Multi-platform social media interactive message hierarchical processing method, device and electronic device

Abstract

The embodiment of the present invention provides a multi-platform social media interactive message hierarchical processing method, device and electronic device, belonging to the field of data processing technology, the method comprising: real-time collection of interactive messages of multiple social media platforms within a time period t to form an interactive message set C(t); standardization processing of the interactive message set C(t); feature extraction for each message m∈S(C(t)) in the standard data set S(C(t)); priority calculation of the feature vector F(m) to obtain the priority value p of the message m; when the priority value p is less than a threshold value p0, generating reply suggestion information R1(p) corresponding to the message m, and when the priority value p is greater than or equal to the threshold value p0, generating automatic reply information R2(p) corresponding to the message m; using an interface display function U to process the reply suggestion information R1(p) and the automatic reply information R2(p), and displaying the processed results on a preset interface. The present invention can effectively improve the efficiency of interactive message processing.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a multi-platform social media interactive message hierarchical processing method, device and electronic equipment. Background Art

[0002] Social media interactive messages refer to various forms of information that users communicate and interact with on social media platforms. These messages can be text, pictures, videos or other media formats, and are usually used to connect with other users, share opinions or participate in discussions.

[0003] Interactive messages on social media platforms include but are not limited to the following types:

[0004] 1. Comments: Users can leave comments under the post to express their opinions or ask questions.

[0005] 2. Like: Users can express their attitude or emotion towards a post by clicking the "Like" button or selecting an emoji reaction.

[0006] 3. Share / Forward: Users can share content to their own pages or forward it to other users.

[0007] 4. Private message / direct message: Users can send private messages directly to communicate one-on-one.

[0008] 5. Tags / Mentions: Users can mention other users in their posts by using the @ symbol followed by their username.

[0009] 6. Live Interaction: During the live broadcast, viewers can interact with the host by sending real-time comments.

[0010] 7. Survey / Poll: Creators can initiate polls to allow users to participate and express their opinions.

[0011] These interactive methods not only enhance communication between users, but also help social media platforms become vibrant communities.

[0012] The prior art has the following problems in processing apartment type information:

[0013] 1. Interactive messages are processed on different platforms, with poor scalability: Interactive messages on different social media platforms have different data structures and collection methods, which means that each platform needs to be processed separately. When a new platform is added, the development workload is large, the scalability is poor, and the cost is high.

[0014] 2. Low efficiency and untimely processing of interactive messages: A large number of interactive messages generated by social media accounts are all processed manually in chronological order, without being able to distinguish priorities. A large amount of manpower and time are consumed to process unimportant and non-urgent interactive content, while important and urgent content cannot be processed in a timely manner.

[0015] 3. Automatic reply content templates, poor user experience: Automatic reply template messages triggered by keywords in interactive messages cannot interact with users with accurate content, and answer irrelevant questions, resulting in poor user experience.

[0016] The above problems become technical problems that need to be solved. Summary of the invention

[0017] In view of this, an embodiment of the present invention provides a multi-platform social media interactive message hierarchical processing method, device and electronic device, which at least partially solve the problems existing in the prior art.

[0018] In a first aspect, an embodiment of the present invention provides a multi-platform social media interactive message hierarchical processing method, comprising:

[0019] Collect interactive messages from multiple social media platforms in time period t in real time to form an interactive message set C(t);

[0020] Standardizing the interactive message set C(t) to obtain a standard data set S(C(t));

[0021] For each message m∈S(C(t)) in the standard data set S(C(t)), feature extraction is performed to obtain a feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], where the feature vector F(m) includes user gender GE, user age AG, number of followed users FL, regional information LC, language information LN, message evaluation value EM and message semantics SE;

[0022] The priority calculation function G() is used to calculate the priority of the feature vector F(m) to obtain the priority value p of the message m, p=G(F(m));

[0023] When the priority value p is less than the threshold value p0, the reply suggestion information R1(p) corresponding to the message m is generated; when the priority value p is greater than or equal to the threshold value p0, the automatic reply information R2(p) corresponding to the message m is generated;

[0024] The reply suggestion information R1(p) and the automatic reply information R2(p) are processed using the interface display function U, and the processed result U(R1(p), R2(p)) is displayed on the preset interface.

