet screen sample set meets the preset number, and the number of the bullet screen samples in the two categories in the bullet screen sample set is the same.
105. The electronic equipment trains the neural network by adopting the bullet screen sample set to obtain a general bullet screen classification model.
Optionally, in this embodiment of the present disclosure, the electronic device may perform preprocessing on each bullet screen sample in the bullet screen sample set, and specifically may filter special characters or messy codes in the bullet screen text. Then, the electronic device may predict the core character based on the context character, obtain a word vector corresponding to each bullet screen sample, and represent the bullet screen samples into a sentence matrix according to the word vector. Then, the electronic device can perform feature extraction on the sentence matrix through the one-dimensional convolutional layer, change sentences with different lengths into fixed-length representations through the pooling layer, and output the probability that the bullet screen sample belongs to each category through the softmax function of the full connection layer.
It should be noted that, in the embodiment of the present disclosure, the general barrage classification model may be obtained by training the electronic device itself, or may be obtained by training the server and sent to the electronic device.
106. And when the ratio of the number of the newly added samples in the bullet screen sample set to the total number of the samples before the addition is greater than or equal to the preset percentage, the electronic equipment retrains the neural network by adopting the bullet screen sample set after the samples are added, and obtains a target bullet screen classification model.
The newly added sample refers to the barrage information meeting the preset condition in the target live broadcast room, that is, refers to the mistakenly deleted positive barrage information or the negative barrage information which is not identified by the model.
It should be noted that, steps 104 to 106 are processes for obtaining a target bullet screen classification model for the first time based on the general bullet screen classification model. And then, in the live broadcasting process, continuous iteration can be carried out based on the target barrage classification model.
The technical scheme provided by the embodiment at least has the following beneficial effects: through training general barrage classification model earlier, then constantly add the distinctive barrage information that satisfies the preset condition in the live broadcast room, carry out the training of model, realized the shielding to the barrage information in different live broadcast rooms.
Fig. 3 is a block diagram illustrating a logical structure of a bullet screen information processing apparatus according to an exemplary embodiment. Referring to fig. 3, the apparatus for processing bullet screen information is applied to an electronic device, and includes: an acquisition module 31, a determination module 32 and a deletion module 33.
An obtaining module 31 configured to perform obtaining target barrage information of a target live broadcast room;
the determining module 32 is configured to perform classification on the target bullet screen information acquired by the acquiring module 31 by using a target bullet screen classification model corresponding to the target live broadcast room, and determine a target category corresponding to the target bullet screen information, wherein the target bullet screen classification model is obtained by adding bullet screen information meeting preset conditions in the target live broadcast room on the basis of a general bullet screen classification model;
and the deleting module 33 is configured to delete the target barrage information from the target live broadcast room when the target category determined by the determining module 32 is the first category, where the target barrage information of the first category is barrage information meeting a negative feedback behavior triggering condition of the user.
Optionally, as shown in fig. 4, the apparatus for processing bullet screen information further includes: a module 34 is added.
An adding module 34 configured to perform: and when the target bullet screen information is determined to be the bullet screen information deleted by mistake, adding the target bullet screen information into the target bullet screen sample set as a forward sample.
Optionally, the target category is a second category, and the target bullet screen information of the second category is normal bullet screen information.
An adding module 34 configured to perform: and responding to the operation that the account marks the target bullet screen information into the first category, and adding the target bullet screen information into the target bullet screen sample set as a negative sample.
Optionally, as shown in fig. 4, the apparatus for processing bullet screen information further includes: a training module 35.
And the training module 35 is configured to perform retraining of the neural network by using the target bullet screen sample set after the sample is added when the ratio of the number of the newly added samples in the target bullet screen sample set to the total number of the samples before the addition is greater than or equal to a preset percentage, so as to obtain a new target bullet screen classification model.
Optionally, the obtaining module 31 is further configured to perform obtaining of bullet screen samples of a first category from the multiple live broadcast rooms, and obtaining bullet screen samples of a second category from the multiple live broadcast rooms, so as to obtain a bullet screen sample set;
the training module 35 is configured to perform training on the neural network by using the bullet screen sample set acquired by the acquisition module 31, so as to obtain a general bullet screen classification model; and when the ratio of the number of the newly added samples in the bullet screen sample set to the total number of the samples before addition is greater than or equal to a preset percentage, retraining the neural network by adopting the bullet screen sample set after adding the samples to obtain a target bullet screen classification model.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 5 is a block diagram illustrating a structure of an electronic device according to an exemplary embodiment, where the electronic device may be a bullet screen information processing apparatus, and the bullet screen information processing apparatus may be: a smartphone, a tablet, a laptop, or a desktop computer.
The apparatus for processing bullet screen information may include at least one processor 51, a communication bus 52, a memory 53, and at least one communication interface 54.
The processor 51 may be a Central Processing Unit (CPU), a micro-processing unit, or one or more integrated circuits for controlling the execution of programs in accordance with the disclosed aspects.
