Bullet screen analysis method and related device

Abstract

The embodiment of the present application discloses a bullet screen analysis method and related device, the method includes obtaining multiple product consultation bullet screens, extracting multiple bullet screen keywords of each bullet screen, matching word sets for multiple bullet screen keywords corresponding to each bullet screen, thereby determining multiple target product consultation bullet screens, wherein there are multiple words with the same attribute characteristics in the word set, and multiple word sets matched by each keyword of the multiple target product consultation bullet screens are the same; performing grammatical conversion processing according to multiple word sets corresponding to the multiple target product consultation bullet screens to obtain target bullet screens; obtaining target bullet screen associated information corresponding to the target bullet screen, the information including at least one of user name, user level, and bullet screen sending time; sending an indication message to the anchor end, the indication message is used to instruct the anchor end to display the target bullet screen, and to respond to the anchor's selection operation for the target bullet screen to display the target bullet screen associated information. The present application is conducive to improving the convenience of the anchor to view audience questions.

Description

Technical Field

[0001] The present application relates to the general data processing technology field of the Internet industry, and specifically to a bullet comment analysis method and related devices. Background Art

[0002] At present, the sales anchors of the product-selling type generally face a large audience group during live broadcasts, and the barrage interactions are frequent. In addition to a large number of meaningless barrages, there are also quite a few barrages that raise questions about the products being sold. However, at this stage, apart from the anchor's visual inspection and simple screen filtering methods, it is impossible to effectively extract the consultation questions about the products being sold in the audience's barrages. Summary of the invention

[0003] The embodiments of the present application provide a bullet screen analysis method and related devices, in order to improve the convenience for the host to view audience questions.

[0004] In a first aspect, an embodiment of the present application provides a bullet screen analysis method, which is applied to a bullet screen analyzer in a live broadcast system, wherein the live broadcast system includes the bullet screen analyzer, a platform server, an anchor terminal, and a user terminal, wherein the bullet screen analyzer, the anchor terminal, and the user terminal are all connected to the platform server in communication, and the method includes:

[0005] Get multiple product consultation comments from multiple users;

[0006] Extracting multiple bullet-screen keywords from each of the product inquiry bullet-screens;

[0007] Matching the multiple bullet-screen keywords of each of the product inquiry bullet-screens with an adapted word set, wherein the word set stores multiple words with the same attribute characteristics, and the word set is named according to the attribute characteristics;

[0008] Determine a plurality of target product consultation bullet screens, wherein the plurality of bullet screen keywords respectively corresponding to the plurality of target product consultation bullet screens have the same plurality of word sets matched;

[0009] Performing grammatical conversion processing on the names of the plurality of word sets corresponding to the plurality of target product consultation bullet screens to obtain a target bullet screen;

[0010] Acquire target bullet screen association information associated with the plurality of target product consultation bullet screens corresponding to the target bullet screen, wherein the target bullet screen association information includes at least one of a user name, a user level, and a bullet screen sending time of a user initiating the bullet screen;

[0011] An instruction message carrying the target barrage and the target barrage association information is sent to the anchor end, wherein the instruction message is used to instruct the anchor end to display the target barrage on the screen, and to display the target barrage association information on the screen in response to the anchor's selection operation for the target barrage.

[0012] In a second aspect, an embodiment of the present application provides a barrage analysis device, which is applied to an anchor end in a live broadcast system, wherein the live broadcast system includes the anchor end, a barrage analyzer, a platform server, and a user end, wherein the barrage analyzer, the anchor end, and the user end are all connected to the platform server in communication, and the method includes:

[0013] Receiving an indication message carrying a target barrage and target barrage association information from a barrage analyzer, wherein the target barrage association information is content associated with a plurality of target product consultation barrages corresponding to the target barrage, and the target barrage association information includes at least one of a user name, a user level, and a barrage sending time of a user initiating the barrage;

[0014] Displaying the target bullet screen on the screen according to the instruction message;

[0015] Obtaining the host's selection operation for the target bullet screen;

[0016] In response to the selection operation, displaying the bullet screen-related information on the screen;

[0017] The target bullet screen is a result obtained by the bullet screen analyzer performing a bullet screen analysis operation on a plurality of product consultation bullet screens from a plurality of user terminals, and the bullet screen analysis operation includes:

[0018] The bullet comment analyzer extracts multiple bullet comment keywords from each of the product inquiry bullet comments;

[0019] The bullet comment analyzer matches each of the multiple bullet comment keywords of each of the product inquiry bullet comments with an adapted word set, wherein the word set stores multiple words with the same attribute characteristics, and the word set is named according to the attribute characteristics;

[0020] The bullet-screen analyzer determines that a plurality of target product consultation bullet-screens have the same plurality of word sets matched by the plurality of bullet-screen keywords corresponding to the plurality of target product consultation bullet-screens;

[0021] The bullet comment analyzer performs grammatical conversion processing on the names of multiple word sets corresponding to multiple target product consultation bullet comments to obtain a target bullet comment.

[0022] In a third aspect, an embodiment of the present application provides a server, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.

[0023] In a fourth aspect, an embodiment of the present application provides a computer storage medium that stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute some or all of the steps described in the first aspect of this embodiment.

[0024] It can be seen that in the present application, the barrage analyzer first obtains multiple product consultation barrages from multiple user terminals; then extracts multiple barrage keywords of each product consultation barrage; thereby matching the multiple barrage keywords of each product consultation barrage with an adapted word set, and determining multiple target product consultation barrages, the word set stores multiple words with the same attribute characteristics, the word set is named according to the attribute characteristics, and the multiple word sets matched by the multiple barrage keywords corresponding to the multiple target product consultation barrages are the same; then the names of the multiple word sets corresponding to the multiple target product consultation barrages are subjected to grammatical conversion processing to obtain the target barrage; the barrage analyzer will also obtain the target barrage associated information associated with the multiple target product consultation barrages corresponding to the target barrage, the target barrage associated information includes at least one of the user name, user level, and barrage sending time of the barrage initiating user; and finally sends an indication message carrying the target barrage and the target barrage associated information to the anchor terminal, the indication message is used to instruct the anchor terminal to display the target barrage on the screen, and to display the target barrage associated information on the screen in response to the anchor's selection operation for the target barrage. The content of the target barrage determined in this way is highly accurate, which is conducive to improving the effectiveness of the host's interaction. In addition, by processing multiple target consultation barrages to obtain the target barrage, the product consultation barrage with repeated content can be reduced, which is conducive to improving the convenience of the host to view the audience's questions, so as to promptly answer and feedback the audience's questions, improve the experience of both parties, and promote transactions. In addition, in this application, the host end can also display the target barrage associated information corresponding to the target barrage through the selection operation for the target barrage, so that the host can view the user information of the barrage initiator associated with the target barrage, so as to be able to answer questions related to the target barrage for a specific user, which is conducive to improving the interactivity between the host and the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1a is a schematic diagram of the architecture of an example live broadcast system provided in an embodiment of the present application;