[0025] According to a specific implementation of the embodiment of the present disclosure, the real-time collection of interactive messages on multiple social media platforms within a time period t includes:

[0026] Determine a set P of social media platforms that contain all the data to be collected;

[0027] For each platform p∈P, define a collection function Cp(t), which returns all messages collected from platform p within time t;

[0028] Aggregate the data collected by all platforms to form an interactive message set C(t).

[0029] According to a specific implementation of the embodiment of the present disclosure, the interaction message set C(t) is standardized to obtain a standard data set S(C(t)), including:

[0030] Define n(v) as the value of field v contained in the interactive message set C(t), min(v) and max(v) as the minimum and maximum values ​​of v respectively, perform numerical conversion on the interactive message set C(t) to obtain a standardized numerical set :

[0031]

[0032] Define c(m) as the interactive message set The category field of message m, e c is the encoding vector of category c, the standardized numerical set Perform encoding conversion to obtain a set of encoding vectors :

[0033] ;

[0034] Define z(v) as the set of encoding vectors The Z score of field v in , μ(v) is the mean of v, σ(v) is the standard deviation of v, and the encoding vector set Perform anomaly detection and obtain the standard data set S(C(t)):

[0035]

[0036] Among them, τ is the set threshold.

[0037] According to a specific implementation of the embodiment of the present disclosure, the feature extraction is performed on each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m), including:

[0038] Set up an evaluation analysis model SF. The evaluation analysis model SF accepts a message m as input and outputs an evaluation polarity p and an evaluation strength s. The evaluation polarity p is a positive or negative number, and the evaluation strength s is a real number between 0 and 1. The message evaluation value EM(m) of the message m is:

[0039] EM(m)=p(m)×s(m).

[0040] According to a specific implementation of the embodiment of the present disclosure, the feature extraction is performed on each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m), and further includes:

[0041] Set up a semantic analysis model SY. The semantic analysis model SY accepts message m as input and outputs the score of one or more semantic categories. Each semantic category YC has a score sc, which indicates the degree to which message m is related to each semantic category. Define the semantic feature SE as:

[0042] SE(m)=s(m)=[s1,s2,…,sn]

[0043] Where s(m) is a vector, the i-th element si in s(m) represents the relevance score between message m and the i-th semantic category, and n is the number of semantic categories.

[0044] According to a specific implementation of an embodiment of the present disclosure, the priority calculation function G() is used to perform priority calculation on the feature vector F(m) to obtain the priority value p of the message m, including:

[0045] Define the priority calculation function G:

[0046]

[0047] Among them α, β, γ, δ, η, ζ, θ, κ, λ i , is the adjustment coefficient, and LNi represents the information of the i-th language.

[0048] According to a specific implementation of the embodiment of the present disclosure, when the priority value p is less than the threshold value p0, generating the reply suggestion information R1(p) corresponding to the message m includes:

[0049] Define the formula for generating reply suggestion information:

[0050] R1(p)=D(m,p)

[0051] Where p is a message parameter, including the context information, conversation history, and user preferences corresponding to message m. D is a conversation management system that generates reply suggestion information based on message m and parameter p.

[0052] The dialog management system extracts key information and intent from the message m, understands the context of the current dialog based on the context information contained in the parameter p, and generates reply suggestions based on the results of semantic understanding and context modeling.

[0053] According to a specific implementation of the embodiment of the present disclosure, the processing of the reply suggestion information R1(p) and the automatic reply information R2(p) by using the interface display function U includes:

[0054] The interface display function U displays the reply suggestion information R1(p) to the user, and accepts the user's modification and confirmation of the reply suggestion information R1(p), and after the user completes the modification and confirmation, automatically sends the user's confirmation message to the target social media;

[0055] After the interface display function U sends the automatic reply message R2(p) directly to the target social media, it displays a message sending completion prompt to the user.

[0056] In a second aspect, an embodiment of the present invention provides a multi-platform social media interactive message hierarchical processing device, comprising:

[0057] The collection module collects interactive messages from multiple social media platforms in a time period t in real time to form an interactive message set C(t);

[0058] A processing module performs standardization processing on the interactive message set C(t) to obtain a standard data set S(C(t));

[0059] The extraction module extracts features from each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], where the feature vector F(m) includes user gender GE, user age AG, number of followed users FL, regional information LC, language information LN, message evaluation value EM and message semantics SE;

[0060] A calculation module performs priority calculation on the feature vector F(m) through a priority calculation function G() to obtain a priority value p of the message m, p=G(F(m));

[0061] A generating module, when the priority value p is less than a threshold value p0, generates a reply suggestion information R1(p) corresponding to the message m, and when the priority value p is greater than or equal to the threshold value p0, generates an automatic reply information R2(p) corresponding to the message m;

[0062] The display module processes the reply suggestion information R1(p) and the automatic reply information R2(p) using the interface display function U, and displays the processed result U(R1(p), R2(p)) on the preset interface.

[0063] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0064] at least one processor; and,

[0065] a memory communicatively connected to the at least one processor; wherein,

[0066] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the multi-platform social media interactive message hierarchical processing method in any of the aforementioned first aspects or any of the implementations of the first aspect.

[0067] In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the multi-platform social media interactive message hierarchical processing method in the aforementioned first aspect or any implementation method of the first aspect.

[0068] In the fifth aspect, an embodiment of the present invention further provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the multi-platform social media interactive message hierarchical processing method in the aforementioned first aspect or any implementation of the first aspect.

[0069] The multi-platform social media interactive message hierarchical processing scheme in the embodiment of the present invention includes: real-time collection of interactive messages of multiple social media platforms within a time period t to form an interactive message set C(t); standardization processing of the interactive message set C(t) to obtain a standard data set S(C(t)); feature extraction for each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], wherein the feature vector F(m) includes user gender GE, user age AG, number of followed users FL, regional information LC, language information LN, Message evaluation value EM and message semantics SE; perform priority calculation on the feature vector F(m) through the priority calculation function G() to obtain the priority value p of the message m, p=G(F(m)); when the priority value p is less than the threshold value p0, generate the reply suggestion information R1(p) corresponding to the message m, and when the priority value p is greater than or equal to the threshold value p0, generate the automatic reply information R2(p) corresponding to the message m; use the interface display function U to process the reply suggestion information R1(p) and the automatic reply information R2(p), and display the processed result U(R1(p), R2(p)) on the preset interface.

[0070] The solution of this application has the following beneficial effects:

[0071] 1) Multi-platform support: Able to collect interactive messages from multiple social media platforms in real time to ensure the comprehensiveness and real-time nature of the data.

[0072] 2) Data standardization and normalization: Extract key features of interactive messages with different data structures, unify data formats, standardize processing, and improve data consistency and processability.

[0073] 3) Improve management efficiency: Through intelligent processing and classification, automatic reply content generation, improved message processing response speed, timely reply to users, and improved efficiency of social media community management.

[0074] 4) Improve user experience: Identify the characteristics of interactive users through natural language algorithms and understand semantics to achieve personalized interaction and timely and effective responses, thereby improving user satisfaction and stickiness. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0076] Figure 1 A schematic flow chart of a method for hierarchical processing of interactive messages on multiple platforms of social media provided by an embodiment of the present invention;

[0077] Figure 2 A schematic flow chart of another method for hierarchical processing of multi-platform social media interactive messages provided by an embodiment of the present invention;

[0078] Figure 3 A schematic diagram of the structure of a device for hierarchical processing of interactive messages on multiple platforms of social media provided by an embodiment of the present invention;

[0079] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0081] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0082] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0083] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0084] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0085] The embodiment of the present disclosure provides a multi-platform social media interactive message hierarchical processing method. The multi-platform social media interactive message hierarchical processing method provided in this embodiment can be executed by a computing device, which can be implemented as software, or as a combination of software and hardware, and the computing device can be integrated in a server, terminal device, etc.

[0086] See also Figure 1 and Figure 2 The present disclosure provides a method for hierarchical processing of interactive messages on multi-platform social media, including:

[0087] S101, collect interactive messages from multiple social media platforms within a time period t in real time to form an interactive message set C(t).

[0088] Collecting interactive messages on multiple social media platforms in a specific time period t in real time and forming an interactive message set C(t) can include the following steps:

[0089] After determining the social media platforms that need to be collected, define the message types to be collected (such as posts, comments, replies, etc.) and set the collection time period t. You can define filtering conditions to select interactive messages of interest, such as keyword searches, activities of specific users, etc.

[0090] You can use the API interfaces provided by various social media platforms to obtain interactive messages. Pull the latest messages from each platform and ensure that the entire time period t is covered. Combine the collected data into a unified set C(t) and store the collected data in a database for subsequent retrieval and analysis.

[0091] S102, standardize the interactive message set C(t) to obtain a standard data set S(C(t)).

[0092] The interactive message set C(t) is standardized to obtain a standard data set S(C(t)), which includes multiple steps such as cleaning, conversion and standardization of data. Specifically, the following steps are performed:

[0093] S1021. Data cleaning

[0094] Remove invalid data: Delete empty messages or messages that do not conform to the format.

[0095] Deduplication: Remove duplicate messages from a data set.

[0096] Fill missing values: For missing fields, you can use the mean, median, or mode to fill in the missing values.

[0097] S1022.Format Conversion

[0098] Standardize date / time format: Convert all date / time fields to a unified format.

[0099] Standardize text format: unify text case, remove punctuation, and perform stemming or word form restoration.

[0100] Convert numeric data: Make all numeric fields consistent in format.

[0101] S1023. Feature extraction

[0102] Extract key information: Extract important information such as username, message text, timestamp, etc.

[0103] Word segmentation: Segment the text and extract keywords.

[0104] Sentiment Analysis: Use sentiment analysis tools to assess the sentiment polarity of messages.

[0105] Semantic analysis: extract the core semantics of the message.

[0106] S1024.Data conversion

[0107] Numerical normalization: converting numerical features to the same scale.

[0108] Categorical variable encoding: One-hot encoding or label encoding is performed on categorical variables.

[0109] Outlier detection: Identify and handle outliers.

[0110] S1025.Data Integration

[0111] Merge data: Combine the processed data into one dataset.

[0112] S103, perform feature extraction on each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], where the feature vector F(m) includes user gender GE, user age AG, number of followed users FL, geographical information LC, language information LN, message evaluation value EM and message semantics SE.

[0113] In order to extract features for each message m∈S(C(t)) in the standard data set S(C(t)) and obtain the feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], the method steps can be defined to extract these features:

[0114] User gender GE can be male or female, and user gender information can be obtained from each message m. Gender can be represented by binary code, such as 0 for male and 1 for female. User age AG can obtain user age information from each message m, and age can be represented by an integer.

[0115] For the number of followed users FL, the number of followers information can be obtained from each message m, and the number of followers is represented by an integer.

[0116] For the geographical information LC, the geographical location of the user can be obtained, and the geographical location information of the user can be obtained from each message m. The geographical location can be represented by a character string or by longitude and latitude coordinates.

[0117] For the language information LN, natural language processing technology can be used to identify the language of each message m, and a string is used to represent the language, for example, "en" represents English and "zh" represents Chinese.

[0118] The message evaluation value EM can be calculated by the number of likes, ratings, etc. For example, the evaluation value information can be obtained from each message m, and the evaluation value can be represented by an integer or floating point number. An evaluation analysis model SF can be set. The evaluation analysis model SF accepts the message m as input and outputs an evaluation polarity p and an evaluation strength s. The evaluation polarity p is a positive or negative number, and the evaluation strength s is a real number between 0 and 1. Then the message evaluation value EM(m) of the message m is:

[0119] EM(m)=p(m)×s(m).

[0120] For the message semantics SE, the specific content of the semantics can be obtained through semantic analysis. For example, a semantic analysis model SY can be set. The semantic analysis model SY accepts the message m as input and outputs the score of one or more semantic categories. Each semantic category YC has a score sc, which indicates the degree of relevance of the message m to each semantic category. The semantic feature SE is defined as:

[0121] SE(m)=s(m)=[s1, s2, …, sn]

[0122] Where s(m) is a vector, and the i-th element si in s(m) represents the association degree score between message m and the i-th semantic category, and n is the number of semantic categories.

[0123] Through the above steps, the feature vector F(m) can be extracted from each message m in the standard data set S(C(t)). These feature vectors can be used for subsequent data analysis and processing, such as sentiment analysis, data grading processing, etc.

[0124] S104. Through the priority calculation function G(), the priority of the feature vector F(m) is calculated to obtain the priority value p of message m, p = G(F(m)).

[0125] By calculating the priority of the feature vector F(m) through the priority calculation function G() and obtaining the priority value p of message m, a specific mathematical formula can be defined. This formula considers the importance of each feature in the feature vector F(m) and calculates a priority value p based on these features.

[0126] The priority calculation function G can be defined as:

[0127]

[0128] Where α, β, γ, δ, η, ζ, θ, κ, λ i , are adjustment coefficients, and LNi represents the information of the i-th language

[0129] S105. When the priority value p is less than the threshold p0, the reply suggestion information R1(p) corresponding to the message m is generated. When the priority value p is greater than or equal to the threshold p0, the automatic reply information R2(p) corresponding to the message m is generated.

[0130] The type of reply information to be generated can be determined based on the comparison result between the priority value p and the threshold p0. When p < p0, the reply suggestion information R1(p) is generated. When p ≥ p0, the automatic reply information R2(p) is generated.

[0131] When the priority value p of message m is less than the threshold p0, the system should generate a reply suggestion information R1(p). Define the reply suggestion information generation formula:

[0132] R1(p) = D(m, p)

[0133] Among them, p is the message parameter, including the context information, conversation history and user preferences corresponding to the message m, and D is the conversation management system, which generates reply suggestion information based on the message m and parameter p.

[0134] The dialog management system extracts key information and intent from the message m, understands the context of the current dialog based on the context information contained in the parameter p, and generates reply suggestions based on the results of semantic understanding and context modeling.

[0135] When the priority value p of message m is greater than or equal to the threshold p0, based on the content and context of the message, the system generates an automatic reply message R2(p). This message can be sent directly to the user without human intervention.

[0136] S106, using the interface display function U to process the reply suggestion information R1(p) and the automatic reply information R2(p), and display the processed result U(R1(p), R2(p)) on the preset interface.

[0137] An interface display function U may be defined, which can handle two types of reply information: reply suggestion information R1(p) and automatic reply information R2(p).

[0138] For the reply suggestion information R1(p), the interface display function U can be used to display R1(p) to the user. This display should contain the complete suggested reply content and provide the user with the option to modify it. Allow the user to modify the content in R1(p). The user can edit the suggested reply through a text box or other input method. After the user completes the modification, confirm the final reply content through a confirmation button or option. Once the user clicks the confirmation button, the interface display function U will capture the final message confirmed by the user and automatically send it to the target social media platform.

[0139] For generating automatic reply messages, the interface displays function U directly sending R2(p) to the target social media platform without user intervention. Once R2(p) is sent successfully, U will show the user a prompt that the message has been sent.

[0140] The solution of this application has the following beneficial effects:

[0141] 1) Multi-platform support: Able to collect interactive messages from multiple social media platforms in real time to ensure the comprehensiveness and real-time nature of the data.

[0142] 2) Data standardization and normalization: Extract key features of interactive messages with different data structures, unify data formats, standardize processing, and improve data consistency and processability.

[0143] 3) Improve management efficiency: Through intelligent processing and classification, automatic reply content generation, improved message processing response speed, timely reply to users, and improved efficiency of social media community management.

[0144] 4) Improve user experience: Identify the characteristics of interactive users through natural language algorithms and understand semantics to achieve personalized interaction and timely and effective responses, thereby improving user satisfaction and stickiness.

[0145] According to a specific implementation of the embodiment of the present disclosure, the real-time collection of interactive messages on multiple social media platforms within a time period t includes:

[0146] Determine a set P of social media platforms that contain all the data to be collected;

[0147] For each platform p∈P, define a collection function Cp(t), which returns all messages collected from platform p within time t;

[0148] Aggregate the data collected by all platforms to form an interactive message set C(t).

[0149] According to a specific implementation of the embodiment of the present disclosure, the interaction message set C(t) is standardized to obtain a standard data set S(C(t)), including:

[0150] Define n(v) as the value of field v contained in the interactive message set C(t), min(v) and max(v) as the minimum and maximum values ​​of v respectively, perform numerical conversion on the interactive message set C(t) to obtain a standardized numerical set :

[0151]

[0152] Define c(m) as the interactive message set The category field of message m, e c is the encoding vector of category c, the standardized numerical set Perform encoding conversion to obtain a set of encoding vectors :

[0153] ;

[0154] Define z(v) as the set of encoding vectors The Z score of field v in , μ(v) is the mean of v, σ(v) is the standard deviation of v, and the encoding vector set Perform anomaly detection and obtain the standard data set S(C(t)):

[0155]

[0156] Among them, τ is the set threshold.

[0157] According to a specific implementation of the embodiment of the present disclosure, the feature extraction is performed on each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m), including:

[0158] Set up an evaluation analysis model SF. The evaluation analysis model SF accepts a message m as input and outputs an evaluation polarity p and an evaluation strength s. The evaluation polarity p is a positive or negative number, and the evaluation strength s is a real number between 0 and 1. The message evaluation value EM(m) of the message m is:

[0159] EM(m)=p(m)×s(m).

[0160] According to a specific implementation of the embodiment of the present disclosure, the feature extraction is performed on each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m), and further includes:

[0161] Set up a semantic analysis model SY. The semantic analysis model SY accepts message m as input and outputs the score of one or more semantic categories. Each semantic category YC has a score sc, which indicates the degree to which message m is related to each semantic category. Define the semantic feature SE as:

[0162] SE(m)=s(m)=[s1,s2,…,sn]

[0163] Where s(m) is a vector, the i-th element si in s(m) represents the relevance score between message m and the i-th semantic category, and n is the number of semantic categories.

[0164] According to a specific implementation of the embodiment of the present disclosure, the priority calculation function G() is used to perform priority calculation on the feature vector F(m) to obtain the priority value p of the message m, including:

[0165] Define the priority calculation function G:

[0166]

[0167] Among them α, β, γ, δ, η, ζ, θ, κ, λ i , is the adjustment coefficient, and LNi represents the information of the i-th language.

[0168] According to a specific implementation of the embodiment of the present disclosure, when the priority value p is less than the threshold value p0, generating the reply suggestion information R1(p) corresponding to the message m includes:

[0169] Define the formula for generating reply suggestion information:

[0170] R1(p)=D(m,p)

[0171] Where p is a message parameter, including the context information, conversation history, and user preferences corresponding to message m. D is a conversation management system that generates reply suggestion information based on message m and parameter p.

[0172] The dialog management system extracts key information and intent from the message m, understands the context of the current dialog based on the context information contained in the parameter p, and generates reply suggestions based on the results of semantic understanding and context modeling.

[0173] According to a specific implementation of the embodiment of the present disclosure, the processing of the reply suggestion information R1(p) and the automatic reply information R2(p) by using the interface display function U includes:

[0174] The interface display function U displays the reply suggestion information R1(p) to the user, and accepts the user's modification and confirmation of the reply suggestion information R1(p), and after the user completes the modification and confirmation, automatically sends the user's confirmation message to the target social media;

[0175] After the interface display function U sends the automatic reply message R2(p) directly to the target social media, it displays a message sending completion prompt to the user.

[0176] Corresponding to the above method embodiment, see Figure 3 The embodiment of the present invention further discloses a multi-platform social media interactive message hierarchical processing device 30, comprising:

[0177] A collection module 301 collects interactive messages from multiple social media platforms in a time period t in real time to form an interactive message set C(t);

[0178] The processing module 302 performs standardization processing on the interactive message set C(t) to obtain a standard data set S(C(t));

[0179] The extraction module 303 performs feature extraction on each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], where the feature vector F(m) includes user gender GE, user age AG, number of followed users FL, regional information LC, language information LN, message evaluation value EM and message semantics SE;

[0180] The calculation module 304 performs priority calculation on the feature vector F(m) through the priority calculation function G() to obtain the priority value p of the message m, p=G(F(m));

[0181] The generating module 305 generates reply suggestion information R1(p) corresponding to the message m when the priority value p is less than the threshold value p0, and generates automatic reply information R2(p) corresponding to the message m when the priority value p is greater than or equal to the threshold value p0;

[0182] The display module 306 processes the reply suggestion information R1(p) and the automatic reply information R2(p) using the interface display function U, and displays the processed result U(R1(p), R2(p)) on a preset interface.

[0183] See also Figure 4 The embodiment of the present invention further provides an electronic device 60, the electronic device comprising:

[0184] at least one processor; and,

[0185] a memory communicatively connected to the at least one processor; wherein,

[0186] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the multi-platform social media interactive message hierarchical processing method in the aforementioned method embodiment.

[0187] An embodiment of the present invention further provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the aforementioned method embodiment.

[0188] An embodiment of the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the multi-platform social media interactive message hierarchical processing method in the aforementioned method embodiment.

[0189] Reference below Figure 4 , which shows a schematic diagram of the structure of an electronic device 60 suitable for implementing the embodiment of the present disclosure. The electronic device in the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0190] like Figure 4As shown, the electronic device 60 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 60 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0191] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, a gage, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 60 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 60 is shown with various devices, but it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have instead.

[0192] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0193] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A multi-platform social media interactive message hierarchical processing method, characterized in that: include: Collect interactive messages from multiple social media platforms in time period t in real time to form an interactive message set C(t); Standardizing the interactive message set C(t) to obtain a standard data set S(C(t)); For each message m∈S(C(t)) in the standard data set S(C(t)), feature extraction is performed to obtain a feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], where the feature vector F(m) includes user gender GE, user age AG, number of followed users FL, regional information LC, language information LN, message evaluation value EM and message semantics SE; The priority calculation function G() is used to calculate the priority of the feature vector F(m) to obtain the priority value p of the message m, p=G(F(m)); When the priority value p is less than the threshold value p0, the reply suggestion information R1(p) corresponding to the message m is generated; when the priority value p is greater than or equal to the threshold value p0, the automatic reply information R2(p) corresponding to the message m is generated; The reply suggestion information R1(p) and the automatic reply information R2(p) are processed using the interface display function U, and the processed result U(R1(p), R2(p)) is displayed on the preset interface; wherein, The process of processing the reply suggestion information R1(p) and the automatic reply information R2(p) by using the interface display function U includes: The interface display function U displays the reply suggestion information R1(p) to the user, and accepts the user's modification and confirmation of the reply suggestion information R1(p), and after the user completes the modification and confirmation, automatically sends the user's confirmation message to the target social media; After the interface display function U sends the automatic reply message R2(p) directly to the target social media, it displays a message sending completion prompt to the user.

2. The method according to claim 1, characterized in that The real-time collection of interactive messages on multiple social media platforms within a time period t includes: Determine a set P of social media platforms that contain all the data to be collected; For each platform p∈P, define a collection function Cp(t), which returns all messages collected from platform p within time t; Aggregate the data collected by all platforms to form an interactive message set C(t).

3. The method according to claim 2, characterized in that The interactive message set C(t) is subjected to standardization processing to obtain a standard data set S(C(t)), including: Define n(v) as the value of field v contained in the interactive message set C(t), min(v) and max(v) as the minimum and maximum values ​​of v respectively, perform numerical conversion on the interactive message set C(t) to obtain a standardized numerical set : ； Define c(m) as the interactive message set The category field of message m, e c is the encoding vector of category c, the standardized numerical set Perform encoding conversion to obtain a set of encoding vectors : ； Define z(v) as the set of encoding vectors The Z score of field v in , μ(v) is the mean of v, σ(v) is the standard deviation of v, and the encoding vector set Perform anomaly detection and obtain the standard data set S(C(t)): , where τ is the set threshold.

4. The method according to claim 3, characterized in that The feature extraction is performed for each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m), including: Set up an evaluation analysis model SF. The evaluation analysis model SF accepts a message m as input and outputs an evaluation polarity p and an evaluation strength s. The evaluation polarity p is a positive or negative number, and the evaluation strength s is a real number between 0 and 1. The message evaluation value EM(m) of the message m is: EM(m)=p(m)×s(m).

5. The method according to claim 4, characterized in that The method further comprises: extracting features from each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m); Set up a semantic analysis model SY. The semantic analysis model SY accepts message m as input and outputs the score of one or more semantic categories. Each semantic category YC has a score sc, which indicates the degree to which message m is related to each semantic category. Define the semantic feature SE as: SE(m)=s(m)=[s1,s2,…,sn] Where s(m) is a vector, the i-th element si in s(m) represents the relevance score between message m and the i-th semantic category, and n is the number of semantic categories.

6. The method according to claim 5, characterized in that The priority calculation function G() is used to perform priority calculation on the feature vector F(m) to obtain the priority value p of the message m, including: Define the priority calculation function G: , where α, β, γ, δ, η, ζ, θ, κ, λ i , is the adjustment coefficient, and LNi represents the information of the i-th language.

7. The method according to claim 6, characterized in that When the priority value p is less than the threshold value p0, the reply suggestion information R1(p) corresponding to the message m is generated, including: Define the formula for generating reply suggestion information: R1(p)=D(m,p) Where p is a message parameter, including the context information, conversation history, and user preferences corresponding to message m. D is a conversation management system that generates reply suggestion information based on message m and parameter p. The dialog management system extracts key information and intent from the message m, understands the context of the current dialog based on the context information contained in the parameter p, and generates reply suggestions based on the results of semantic understanding and context modeling.

8. A multi-platform social media interactive message hierarchical processing device, characterized in that: include: The collection module collects interactive messages from multiple social media platforms in a time period t in real time to form an interactive message set C(t); A processing module performs standardization processing on the interactive message set C(t) to obtain a standard data set S(C(t)); The extraction module extracts features from each message m∈S(C(t)) in the standard data set S(C(t)) to obtain a feature vector F(m)=[GE,AG,FL,LC,LN,EM,SE], where the feature vector F(m) includes user gender GE, user age AG, number of followed users FL, regional information LC, language information LN, message evaluation value EM and message semantics SE; A calculation module performs priority calculation on the feature vector F(m) through a priority calculation function G() to obtain a priority value p of the message m, p=G(F(m)); A generating module, when the priority value p is less than a threshold value p0, generates a reply suggestion information R1(p) corresponding to the message m, and when the priority value p is greater than or equal to the threshold value p0, generates an automatic reply information R2(p) corresponding to the message m; The display module processes the reply suggestion information R1(p) and the automatic reply information R2(p) using the interface display function U, and displays the processed result U(R1(p), R2(p)) on the preset interface; wherein, The process of processing the reply suggestion information R1(p) and the automatic reply information R2(p) by using the interface display function U includes: The interface display function U displays the reply suggestion information R1(p) to the user, and accepts the user's modification and confirmation of the reply suggestion information R1(p), and after the user completes the modification and confirmation, automatically sends the user's confirmation message to the target social media; After the interface display function U sends the automatic reply message R2(p) directly to the target social media, it displays a message sending completion prompt to the user.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the multi-platform social media interactive message hierarchical processing method described in any one of claims 1 to 7.

Images