The communication bus 52 may include a path for communicating information between the aforementioned components.
The communication interface 54 may be any device, such as a transceiver, for communicating with other devices or communication networks, such as an ethernet, a Radio Access Network (RAN), a Wireless Local Area Network (WLAN), etc.
The memory 53 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device that may store static information and instructions, a Random Access Memory (RAM) or other type of dynamic storage device that may store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disk read-only memory (CD-ROM) or other optical disk storage, optical disk storage (including compact disk, laser disk, optical disk, digital versatile disk, blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be self-contained and connected to the processing unit by a bus. The memory may also be integrated with the processing unit.
The memory 53 is used for storing application program codes for executing the disclosed scheme, and is controlled by the processor 51 to execute. The processor 51 is configured to execute application program code stored in the memory 53 to implement the functions of the disclosed method.
In particular implementations, processor 51 may include one or more CPUs such as CPU0 and CPU1 in fig. 5, for example, as one embodiment.
In a specific implementation, the processing device of the bullet screen information may include a plurality of processors, such as the processor 51 and the processor 55 in fig. 5, as an example. Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor.
In a specific implementation, the apparatus for processing bullet screen information may further include an input device 56 and an output device 57 as an embodiment. The input device 56 is in communication with the processor 51 and can accept user input in a variety of ways. For example, the input device 56 may be a mouse, a keyboard, a touch screen device, or a sensing device, among others. The output device 57 is in communication with the processor 51 and may display information in a variety of ways. For example, the output device 57 may be a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display device, or the like.
Those skilled in the art will appreciate that the configuration shown in fig. 5 is not limiting to electronic devices and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components may be used.
The present disclosure also provides a computer-readable storage medium including instructions stored thereon, which, when executed by a processor of a computer, enable the computer to perform the processing method of bullet screen information provided by the above-described illustrated embodiment. For example, the computer readable storage medium may be a memory 53 comprising instructions executable by the processor 51 of the electronic device to perform the above-described method. Alternatively, the computer readable storage medium may be a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
The present disclosure also provides a computer program product containing instructions, which when run on an electronic device, causes the electronic device to execute the processing method of bullet screen information provided by the above-described illustrated embodiments.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice in the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. A method for processing bullet screen information is characterized by comprising the following steps:
acquiring target bullet screen information of a target live broadcast room;
classifying the target bullet screen information by adopting a target bullet screen classification model corresponding to the target live broadcast room, and determining a target category corresponding to the target bullet screen information, wherein the target bullet screen classification model is obtained by adding bullet screen information meeting preset conditions in the target live broadcast room on the basis of a general bullet screen classification model through training;
and when the target category is a first category, deleting the target barrage information from the target live broadcast room, wherein the target barrage information of the first category is barrage information meeting negative feedback behavior triggering conditions of users.
2. The method for processing barrage information according to claim 1, further comprising:
and when the target bullet screen information is determined to be the bullet screen information deleted by mistake, adding the target bullet screen information into a target bullet screen sample set as a forward sample.
3. The method for processing bullet screen information according to claim 1, wherein the target category is a second category, the target bullet screen information of the second category is normal bullet screen information, and the method for processing bullet screen information further comprises:
and responding to the operation that the account marks the target bullet screen information as the first category, and adding the target bullet screen information as a negative sample into a target bullet screen sample set.
4. The method for processing bullet screen information according to claim 2 or 3, characterized in that the method for processing bullet screen information further comprises:
and when the ratio of the number of the newly added samples in the target bullet screen sample set to the total number of the samples before addition is greater than or equal to a preset percentage, retraining the neural network by adopting the target bullet screen sample set after the samples are added to obtain a new target bullet screen classification model.
5. The method for processing barrage information according to any one of claims 1 to 3, wherein the method for processing barrage information further comprises:
acquiring the barrage samples of the first category from a plurality of live broadcast rooms, and acquiring the barrage samples of the second category from the plurality of live broadcast rooms to obtain a barrage sample set;
training a neural network by adopting the bullet screen sample set to obtain the general bullet screen classification model;
and when the ratio of the number of the newly added samples in the bullet screen sample set to the total number of the samples before addition is greater than or equal to a preset percentage, retraining the neural network by adopting the bullet screen sample set after adding the samples to obtain the target bullet screen classification model.
6. A bullet screen information processing device, comprising:
the acquisition module is configured to execute acquisition of target barrage information of a target live broadcast room;
the determining module is configured to perform classification on the target bullet screen information acquired by the acquiring module by using a target bullet screen classification model corresponding to the target live broadcast room, and determine a target category corresponding to the target bullet screen information, wherein the target bullet screen classification model is obtained by adding bullet screen information meeting preset conditions in the target live broadcast room on the basis of a general bullet screen classification model through training;
and the deleting module is configured to delete the target barrage information from the target live broadcast room when the target category determined by the determining module is a first category, wherein the target barrage information of the first category is barrage information meeting a negative feedback behavior triggering condition of a user.