[0027] Figure 1b This is a composition example diagram of a server provided in an embodiment of the present application;

[0028] Figure 2 It is a flowchart of a bullet comment analysis method provided in an embodiment of the present application;

[0029] Figure 3 This is an interface display diagram of a host terminal provided in an embodiment of the present application;

[0030] Figure 4 This is another interface display diagram of the anchor terminal provided in an embodiment of the present application;

[0031] Figure 5 This is another interface display diagram of the anchor terminal provided in the embodiment of the present application;

[0032] Figure 6a This is a block diagram of the functional units of a bullet comment analysis device provided in an embodiment of the present application;

[0033] Figure 6b This is a block diagram of the functional units of another barrage analysis device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0035] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0036] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0037] The embodiments of the present application are described below with reference to the accompanying drawings.

[0038] The technical solution of this application can be applied to Figure 1a The live broadcast system shown includes a barrage analyzer 100, a platform server 200, an anchor terminal 300, and a user terminal 400. The barrage analyzer 100, the anchor terminal 300, and the user terminal 400 are all communicatively connected to the platform server 200.

[0039] In a specific implementation, the barrage analyzer 100 and the platform server 200 can be independently deployed in two servers respectively, or the barrage analyzer 100 and the platform server 200 can be deployed in the same server, without further limitation here.

[0040] The composition structure of the electronic device in this application can be as follows Figure 1b As shown, the electronic device may be a barrage analyzer 100, a platform server 200, a host terminal 300, or a user terminal 400. The electronic device may include a processor 110, a memory 120, a communication interface 130, and one or more programs 121, wherein the one or more programs 121 are stored in the above-mentioned memory 120 and are configured to be executed by the above-mentioned processor 110, and the one or more programs 121 include instructions for executing any step in the above-mentioned method embodiment.

[0041] Among them, the communication interface 130 is used to support the communication between the electronic device and other devices. The processor 110 can be, for example, a central processing unit (CPU), a general processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logic blocks, units and circuits described in conjunction with the disclosure of the embodiments of the present application. The processor can also be a combination that implements a computing function, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0042] The memory 120 may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. Among them, the nonvolatile memory may be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DR RAM).

[0043] In a specific implementation, the processor 110 is used to execute any step performed by the barrage analyzer, platform server, anchor terminal, or user terminal in the following method embodiment, and when performing operations such as sending indication messages, the communication interface 130 can be selectively called to complete the corresponding operation.

[0044] It should be noted that the structural diagram of the above electronic device is only an example, and the specific devices included may be more or less, and no sole limitation is made here.

[0045] See also Figure 2 , Figure 2 is a flow chart of a bullet comment analysis method provided in an embodiment of the present application, which can be applied in the following aspects: Figure 1a The bullet screen analyzer 100 in the live broadcast system shown in FIG. Figure 2 As shown, the bullet comment analysis method includes:

[0046] S210, obtaining multiple product consultation bullet comments from multiple users.

[0047] Among them, the product consultation barrage is the consultation content initiated by users watching the live broadcast through the user terminal regarding the products sold in the live broadcast.

[0048] In a specific implementation, multiple users watching the live broadcast can send bullet comments to the platform server, and the platform server can send the bullet comments to the anchor terminal and the bullet comment analyzer. The bullet comments obtained by the bullet comment analyzer from the platform server can be product consultation bullet comments, or the bullet comments obtained by the bullet comment analyzer from the platform server can be bullet comments sent by the user terminal, and the bullet comment analyzer can filter the bullet comments to obtain product consultation bullet comments.

[0049] Specifically, the barrage analyzer can periodically obtain barrages from the platform server according to a preset duration.

[0050] S220, extracting multiple bullet screen keywords of each product inquiry bullet screen.

[0051] In a specific implementation, the bullet-screen keywords can be words directly extracted from the product consultation bullet-screen to improve the convenience of the bullet-screen analyzer. Alternatively, the bullet-screen keywords can be words extracted from the semantic information obtained by parsing the product consultation bullet-screen, so as to ensure the accuracy of the extracted bullet-screen keywords.

[0052] S230, matching the multiple bullet screen keywords of each product inquiry bullet screen with an adapted word set, wherein the word set stores multiple words with the same attribute characteristics, and the word set is named according to the attribute characteristics.

[0053] The word set refers to a set of words with the same attribute characteristics. The name of the word set is named according to the corresponding attribute characteristics. For example, the set of size-related words such as "M code", "S code", and "L code" can be the words in the word set "size".

[0054] Specifically, the word set is pre-stored content.

[0055] In the specific implementation, after the barrage analyzer extracts the barrage keywords for each product consultation barrage, the barrage analyzer will match them according to the barrage keywords, and match the word set in the word set that has words corresponding to the barrage keywords to the word set that is adapted to the barrage keywords. In this way, each product consultation information can be matched to at least one word set.

[0056] In a specific implementation, when the bullet chat analyzer matches according to bullet chat keywords, the bullet chat analyzer can match synonyms and near synonyms for the bullet chat keywords according to the semantics of the commodity consultation bullet chat, and match the adapted word set according to the bullet chat keywords and the corresponding synonyms and near synonyms. At this time, the word set includes the bullet chat keywords and at least one of the corresponding synonyms and near synonyms. Matching the word set together with the bullet chat keywords and the corresponding synonyms and near synonyms can improve the matching success rate.

[0057] Specifically, if the bullet-screen keywords and their corresponding synonyms and near-synonyms match different sets of words, the word set that matches the bullet-screen keywords can be determined according to the preset priority, and the preset priority can be that the word set corresponding to the bullet-screen keywords is higher than the word set corresponding to the synonyms, and the word set corresponding to the synonyms is higher than the word set corresponding to the near-synonyms. In this way, the accuracy of the matched word set can be guaranteed.

[0058] S240, determining a plurality of target product consultation bullet screens, wherein the plurality of word sets matched by the plurality of bullet screen keywords respectively corresponding to the plurality of target product consultation bullet screens are the same.

[0059] The target product consultation barrage refers to one of a plurality of product consultation barrages having the same set of words matched by each keyword.

[0060] In the specific implementation, the bullet comment analyzer extracts the bullet comment keywords of each product consultation bullet comment, and after matching them with an adapted word set, the product consultation bullet comments with the same matching word set will be classified into one category. The bullet comment analyzer can classify all the obtained product consultation bullet comments into multiple categories, each of which includes at least one product consultation bullet comment. The product consultation bullet comments under each category are a group of target product consultation bullet comments.

[0061] S250, performing grammatical conversion processing on the names of the multiple word sets corresponding to the multiple target product consultation bullet screens to obtain a target bullet screen.

[0062] Specifically, the barrage analyzer may perform grammatical conversion processing on the names of multiple word sets corresponding to multiple target product barrages, thereby generating a sentence according to the names of the multiple word sets, and this sentence is the processed target barrage.

[0063] S260, obtaining target barrage association information associated with the multiple target product consultation barrages corresponding to the target barrage, the target barrage association information including at least one of the user name, user level, and barrage sending time of the barrage initiator user.

[0064] Among them, the user name refers to the user's user name in the live broadcast service provided by the platform server, the user level refers to the user's fan group level in the anchor's live broadcast room, and the barrage sending time refers to the time when the user end sends the product consultation barrage to the platform server.

[0065] The target barrage association information includes the barrage association information of each target product consultation barrage corresponding to the target barrage. Specifically, the barrage analyzer can obtain the barrage association information corresponding to each product consultation barrage when obtaining the product consultation barrage. When the target barrage is determined, the barrage association information of each target product consultation barrage corresponding to the target barrage is the content of the target barrage association information corresponding to the target barrage.

[0066] S270, sending an instruction message carrying the target barrage and the target barrage association information to the anchor end, wherein the instruction message is used to instruct the anchor end to display the target barrage on the screen, and to display the target barrage association information on the screen in response to the anchor's selection operation for the target barrage.

[0067] The instruction message is used to instruct the anchor to perform corresponding operations according to the instruction message.

[0068] In a specific implementation, after receiving the indication message, the host side will directly display the content of the target barrage on the screen for the host to view. If the host side detects that the host has performed a selection operation on the target barrage, the target barrage related information can be displayed on the screen. Among them, the host's selection operation on the target barrage can be for the host to click on the area on the screen where the target barrage is located, so that the host can operate. The target barrage related information can be displayed on the screen in the form of a pop-up window. Alternatively, the target barrage related information can be displayed in the form of an indicator frame in a position adjacent to the target barrage, so that the target barrage related information is more targeted, which can avoid the problem of inconsistent information caused by the host's wrong click.

[0069] Specifically, the bullet screen analyzer can send the instruction message to the platform server, which is then forwarded to the anchor terminal. Alternatively, the bullet screen analyzer can directly communicate with the anchor terminal, thereby directly sending the instruction message to the anchor terminal to improve the efficiency of the target bullet screen display.

[0070] Furthermore, the indication message may also carry multiple target product consultation bullets corresponding to the target bullet, and the indication message may also be used to respond to the host's selection operation for a specific user in the target bullet information to display the original product consultation bullet sent by the specific user on the screen. This allows the host to understand the actual bullet content sent by the user, which is conducive to improving functionality.

[0071] It can be seen that in the present application, the barrage analyzer first obtains multiple product consultation barrages from multiple user terminals; then extracts multiple barrage keywords of each product consultation barrage; thereby matching the multiple barrage keywords of each product consultation barrage with an adapted word set, and determining multiple target product consultation barrages, the word set stores multiple words with the same attribute characteristics, the word set is named according to the attribute characteristics, and the multiple word sets matched by the multiple barrage keywords corresponding to the multiple target product consultation barrages are the same; then the names of the multiple word sets corresponding to the multiple target product consultation barrages are subjected to grammatical conversion processing to obtain the target barrage; the barrage analyzer will also obtain the target barrage associated information associated with the multiple target product consultation barrages corresponding to the target barrage, the target barrage associated information includes at least one of the user name, user level, and barrage sending time of the barrage initiating user; and finally sends an indication message carrying the target barrage and the target barrage associated information to the anchor terminal, the indication message is used to instruct the anchor terminal to display the target barrage on the screen, and to display the target barrage associated information on the screen in response to the anchor's selection operation for the target barrage. The content of the target barrage determined in this way is highly accurate, which is conducive to improving the effectiveness of the host's interaction. In addition, by processing multiple target consultation barrages to obtain the target barrage, the product consultation barrage with repeated content can be reduced, which is conducive to improving the convenience of the host to view the audience's questions, so as to promptly answer and feedback the audience's questions, improve the experience of both parties, and promote transactions. In addition, in this application, the host end can also display the target barrage associated information corresponding to the target barrage through the selection operation for the target barrage, so that the host can view the user information of the barrage initiator associated with the target barrage, so as to be able to answer questions related to the target barrage for a specific user, which is conducive to improving the interactivity between the host and the user.

[0072] In one possible example, when there are multiple target barrages, the indication message is used to instruct the anchor end to display multiple target barrages on the screen according to the number of target product consultation barrages corresponding to the target barrages, and the more target product consultation barrages corresponding to the target barrages, the closer the target barrages are displayed.

[0073] In the specific implementation, the barrage analyzer can sort each target barrage according to the number of target product consultation barrages corresponding to each target barrage, and send an indication message of multiple target barrages carrying the sorting order to the anchor end, so that the display screen of the anchor end can give priority to displaying the target barrages with a large number of target consultation barrages according to the sorting order, so that the anchor can give priority to answering the questions asked by users more frequently, thereby giving priority to serving most users and meeting the needs of most users.

[0074] In particular, when the number of target consultation barrages corresponding to target barrages is the same, the barrage analyzer can further sort the target barrages with reference to the user level corresponding to each target barrage when sorting the target barrages. For example, according to the settings of the anchor end, the target barrages with higher user levels can be sorted in front, so as to give priority to old customers; or, according to the settings of the anchor end, the target barrages with lower user levels can be sorted in front, which is conducive to cultivating new customers. It can be set according to the actual needs of the anchor, and no further restrictions are made here. Specifically, the anchor end can send a sorting message to the barrage analyzer, and the sorting message is used to instruct the barrage analyzer to sort the user level according to the anchor's demand information. The anchor's demand information can be low-level or high-level in front. Among them, the user level of the target barrage refers to the sum of the user levels corresponding to all target product consultation barrages corresponding to the target barrage.

[0075] In a specific implementation, when there are multiple target bullet comments, all target bullet comments can be displayed on the screen of the host at the same time according to the number of target bullet comments and the screen size of the host's device, or all target bullet comments can be displayed on the screen of the host multiple times. It can be understood that when all target bullet comments are displayed on the screen of the host multiple times, at least one target bullet comment is displayed each time. Figure 5 As shown, multiple target bullet comments can be displayed on the host's screen.

[0076] It can be seen that in this example, by arranging the display order of the target barrage according to the user level corresponding to the target barrage on the anchor side, it is convenient for the anchor to identify the consultation situation of each target barrage, and can promptly send the target barrage of high-level users to provide more thoughtful services to old customers and improve user experience.

[0077] In a possible example, the instruction message is used to instruct the anchor end to display the user name on the screen according to the user level in response to the anchor's selection operation for the target barrage.

[0078] In actual applications, when answering questions in the bullet screen, the anchor can answer questions for a specific user in order to improve the interactivity with the users watching the live broadcast. Figure 5 For the question "What sizes are available for the product you are explaining now?", the anchor can answer "Baby Zhang San, this dress is available in sizes from XS to XL, so you can buy it with confidence" in order to improve the interactivity with the user.

[0079] In a specific implementation, the barrage analyzer may sort the barrage association information corresponding to each target product consultation barrage according to the user levels corresponding to the multiple target product consultation barrages corresponding to the target barrage.

[0080] Specifically, if the business model of the live broadcast room pays more attention to the cultivation of new customers, the barrage analyzer can sort the barrage-related information corresponding to each target product consultation barrage according to the user level corresponding to the multiple target product consultation barrages corresponding to the target barrage from low to high. The lower the level of the target product consultation barrage, the higher the barrage-related information corresponding to the barrage will be. In this way, the host can answer questions targeting a user with a low level, which is conducive to cultivating new customers.

[0081] Or, specifically, if the business model of the live broadcast room focuses more on maintaining old customers, the barrage analyzer can sort the barrage-related information corresponding to each target product consultation barrage according to the user levels corresponding to the multiple target product consultation barrages corresponding to the target barrage, from high to low. The higher the level of the target product consultation barrage, the higher the barrage-related information corresponding to the barrage will be. In this way, the host can answer questions targeting a user with a high user level, which is conducive to maintaining old customers.

[0082] Alternatively, specifically, the bullet comment analyzer can also divide the user levels into stages, for example, levels 0 to 5 can be divided into one stage, levels 6 to 10 can be divided into one stage, and so on. In this way, the bullet comment analyzer can randomly select at least one user in each stage, and display the corresponding bullet comment related information in the current interface, so as to better serve users at each user level stage.

[0083] It can be seen that in this example, by sorting and displaying the corresponding user names according to the user levels of the users corresponding to each target consultation barrage on the anchor side, better services can be provided to the target customer group, which is conducive to improving the transaction success rate with the target customer group.

[0084] In a possible example, the instruction message is used to instruct the anchor end to display the user name and the user level corresponding to the user name on the screen according to the time when the barrage was sent in response to the anchor's selection operation for the target barrage.

[0085] In the specific implementation, the barrage analyzer can sort the barrage related information corresponding to the target product consultation barrage with a later barrage sending time in front, so that the barrage related information corresponding to the target product consultation barrage with a later barrage sending time is displayed first on the screen of the anchor end, thereby reducing the possibility that the user corresponding to the displayed barrage related information will leave the live broadcast room.

[0086] It can be seen that in this example, by sorting and displaying the corresponding user names and user levels according to the sending time of each target consultation barrage on the anchor side, better services can be provided to customers. This can avoid the situation where customers leave the live broadcast room and cannot hear the anchor's response due to long waiting time, which is conducive to ensuring that every answer from the anchor is effective.

[0087] In a possible example, when the bullet comment analyzer forwards the indication message to the anchor end through the platform server, the platform server can highlight the bullet comment associated information of the users who are still in the live broadcast room in the sorted target bullet comment associated information according to the information of the users who are currently in the live broadcast room, so as to facilitate the anchor to distinguish.

[0088] In a possible example, extracting multiple barrage keywords of each product consultation barrage includes: parsing the multiple product consultation barrages to obtain semantic information of each product consultation barrage; extracting the barrage keywords of the product consultation barrage from the semantic information corresponding to the product consultation barrage.

[0089] Among them, the semantic information is the content obtained by the barrage analyzer through semantic recognition processing of the product consultation barrage.

[0090] In a specific implementation, after obtaining the product consultation barrage, the barrage analyzer can perform semantic recognition processing on the product consultation barrage, thereby obtaining semantic information corresponding to the product consultation barrage, and then extract multiple barrage keywords corresponding to the product consultation barrage from the semantic information.

[0091] Specifically, in the process of semantic recognition processing of multiple product consultation barrages, the barrage analyzer will filter out product consultation barrages with unclear semantics. And parse to obtain the semantic information of the remaining product consultation barrages. For example, the product consultation barrage is "Is there any blue one?", and one possible result after semantic recognition processing is "Are there any blue products currently being explained?" Then the barrage analyzer will perform stop word removal processing on each product consultation barrage. This process is used to remove non-important words (for example: de, di, de) or punctuation marks in the semantic information corresponding to each product consultation barrage. For example: "Are there any blue products currently being explained?" After the stop word removal process, one possible result is: "Are there any blue products currently being explained?" The above-mentioned stop word removal processing operation is performed based on a stop word removal dictionary, which can use an existing stop word removal dictionary, or can be obtained by training based on the data stored in the database to obtain a stop word removal dictionary suitable for the current live broadcast product type.

[0092] Furthermore, the semantic information corresponding to each product consultation bullet screen after the stop word processing is removed is subjected to syntactic analysis, and word segmentation processing is performed according to the analysis result, so as to obtain multiple bullet screen keywords to be extracted. Specifically, for example, the bullet screen keywords obtained in advance from the above data may include "current product explained" and "blue".

[0093] It can be seen that in this example, the barrage analyzer obtains the semantic information of the product consultation barrage by parsing the product consultation barrage, and then extracts the barrage keywords from the semantic information, which is conducive to ensuring the reliability of the correspondence between the extracted barrage keywords and the product consultation barrage content, and avoiding the influence of the barrage analysis results due to user typing errors.

[0094] In a possible example, the names of the multiple word sets corresponding to the multiple target product consultation barrages are subjected to grammatical conversion processing, including: according to the semantic information of any one of the multiple target product consultation barrages, obtaining the association relationship between each word set corresponding to the target barrage, the association relationship is used to characterize the positional relationship between each word set in the semantic information; according to the association relationship, sentence reorganization is performed on the multiple word sets corresponding to the multiple target product consultation barrages.

[0095] In a specific implementation, the barrage analyzer can select any semantic information corresponding to a target product consultation barrage as the sentence pattern corresponding to the target barrage. Then, the association relationship of each word set is obtained based on the semantic information, that is, the position of the barrage keywords in the target product consultation barrage corresponding to the semantic information in the semantic information is determined. Then, the multiple word sets are reorganized according to the association relationship to obtain the target barrage. Specifically, the specific operation of reorganizing multiple word sets according to the association relationship is: bring each word set into the position of the corresponding barrage keyword in the above semantic information.

[0096] Exemplarily, the semantic information of multiple target product consultation bullet comments at least includes: "Does the product currently being explained have size S?", "What models are available for the product currently being explained?", "What sizes are available for the product currently being explained?". Bullet comment keywords are extracted for each of the above target product consultation bullet comments, and the matching word sets are: "What models are available for the product currently being explained?", "Sizes." The bullet comment analyzer can use the semantic information "What models are available for the product currently being explained" as the sentence pattern of the target bullet comment, and then determine the positions of the bullet comment keyword "What products are currently being explained" corresponding to the word set "What products are currently being explained" and the bullet comment keyword "model" corresponding to the word set "Sizes" in the semantic information. Then, the two bullet comment keywords "What products are currently being explained" and "Sizes" are brought into the corresponding positions in the semantic information to generate the target bullet comment "What sizes are available for the product currently being explained?".

[0097] It can be seen that in this example, the barrage analyzer determines the association relationship between the various word sets corresponding to the target barrage through the semantic information of the target product consultation barrage, and reorganizes the sentences of multiple word sets according to the association relationship, so as to classify the contents of multiple target product consultation barrages and generate target barrages from the classified contents, thereby ensuring that the host can understand the content of a large amount of barrage in a timely manner and facilitate the host's review.

[0098] In one possible example, before extracting multiple barrage keywords of each of the product consultation barrages, the method also includes: determining whether the product consultation barrage includes a product number or a demonstrative pronoun; if so, determining the product corresponding to the product number or the demonstrative pronoun as the target product corresponding to the product consultation barrage according to a shopping list, the shopping list including the currently explained product, historically explained products, and to-be-explained products; if not, determining the currently explained product as the target product corresponding to the product consultation barrage.

[0099] In actual applications, for the sake of convenience, users will not directly enter the name of the product. They may use some abbreviated indicative words or use the product number to refer to the product. In order to make it easier for the anchor to answer questions about a specific product, it is necessary to further determine whether the product is clear before extracting the barrage keywords.

[0100] In the specific implementation, the barrage analyzer will first determine whether the product consultation barrage includes a specific product name. If so, it indicates that the target product is clear and subsequent operations can be performed. If not, it is determined whether the product consultation barrage includes a product number. If so, the product corresponding to the product number is determined as the target product corresponding to the product consultation barrage according to the shopping list, and subsequent operations are performed. If not, it is determined whether the product consultation barrage includes a demonstrative pronoun. If so, the product corresponding to the demonstrative pronoun is determined as the target product corresponding to the product consultation barrage according to the shopping list, and subsequent operations are performed. Exemplarily, if demonstrative pronouns such as "this" and "this" are used in the barrage consultation information, the current explanation product can be determined according to the shopping list, and the current explanation product can be determined as the target product. Alternatively, if demonstrative pronouns such as "the pair of shoes just now" are used in the barrage consultation information, the shoe type product before the current explanation product can be queried according to the list order of the shopping list, and the product is determined as the target product, which is also a historical explanation product. If the product consultation barrage does not include demonstrative pronouns, it is determined that the product consultation barrage uses an abbreviated expression, such as "Do you have a blue one?". At this time, the product currently being explained can be determined as the target product, thereby completing the semantics of the product consultation barrage.

[0101] It can be seen that in this example, the barrage analyzer can determine the target product corresponding to the product consultation barrage based on the live broadcast situation of the live broadcast room, so that the anchor can answer questions about a specific product, so that users can understand the information of each product more clearly.

[0102] The embodiment of the present application provides a bullet comment analysis method, which can be applied in the following aspects: Figure 1a The anchor end 300 in the live broadcast system shown, the barrage analysis method includes: receiving a target barrage and target barrage association information from a barrage analyzer, the target barrage association information being content associated with a plurality of target product consultation barrages corresponding to the target barrage, the target barrage association information including at least one of the user name, user level, and barrage sending time of the user initiating the barrage; displaying the target barrage; obtaining a selection operation of the live broadcast for the target barrage; and displaying the barrage association information according to the selection operation.

[0103] Among them, the target barrage is the result obtained by the barrage analyzer performing a barrage analysis operation on multiple product consultation barrages from multiple user terminals, and the barrage analysis operation includes: the barrage analyzer extracts multiple barrage keywords for each of the product consultation barrages; the barrage analyzer matches the multiple barrage keywords of each of the product consultation barrages with an adapted word set, and the word set includes multiple words with the same attribute characteristics, and the word set is named according to the attribute characteristics; the barrage analyzer determines multiple target product consultation barrages, and the multiple word sets matched by the multiple barrage keywords corresponding to the multiple target product consultation barrages are the same; the barrage analyzer performs grammatical conversion processing on the names of the multiple word sets corresponding to the multiple target product consultation barrages to obtain the target barrage.

[0104] Specifically, after receiving the instruction message, the host will directly display the content of the target bullet screen on the screen for the host to view. If the host detects that the host has performed a selection operation on the target bullet screen, the target bullet screen associated information can be displayed on the screen. The host can refer to the above example to display the content of the target bullet screen and the target bullet screen associated information, which will not be further described here.

[0105] See also Figures 3 to 5 , the anchor will receive the user's bullet comments from the platform server in real time and display them on the anchor's screen, such as Figure 3 As shown. The real-time user bullet screen can be typed by multiple user terminals and sent to the platform server. The server can also send the user bullet screen to the bullet screen analyzer, and the bullet screen analyzer performs bullet screen analysis and processing to obtain multiple target bullet screens. The bullet screen analyzer can send an indication message carrying the target bullet screen and the target bullet screen association information corresponding to the target bullet screen to the anchor terminal. After receiving the indication message, the anchor terminal will display the target bullet screen on the screen, as shown. Figure 4 As shown. The target bullet screen can be displayed on top of the user bullet screen, or can be displayed floating above the user bullet screen, without further restriction. Afterwards, if the anchor detects that the anchor has selected a specific target bullet screen, the target bullet screen related information for the target bullet screen will be displayed on the screen. For example, see Figure 5 When the host detects a selection operation for the target bullet screen "What sizes are available for the product being explained?", the host will display the target bullet screen related information corresponding to the target bullet screen on the screen, and the target bullet screen related information includes the user name, user level, and bullet screen initiation time of the bullet screen initiator of each target product consultation bullet screen corresponding to the Mu Bai Oh bullet screen. Furthermore, based on the host displaying the target bullet screen related information according to the instruction message, if the host detects a selection operation for the target bullet screen "What sizes are available for the product being explained?" again, the target bullet screen related information will be folded (or hidden).

[0106] In the specific implementation, if the host detects a selection operation (such as long press, etc.) on the blank area displayed on the display screen, the host end will temporarily hide the target barrage and the target barrage associated information and display the user barrage.

[0107] In a specific implementation, the host side can display the above target bullet screen and target bullet screen associated information according to the host's on setting for displaying the target bullet screen. And according to the host's off setting for displaying the target bullet screen, the host side can not display the above target bullet screen and target bullet screen associated information. At this time, the display screen of the host side will always display the user bullet screen.

[0108] It can be seen that in this example, the host side displays the target barrage and target barrage related information according to the instruction message to improve the convenience for the host to view the audience's questions, thereby promptly answering and feeding back the audience's questions and improving the experience of both parties.

[0109] The present application can divide the server into functional units according to the above method examples. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiments of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0110] Figure 6a : is a functional unit block diagram of a bullet comment analysis device provided in an embodiment of the present application. The bullet comment analysis device 30 can be applied in Figure 1a The bullet screen analyzer 100 in the live broadcast system shown in the figure includes the bullet screen analyzer, a platform server, an anchor terminal, and a user terminal. The bullet screen analyzer, the anchor terminal, and the user terminal are all connected to the platform server in communication. The device includes:

[0111] An acquisition unit 310 is used to acquire multiple product consultation bullet comments from multiple user terminals;

[0112] An extraction unit 320 is used to extract multiple bullet-screen keywords of each of the product inquiry bullet-screens;

[0113] A matching unit 330 is used to match the multiple bullet-screen keywords of each of the product inquiry bullet-screens with an adapted word set, wherein the word set includes multiple words with the same attribute characteristics, and the word set is named after the attribute characteristics;

[0114] A determination unit 340 is used to determine a plurality of target product consultation bullet screens, wherein the plurality of bullet screen keywords corresponding to the target product consultation bullet screens match the same plurality of word sets;

[0115] The processing unit 350 is used to perform grammatical conversion processing on the multiple word sets corresponding to the multiple target product consultation bullet comments to obtain a target bullet comment;

[0116] The sending unit 360 is used to send the target bullet to the anchor terminal.

[0117] In one possible example, when there are multiple target barrages, the indication message is used to instruct the anchor end to display multiple target barrages on the screen according to the number of target product consultation barrages corresponding to the target barrages, and the more target product consultation barrages corresponding to the target barrages, the closer the target barrages are displayed.

[0118] In one possible example, the indication message is used to instruct the anchor end to display the user name on the screen according to the user level in response to the anchor's selection operation for the target barrage; or, the indication message is used to instruct the anchor end to display the user name on the screen according to the barrage sending time in response to the anchor's selection operation for the target barrage.

[0119] In one possible example, in terms of extracting multiple bullet-screen keywords from each of the product consultation bullet-screens, the extraction unit is specifically used to: parse the multiple product consultation bullet-screens to obtain semantic information of each of the product consultation bullet-screens; extract the bullet-screen keywords of the product consultation bullet-screens from the semantic information corresponding to the product consultation bullet-screens. In one possible example, in terms of performing grammatical conversion processing on the names of the multiple word sets corresponding to the multiple target product consultation bullet-screens, the processing unit is specifically used to: obtain the association relationship of each of the word sets corresponding to the target bullet-screen according to the semantic information of any one of the multiple target product consultation bullet-screens, the association relationship being used to characterize the positional relationship of each of the word sets in the semantic information; and perform sentence reorganization on the multiple word sets corresponding to the multiple target product consultation bullet-screens according to the association relationship.

[0120] In one possible example, the device also includes a product confirmation unit, which is used to determine whether the product consultation barrage includes a product number or a demonstrative pronoun before extracting multiple barrage keywords of each product consultation barrage; if so, the product corresponding to the product number or demonstrative pronoun is determined as the target product corresponding to the product consultation barrage according to a shopping list, and the shopping list includes the current explained product, historical explained products, and to-be-explained products; if not, the current explained product is determined as the target product corresponding to the product consultation barrage.

[0121] In the case of using an integrated unit, the functional unit composition block diagram of the bullet comment analysis device 40 provided in the embodiment of the present application is as follows: Figure 6b As shown. Figure 6b In the embodiment, the bullet-screen analysis device 40 includes: a processing module 420 and a communication module 410. The processing module 420 is used to control and manage the actions of the bullet-screen analysis device 30, for example, the steps performed by the acquisition unit 310, the extraction unit 320, the matching unit 330, the determination unit 340, the processing unit 350, and the sending unit 360, and / or other processes for performing the technology described herein. The communication module 410 is used to support the interaction between the bullet-screen analysis device 40 and other devices. Figure 6b As shown, the barrage analysis device 40 may further include a storage module 430 , which is used to store program codes and data of the barrage analysis device 40 .

[0122] Among them, the processing module 420 can be a processor or a controller, for example, a central processing unit (CPU), a general processor, a digital signal processor (DSP), an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the embodiments of the present application. The processor can also be a combination that implements a computing function, for example, a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. The communication module 410 can be a transceiver, an RF circuit or a communication interface, etc. The storage module 430 can be a memory.

[0123] Among them, all relevant contents of each scenario involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and will not be repeated here. Figure 2 The steps performed by the barrage analyzer in the barrage analysis method shown.

[0124] The embodiment of the present application also provides a bullet screen analysis device. The bullet screen analysis device can be used in Figure 1a On the anchor terminal 300 in the live broadcast system shown, the live broadcast system includes the bullet screen analyzer, the platform server, the anchor terminal, and the user terminal. The bullet screen analyzer, the anchor terminal, and the user terminal are all connected to the platform server in communication. The device includes:

[0125] A receiving unit, the receiving unit is used to receive an indication message carrying a target barrage and target barrage association information from a barrage analyzer, wherein the target barrage association information is content associated with a plurality of target product consultation barrages corresponding to the target barrage, and the target barrage association information includes at least one of a user name, a user level, and a barrage sending time of a user initiating the barrage;

[0126] A display unit, configured to display the target bullet screen on a screen according to the instruction message; and obtain a selection operation of the anchor for the target bullet screen, and display the bullet screen-related information on the screen in response to the selection operation;

[0127] The target bullet screen is a result obtained by the bullet screen analyzer performing a bullet screen analysis operation on a plurality of product consultation bullet screens from a plurality of user terminals, and the bullet screen analysis operation includes:

[0128] The bullet comment analyzer extracts multiple bullet comment keywords from each of the product inquiry bullet comments;

[0129] The bullet comment analyzer matches each of the multiple bullet comment keywords of each of the product inquiry bullet comments with an adapted word set, wherein the word set stores multiple words with the same attribute characteristics, and the word set is named according to the attribute characteristics;

[0130] The bullet-screen analyzer determines that a plurality of target product consultation bullet-screens have the same plurality of word sets matched by the plurality of bullet-screen keywords corresponding to the plurality of target product consultation bullet-screens;

[0131] The bullet comment analyzer performs grammatical conversion processing on the names of multiple word sets corresponding to multiple target product consultation bullet comments to obtain a target bullet comment.

[0132] Furthermore, the bullet-screen analysis device may include: a processing module and a communication module. The processing module is used to control and manage the actions of the bullet-screen analysis device, for example, the steps performed by the receiving unit and the display unit, and / or other processes for performing the technology described herein. The communication module is used to support the interaction between the bullet-screen analysis device and other devices. The bullet-screen analysis device may also include a storage module, which is used to store program code and data of the bullet-screen analysis device.

[0133] Wherein, the processing module can be a processor or a controller, for example, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute various exemplary logic boxes, modules and circuits described in conjunction with the disclosure of the embodiments of the present application. The processor can also be a combination that implements a computing function, for example, a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. The communication module can be a transceiver, an RF circuit or a communication interface, and the like. The storage module can be a memory.

[0134] Among them, all relevant contents of each scene involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and will not be repeated here. The above bullet screen analysis device can execute the steps executed by the anchor end in the above-mentioned bullet screen analysis method.

[0135] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps of any method recorded in the above method embodiments, and the above computer includes a server.

[0136] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0137] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0138] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0139] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0140] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0141] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the above-mentioned methods of each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0142] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0143] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A bullet comment analysis method, characterized in that: A bullet screen analyzer applied to a live broadcast system, the live broadcast system comprising the bullet screen analyzer, a platform server, an anchor terminal, and a user terminal, the bullet screen analyzer, the anchor terminal, and the user terminal are all connected to the platform server in communication, and the method comprises: Get multiple product consultation comments from multiple users; Determine whether there is a product name in each of the product consultation bullet screens, and if not, determine whether the multiple product consultation bullet screens include product numbers or demonstrative pronouns; if included, determine the product corresponding to the product number or demonstrative pronoun as a target product corresponding to the product consultation bullet screen according to a shopping list, the shopping list including the current explained product, the historical explained products, and the products to be explained; if not included, determine the current explained product as the target product corresponding to the product consultation bullet screen; Extracting multiple bullet-screen keywords from each of the product inquiry bullet-screens; According to the semantics of each of the product consultation bullet screens, the synonyms and near synonyms corresponding to each of the bullet screen keywords corresponding to the product consultation bullet screens are obtained, and according to the single bullet screen keyword and the corresponding synonyms and near synonyms, a target word set is matched for each of the bullet screen keywords, wherein the target word set includes the bullet screen keyword and at least one of the corresponding synonyms and near synonyms; a plurality of words having the same attribute characteristics are stored in the word set, and the word set is named according to the attribute characteristics; If at least two word sets are matched for the barrage keyword according to the barrage keyword and the corresponding synonyms and near-synonyms, the target word set adapted to the barrage keyword is determined according to a preset priority, and the preset priority is: the word set adapted to the barrage keyword is superior to the word set adapted to the synonym, and the word set adapted to the synonym is superior to the word set adapted to the near-synonym; Determine a plurality of target product consultation bullet screens, wherein the plurality of target word sets matched by the plurality of bullet screen keywords respectively corresponding to the plurality of target product consultation bullet screens are the same; Performing grammatical conversion processing on the names of the plurality of target word sets corresponding to the plurality of target product consultation bullet screens to obtain a target bullet screen; Acquire target bullet screen association information associated with the plurality of target product consultation bullet screens corresponding to the target bullet screen, wherein the target bullet screen association information includes at least one of a user name, a user level, and a bullet screen sending time of a user initiating the bullet screen; An indication message carrying the target barrage and the target barrage associated information is sent to the anchor end, wherein the indication message is used to instruct the anchor end to display the target barrage above the user barrage on the screen by overlaying or floating, and to respond to the anchor's first selection operation for the target barrage to display the target barrage associated information on the screen, and to respond to the anchor's second selection operation for the target user in the target barrage associated information to display the product consultation barrage sent by the user end of the target user on the screen.

2. The method according to claim 1, characterized in that When there are multiple target barrages, the indication message is used to instruct the anchor end to display multiple target barrages on the screen according to the number of target product consultation barrages corresponding to the target barrages. The more target product consultation barrages corresponding to the target barrages, the closer the target barrages are displayed.

3. The method according to claim 1, characterized in that The instruction message is used to instruct the anchor end to display the user name according to the user level on the screen in response to the anchor's first selection operation on the target bullet screen; Alternatively, the instruction message is used to instruct the anchor end to display the user name on the screen according to the time when the barrage was sent in response to the anchor's first selection operation on the target barrage.

4. The method according to claim 1, characterized in that The extracting of multiple bullet screen keywords of each product inquiry bullet screen includes: Parsing the plurality of product consultation bullet comments to obtain semantic information of each product consultation bullet comment; The bullet screen keywords of the product consultation bullet screen are extracted from the semantic information corresponding to the product consultation bullet screen.

5. The method according to claim 2, characterized in that The grammatical conversion process of the names of the plurality of target word sets corresponding to the plurality of target product consultation bullet screens includes: According to the semantic information of any one of the plurality of target product consultation bullet screens, an association relationship of each of the target word sets corresponding to the target bullet screen is obtained, wherein the association relationship is used to represent the position relationship of each of the target word sets in the semantic information; According to the association relationship, the plurality of target word sets corresponding to the plurality of target product consultation bullet screens are sentence reorganized.

6. A bullet comment analysis method, characterized in that: The method is applied to an anchor terminal in a live broadcast system, wherein the live broadcast system comprises the anchor terminal, a bullet screen analyzer, a platform server, and a user terminal, wherein the bullet screen analyzer, the anchor terminal, and the user terminal are all connected to the platform server in communication, and the method comprises: Receiving an indication message carrying a target barrage and target barrage association information from a barrage analyzer, wherein the target barrage association information is content associated with a plurality of target product consultation barrages corresponding to the target barrage, and the target barrage association information includes at least one of a user name, a user level, and a barrage sending time of a user initiating the barrage; Displaying the target bullet screen on top of the user bullet screen by overlaying or floating according to the instruction message; Obtaining the host's first selection operation for the target bullet comment; In response to the first selection operation, displaying the bullet screen associated information on the screen; Obtaining the host's second selection operation for the target bullet screen; In response to the second selection operation, displaying the product consultation bullet screen sent by the user terminal of the target user on the screen; The target bullet screen is a result obtained by the bullet screen analyzer performing a bullet screen analysis operation on a plurality of product consultation bullet screens from a plurality of user terminals, and the bullet screen analysis operation includes: The bullet-screen analyzer determines whether there is a product name in each of the product consultation bullet-screens, and if not, determines whether the multiple product consultation bullet-screens include a product number or a demonstrative pronoun; if included, determines the product corresponding to the product number or the demonstrative pronoun as a target product corresponding to the product consultation bullet-screen according to a shopping list, wherein the shopping list includes a currently explained product, a historically explained product, and a product to be explained; if not included, determines the currently explained product as a target product corresponding to the product consultation bullet-screen; The bullet comment analyzer extracts multiple bullet comment keywords from each of the product inquiry bullet comments; The barrage analyzer obtains the synonyms and near synonyms corresponding to each barrage keyword corresponding to the product consultation barrage according to the semantics of each product consultation barrage, and matches an adapted target word set for each barrage keyword according to a single barrage keyword and the corresponding synonyms and near synonyms, wherein the target word set includes the barrage keyword and at least one of the corresponding synonyms and near synonyms; a plurality of words with the same attribute characteristics are stored in the word set, and the word set is named according to the attribute characteristics; If at least two word sets are matched for the barrage keyword according to the barrage keyword and the corresponding synonyms and near-synonyms, the barrage analyzer determines the target word set adapted to the barrage keyword according to a preset priority, and the preset priority is: the word set adapted to the barrage keyword is superior to the word set adapted to the synonym, and the word set adapted to the synonym is superior to the word set adapted to the near-synonym; The bullet-screen analyzer determines a plurality of target product consultation bullet-screens, and the plurality of target word sets matched by the plurality of bullet-screen keywords respectively corresponding to the plurality of target product consultation bullet-screens are the same; The barrage analyzer performs grammatical conversion processing on the names of multiple target word sets corresponding to multiple target product consultation barrages to obtain target barrages.

7. A bullet comment analysis device, characterized in that: A bullet screen analyzer applied to a live broadcast system, the live broadcast system comprising the bullet screen analyzer, a platform server, an anchor terminal, and a user terminal, the bullet screen analyzer, the anchor terminal, and the user terminal are all connected to the platform server in communication, and the device comprises: An acquisition unit, used to acquire multiple product consultation bullet comments from multiple user terminals; a product confirmation unit, configured to determine whether a product name exists in each of the product consultation bullet screens, and if not, to determine whether the plurality of product consultation bullet screens include a product number or a demonstrative pronoun; if included, to determine the product corresponding to the product number or the demonstrative pronoun as a target product corresponding to the product consultation bullet screen according to a shopping list, wherein the shopping list includes a currently explained product, a previously explained product, and a product to be explained; if not included, to determine the currently explained product as a target product corresponding to the product consultation bullet screen; An extraction unit, used to extract multiple bullet-screen keywords of each of the product consultation bullet-screens; A matching unit, for obtaining synonyms and near synonyms corresponding to each bullet-screen keyword corresponding to the product consultation bullet-screen according to the semantics of each product consultation bullet-screen, and matching an adapted target word set for each bullet-screen keyword according to a single bullet-screen keyword and the corresponding synonyms and near synonyms, wherein the target word set includes the bullet-screen keyword and at least one of the corresponding synonyms and near synonyms; a plurality of words having the same attribute characteristics are stored in the word set, and the word set is named according to the attribute characteristics; The matching unit is further configured to determine the target word set adapted to the barrage keyword according to a preset priority when at least two word sets are matched to the barrage keyword according to the barrage keyword and the corresponding synonyms and near synonyms, wherein the preset priority is: the word set adapted to the barrage keyword is superior to the word set adapted to the synonym, and the word set adapted to the synonym is superior to the word set adapted to the near synonym; A determination unit, configured to determine a plurality of target product consultation bullet screens, wherein the plurality of bullet screen keywords corresponding to the target product consultation bullet screens match the same plurality of target word sets; A processing unit, configured to perform grammatical conversion processing on the plurality of target word sets corresponding to the plurality of target product consultation bullet screens to obtain a target bullet screen; A sending unit is used to send an indication message of the target barrage and target barrage associated information to the anchor end, wherein the indication message is used to instruct the anchor end to display the target barrage above the user barrage on the screen by overlaying or floating, and to respond to the anchor's first selection operation for the target barrage to display the target barrage associated information on the screen, and to respond to the anchor's second selection operation for the target user in the target barrage associated information to display the product consultation barrage sent by the user end of the target user on the screen.

8. A server, characterized in that: The method comprises a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for executing the steps in the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program for electronic data exchange is stored, wherein the computer program enables a computer to execute the steps in the method according to any one of claims 1 to 6.

Images