7. The apparatus for processing barrage information according to claim 6, further comprising: an add module configured to perform:
and when the target bullet screen information is determined to be the bullet screen information deleted by mistake, adding the target bullet screen information into a target bullet screen sample set as a forward sample.
8. An electronic device, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the bullet screen information processing method according to any one of claims 1-5.
9. A computer-readable storage medium having instructions stored thereon, wherein the instructions in the computer-readable storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the method for processing bullet screen information according to any one of claims 1-5.
10. A computer program product comprising computer instructions, characterized in that the computer instructions, when executed by a processor, implement the method for processing barrage information according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210147409.0A CN114650455B (en) | 2022-02-17 | 2022-02-17 | Barrage information processing method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210147409.0A CN114650455B (en) | 2022-02-17 | 2022-02-17 | Barrage information processing method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114650455A true CN114650455A (en) | 2022-06-21 |
| CN114650455B CN114650455B (en) | 2024-03-26 |
Family
ID=81994114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210147409.0A Active CN114650455B (en) | 2022-02-17 | 2022-02-17 | Barrage information processing method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114650455B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106982377A (en) * | 2017-03-29 | 2017-07-25 | 武汉斗鱼网络科技有限公司 | barrage management method and device |
| CN108537176A (en) * | 2018-04-11 | 2018-09-14 | 武汉斗鱼网络科技有限公司 | Recognition methods, device, terminal and the storage medium of target barrage |
| CN111127079A (en) * | 2019-12-04 | 2020-05-08 | 北京奇艺世纪科技有限公司 | Method, device, computer equipment and storage medium for issuing commodity resources |
| CN111225227A (en) * | 2020-01-03 | 2020-06-02 | 网易（杭州）网络有限公司 | Bullet screen publishing method, bullet screen model generating method and bullet screen publishing device |
| CN112235629A (en) * | 2020-10-15 | 2021-01-15 | 广州博冠信息科技有限公司 | Bullet screen shielding method and device, computer equipment and storage medium |
-
2022
- 2022-02-17 CN CN202210147409.0A patent/CN114650455B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106982377A (en) * | 2017-03-29 | 2017-07-25 | 武汉斗鱼网络科技有限公司 | barrage management method and device |
| CN108537176A (en) * | 2018-04-11 | 2018-09-14 | 武汉斗鱼网络科技有限公司 | Recognition methods, device, terminal and the storage medium of target barrage |
| CN111127079A (en) * | 2019-12-04 | 2020-05-08 | 北京奇艺世纪科技有限公司 | Method, device, computer equipment and storage medium for issuing commodity resources |
| CN111225227A (en) * | 2020-01-03 | 2020-06-02 | 网易（杭州）网络有限公司 | Bullet screen publishing method, bullet screen model generating method and bullet screen publishing device |
| CN112235629A (en) * | 2020-10-15 | 2021-01-15 | 广州博冠信息科技有限公司 | Bullet screen shielding method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114650455B (en) | 2024-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10567329B2 (en) | Methods and apparatus for inserting content into conversations in on-line and digital environments | |
| US20230015606A1 (en) | Named entity recognition method and apparatus, device, and storage medium | |
| CN107832433B (en) | Information recommendation method, device, server and storage medium based on conversation interaction | |
| CN109657054B (en) | Abstract generation method, device, server and storage medium | |
| Xu et al. | Improve Chinese word embeddings by exploiting internal structure | |
| CN113168586A (en) | Text classification and management | |
| CN108537176B (en) | Target barrage identification method and device, terminal and storage medium | |
| CN108874886B (en) | Analyzing multimedia content using knowledge graph embedding | |
| US20160055164A1 (en) | News alert system and method | |
| US9959269B1 (en) | Method and system for implementing a website builder | |
| US20190151758A1 (en) | Unique virtual entity creation based on real world data sources | |
| CN111444357A (en) | Content information determination method and device, computer equipment and storage medium | |
| CN111507097A (en) | Title text processing method and device, electronic equipment and storage medium | |
| CN108055617B (en) | Microphone awakening method and device, terminal equipment and storage medium | |
| JP6932360B2 (en) | Object search method, device and server | |
| CN110069769B (en) | Application label generation method and device and storage device | |
| CN112507167A (en) | Method and device for identifying video collection, electronic equipment and storage medium | |
| US11843843B2 (en) | Bullet screen key content jump method and bullet screen jump method | |
| CN109062905B (en) | Barrage text value evaluation method, device, equipment and medium | |
| CN114722832A (en) | Abstract extraction method, device, equipment and storage medium | |
| CN113407775B (en) | Video searching method and device and electronic equipment | |
| Brassett et al. | The Joke Is on Us: Political Comedy in (Late) Neoliberal Times | |
| CN109145261B (en) | Method and device for generating label | |
| CN114650455A (en) | Barrage information processing method and device, electronic equipment and storage medium | |
| CN113515336B (en) | Live room joining method, creation method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |