Target feature extraction method and device based on social media data

Abstract

The embodiment of the invention provides a target feature extraction method and device based on social media data and electronic equipment, and belongs to the technical field of data processing, and the method comprises the steps: collecting social media data generated by a user in real time from n social media platforms M = {M1, M2,... Mn}; calculating features in the preprocessed data D through a feature calculation function Gf to obtain a feature data set Kt; calculating historical data on n social media platforms M = {M1, M2,... Mn} through a weight calculation function Gs to obtain a platform weight vector KM; performing drift feature calculation on historical data of the target feature on the n social media platforms through the drift calculation function Gp to obtain a drift weight vector KP; and calculating an evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP. According to the invention, the accuracy of target feature extraction is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a target feature extraction method, device and electronic equipment based on social media data. Background Art

[0002] Multi-platform social media data refers to data collected from multiple different social media platforms. This data can include user-generated content, user behavior data, interaction data, etc., and is usually used to analyze user behavior, market trends, brand reputation, content dissemination effects, etc. The collection and analysis of multi-platform social media data is very valuable for companies and researchers because it provides a more comprehensive perspective to understand the target audience and data change dynamics.

[0003] The existing technology has the following problems in the process of implementing social media data feature extraction:

[0004] 1. Single data source: Traditional target feature influence assessment mainly relies on market research and consumer surveys, lacks comprehensive analysis of social media data, and cannot reflect the actual performance of target features on social media.

[0005] 2. Limited analysis dimensions: Existing methods usually only focus on specific indicators (such as the number of target feature mentions or positive reviews), ignoring multi-dimensional factors such as user interaction, emotional tendencies, and social network communication.

[0006] 3. Poor real-time performance: Most target feature influence assessment methods have a long cycle and are difficult to reflect the dynamic changes of target feature influence in real time, and cannot provide timely and effective decision-making support for target feature management.

[0007] 4. Lack of automated processing: Existing evaluation methods mostly rely on manual processing and lack automated data collection, analysis and report generation mechanisms. They are inefficient and easily affected by subjective factors.

[0008] The above problems become technical problems that need to be solved. Summary of the invention

[0009] In view of this, embodiments of the present invention provide a method, device and electronic device for extracting target features based on social media data, which at least partially solve the problems existing in the prior art.

[0010] In a first aspect, an embodiment of the present invention provides a method for extracting target features based on social media data, comprising:

[0011] Collect user-generated social media data from n social media platforms M={M1,M2,…Mn} in real time, clean, format and perform sentiment analysis on the collected social media data to obtain preprocessed data D={D1,D2,…Dn};

[0012] The features in the preprocessed data D are calculated by the feature calculation function Gf to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, where Tn represents the number of occurrences of the target feature, Eq represents the emotional tendency, Ud represents the user interaction amount, and Tf represents the dissemination range;

[0013] The historical data on n social media platforms M={M1,M2,…Mn} are calculated through the weight calculation function Gs to obtain the platform weight vector KM=[KM1,KM2,…KMn];

[0014] The drift calculation function Gp is ​​used to calculate the drift characteristics of the historical data of the target feature on n social media platforms, and the drift weight vector KP=[KP1, KP2, …KPn] is obtained;

[0015] Based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP, the evaluation value of the target feature is calculated.

[0016] According to a specific implementation of an embodiment of the present invention, the social media data generated by users is collected in real time from n social media platforms M={M1, M2, ...Mn}, and the collected social media data is cleaned, formatted, and sentiment analyzed to obtain preprocessed data D, including:

[0017] Obtain API access to the social media platforms where data collection is required;

[0018] Define the keywords, hashtags, and account names for data collection to obtain social media data from the selected social media platforms.

[0019] According to a specific implementation of an embodiment of the present invention, the method of collecting user-generated social media data from n social media platforms M={M1, M2, ...Mn} in real time, cleaning, formatting and sentiment analysis the collected social media data to obtain preprocessed data D also includes:

[0020] Adjust the data from different sources in social media data to a consistent format and fill in default values ​​for fields with missing values;

[0021] Normalize the data attributes in social media to keep the value range consistent;

[0022] Set up a sentiment analysis algorithm to perform sentiment analysis on each text content in the social media data, obtain positive, negative or neutral sentiment scores, and form the final preprocessed data D.

[0023] According to a specific implementation of an embodiment of the present invention, the feature calculation function Gf is used to calculate the features in the preprocessed data D to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, including:

[0024] Calculate the total number of times Tn the target feature appears in the preprocessed data D:

[0025]

[0026] N is the total number of samples in the data set, di is the i-th sample, Is an indicator function, if the di sample contains the target feature, the value is 1, otherwise it is 0;

[0027] Setting the sentiment score function , which returns the calculated value representing the sentiment score Eq of a single sample di:

[0028]

[0029] Calculate user interaction volume Ud:

[0030]

[0031] represents the number of interactions of the i-th sample;

[0032] Calculate the propagation range Tf:

[0033]

[0034] Indicates the number of users covered by the i-th sample.

[0035] According to a specific implementation of an embodiment of the present invention, the historical data on n social media platforms M={M1, M2, ...Mn} is calculated by the weight calculation function Gs to obtain a platform weight vector KM=[KM1, KM2, ...KMn], including:

[0036] Set multiple evaluation indicators X for historical data, and the evaluation indicators X correspond to the evaluation values , i represents the i-th platform, j represents the j-th indicator of the evaluation indicator X;

[0037] Calculate the evaluation value The corresponding conversion value :

[0038]

[0039] and are the minimum and maximum values ​​of all platforms on the jth indicator respectively;

[0040] Assign a weight coefficient to each indicator , for each platform Mi, calculate its comprehensive score S i :

[0041]

[0042] The comprehensive score S of each platform i Normalize and get the platform weight KMi:

[0043] .

[0044] According to a specific implementation of an embodiment of the present invention, the drift feature calculation is performed on the historical data of the target feature on n social media platforms through the drift calculation function Gp to obtain a drift weight vector KP=[KP1, KP2, ...KPn], including:

[0045] Get the data at two time points t1 and t2, and for each feature F and each platform Mi, define the drift value for:

[0046]

[0047] represents the platform at time t The value of the upper feature F;

[0048] For each platform Mi, calculate its comprehensive drift score Pi:

[0049]

[0050] w F is the weight coefficient;

[0051] Quantify the comprehensive drift score Pi of each platform to obtain the drift weight KPi:

[0052] .

[0053] According to a specific implementation of an embodiment of the present invention, the calculation of the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP includes:

[0054] For each platform Mi, define weighted feature data :

[0055]

[0056] in is the value of the jth feature in the feature data set Kt on the platform Mi, is the number of days from the time of the feature data to the current time, α is the time decay factor, 0<α<1;

[0057] Set the feature interaction coefficient matrix I, which is a 4×4 matrix. Represents the interaction coefficient between feature p and feature q, weighted feature data for:

[0058] .

[0059] According to a specific implementation of an embodiment of the present invention, the calculation of the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP also includes:

[0060] For each platform Mi, calculate its comprehensive feature score Si:

[0061]

[0062] Add up the comprehensive feature scores of all platforms to get the evaluation value E of the target feature:

[0063] .

[0064] In a second aspect, an embodiment of the present invention provides a target feature extraction device based on social media data, comprising:

[0065] The collection module collects user-generated social media data from n social media platforms M={M1,M2,…Mn} in real time, cleans, formats and performs sentiment analysis on the collected social media data to obtain pre-processed data D={D1,D2,…Dn};

[0066] The feature module calculates the features in the preprocessed data D through the feature calculation function Gf to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, where Tn represents the number of occurrences of the target feature, Eq represents the emotional tendency, Ud represents the user interaction amount, and Tf represents the dissemination range;

[0067] The weight module calculates the historical data on n social media platforms M={M1,M2,…Mn} through the weight calculation function Gs to obtain the platform weight vector KM=[KM1,KM2,…KMn];

[0068] The drift module calculates the drift characteristics of the target feature on the historical data of n social media platforms through the drift calculation function Gp, and obtains the drift weight vector KP=[KP1, KP2, …KPn];

[0069] The calculation module calculates the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP.

[0070] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0071] at least one processor; and,

[0072] a memory communicatively connected to the at least one processor; wherein,

[0073] 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 target feature extraction method based on social media data in any of the aforementioned first aspects or any of the implementations of the first aspect.

[0074] 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 target feature extraction method based on social media data in the aforementioned first aspect or any implementation of the first aspect.

[0075] In a 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 target feature extraction method based on social media data in the aforementioned first aspect or any implementation of the first aspect.

[0076] The target feature extraction scheme based on social media data in the embodiment of the present invention includes: collecting social media data generated by users from n social media platforms M={M1, M2, ...Mn} in real time, cleaning, formatting and sentiment analysis of the collected social media data to obtain pre-processed data D={D1, D2, ...Dn}; calculating the features in the pre-processed data D by a feature calculation function Gf to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, where Tn represents the number of occurrences of the target feature, Eq represents the sentiment tendency, and Ud represents indicates the amount of user interaction, Tf indicates the scope of dissemination; the historical data on n social media platforms M={M1,M2,…Mn} are calculated by the weight calculation function Gs to obtain the platform weight vector KM=[KM1,KM2,…KMn]; the drift calculation function Gp is ​​used to calculate the drift characteristics of the historical data of the target feature on n social media platforms to obtain the drift weight vector KP=[KP1,KP2,…KPn]; based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP, the evaluation value of the target feature is calculated. The present invention has the following beneficial effects:

[0077] aMulti-dimensional analysis: This algorithm combines multi-dimensional indicators such as the number of mentions of target features (e.g., brand features), sentiment tendency, user interaction, and dissemination range, and provides a comprehensive assessment of the impact of target features, which can more accurately reflect the actual performance of target features on social media.

[0078] b. Strong real-time performance: Through real-time data collection and processing, the algorithm can dynamically reflect changes in the influence of target features, provide users with timely decision support, and help target features maintain agility in competition.

[0079] cHigh accuracy: Through optimized weight distribution and machine learning algorithms, the algorithm can make fine adjustments based on the industry characteristics and market performance of the target feature to ensure the accuracy and reliability of the target feature's influence estimate.

[0080] d. Automated processing: This algorithm realizes the automation of the entire process from data collection, processing, analysis to report generation, greatly improving work efficiency, reducing errors caused by manual intervention, and adapting to the processing needs of large-scale data.

[0081] eVisualization: Through intuitive visualization tools, it helps users quickly understand the analysis results, improves the scientificity and precision of target feature management, and supports customized report output to meet the needs of users at different levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] 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.

[0083] Figure 1 A schematic diagram of a target feature extraction method based on social media data provided by an embodiment of the present invention;

[0084] Figure 2 A schematic flow chart of another target feature extraction method based on social media data provided by an embodiment of the present invention;

[0085] Figure 3 A schematic diagram of the structure of a target feature extraction device based on social media data provided by an embodiment of the present invention;

[0086] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0087] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0088] 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.

[0089] 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.

[0090] 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.

[0091] 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.

[0092] The disclosed embodiment provides a target feature extraction method based on social media data. The target feature extraction method based on social media data 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.

[0093] See also Figure 1 and Figure 2 The present disclosure provides a method for extracting target features based on social media data, including:

[0094] S101, collect user-generated social media data from n social media platforms M={M1,M2,…Mn} in real time, clean, format and perform sentiment analysis on the collected social media data to obtain preprocessed data D={D1,D2,…Dn}.

[0095] You can use the API interface to obtain user-generated content and interaction data related to target features (e.g., brand features) from multiple social media platforms. The specific steps are as follows:

[0096] a Keyword setting:

[0097] Set keywords based on target feature names, product names, related topics, etc. to ensure comprehensive and targeted data collection. You can use Boolean logic to combine different keywords to cover possible target feature mentions.

[0098] b Platform access:

[0099] Connect data through the official API interface of the social media platform, set the frequency and scope of data collection, and ensure real-time data inflow. For social platforms that do not provide APIs, you may need to crawl data through web crawler technology and follow the platform's terms of use.

[0100] c Data storage:

[0101] The collected raw data is stored in a NoSQL database (such as MongoDB) to support efficient storage and fast access of unstructured data, preparing for subsequent processing and analysis.

[0102] Clean, format and perform sentiment analysis on the collected social media data to extract key information related to the influence of the target feature. The specific steps are as follows:

[0103] aData cleaning:

[0104] Use data cleaning tools such as OpenRefine to remove noisy data, duplicate data, and invalid data, such as robot-generated content or advertising spam. Use regular expressions or machine learning models such as BERT to automatically tag and filter low-quality content.

[0105] bData formatting:

[0106] Unify data formats across different platforms, such as converting timestamps to a unified UTC format, handling encoding differences between texts in different languages, and unifying the format of user identifiers. Streaming and formatting of data is achieved through data pipelines such as Apache Kafka.

[0107] cSentiment Analysis:

[0108] Use natural language processing (NLP) technology to perform sentiment analysis on user-generated content. You can use a pre-trained sentiment classification model (such as a model using the Transformer architecture) to classify the text into positive, negative, and neutral sentiments, and fine-tune the results in conjunction with a sentiment lexicon (such as VADER).

[0109] S102, calculate the features in the preprocessed data D through the feature calculation function Gf to obtain the feature data set Kt={Tn,Eq,Ud,Tf}, where Tn represents the number of occurrences of the target feature, Eq represents the emotional tendency, Ud represents the user interaction amount, and Tf represents the dissemination range.

[0110] The preprocessed data D is a data set, the target feature can be a keyword, a phrase, or other interesting patterns, and Tn represents the total number of times the target feature appears in the data.

[0111]

[0112] Where N is the total number of samples in the dataset, di is the i-th sample, and Is an indicator function, which takes value 1 if the condition is true; otherwise it takes value 0.

[0113] Sentiment tendency is usually calculated through sentiment analysis, which can be based on dictionary methods or machine learning models. Suppose there is a sentiment score function , which returns a numeric value representing the sentiment score of a single sample (e.g., between -1 and +1, with negative numbers representing negative sentiment and positive numbers representing positive sentiment).

[0114]

[0115] User interaction can refer to the number of likes, comments, shares, and other indicators of user interaction with content. Represents the number of interactions of the i-th sample.

[0116]

[0117] Reach refers to how many different users have seen or engaged with the content. represents the number of unique users covered by the i-th sample.

[0118]

[0119] If we consider the total number of unique users instead of the number of users per interaction, then:

[0120]

[0121] where Ri is the set of all unique users associated with the i-th sample, and |.| represents the cardinality of the set, i.e., the number of elements in the set.

[0122] S103, calculate the historical data on n social media platforms M={M1,M2,…Mn} through the weight calculation function Gs to obtain the platform weight vector KM=[KM1,KM2,…KMn].

[0123] The weight calculation function Gs is used to determine the weight of each platform based on the historical data of the social media platform, and finally form a platform weight vector KM. These weights can be based on a variety of factors, such as user activity, content dissemination efficiency, influence, etc.

[0124] Each social media platform Mi can be evaluated based on multiple indicators, such as: user activity (such as the number of daily active users), content dissemination speed, average interaction rate (likes, comments, shares, etc.), user growth rate, brand or content exposure, and user satisfaction score.

[0125] In order to ensure the comparability between different indicators, each indicator needs to be standardized. Suppose for an indicator X, its value is , where i represents the i-th platform and j represents the j-th indicator.

[0126] The normalization formula can be:

[0127]

[0128] in are the minimum and maximum values ​​of the j-th indicator for all platforms respectively.

[0129] Assign a weight coefficient w to each indicator j , for each platform Mi, calculate its comprehensive score Si:

[0130]

[0131] Normalize the comprehensive score Si of each platform to get the platform weight KMi:

[0132] .

[0133] S104, performing drift feature calculation on the historical data of the target feature on n social media platforms through the drift calculation function Gp to obtain a drift weight vector KP=[KP1, KP2, ...KPn].

[0134] Data at two time points t1 and t2 can be obtained, and for each feature F and each platform Mi, the drift value is defined for:

[0135]

[0136] represents the platform at time t The value of the upper feature F;

[0137] For each platform Mi, calculate its comprehensive drift score Pi:

[0138]

[0139] w F is the weight coefficient;

[0140] Quantify the comprehensive drift score Pi of each platform to obtain the drift weight KPi:

[0141] .

[0142] S105, calculating the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP.

[0143] For each platform Mi, define weighted feature data as follows:

[0144]

[0145] in is the value of the jth feature in the feature data set Kt on the platform Mi, is the number of days from the current time to the feature data, and α is the time decay factor (0<α<1).

[0146] Set the feature interaction coefficient matrix I to a 4×4 matrix, where represents the interaction coefficient between feature p and feature q. Then the weighted feature data for:

[0147]

[0148] For each platform Mi, calculate its comprehensive feature score Si:

[0149]

[0150] Add up the comprehensive feature scores of all platforms to get the evaluation value E of the target feature:

[0151]

[0152] The present invention has the following beneficial effects:

[0153] aMulti-dimensional analysis: This algorithm combines multi-dimensional indicators such as the number of mentions of target features (e.g., brand features), sentiment tendency, user interaction, and dissemination range, and provides a comprehensive assessment of the impact of target features, which can more accurately reflect the actual performance of target features on social media.

[0154] b. Strong real-time performance: Through real-time data collection and processing, the algorithm can dynamically reflect changes in the influence of target features, provide users with timely decision support, and help target features maintain agility in competition.

[0155] cHigh accuracy: Through optimized weight distribution and machine learning algorithms, the algorithm can make fine adjustments based on the industry characteristics and market performance of the target feature to ensure the accuracy and reliability of the target feature's influence estimate.

[0156] d. Automated processing: This algorithm realizes the automation of the entire process from data collection, processing, analysis to report generation, greatly improving work efficiency, reducing errors caused by manual intervention, and adapting to the processing needs of large-scale data.

[0157] eVisualization: Through intuitive visualization tools, it helps users quickly understand the analysis results, improves the scientificity and precision of target feature management, and supports customized report output to meet the needs of users at different levels.

[0158] According to a specific implementation of an embodiment of the present invention, the social media data generated by users is collected in real time from n social media platforms M={M1, M2, ...Mn}, and the collected social media data is cleaned, formatted, and sentiment analyzed to obtain preprocessed data D, including:

[0159] Obtain API access to the social media platforms where data collection is required;

[0160] Define the keywords, hashtags, and account names for data collection to obtain social media data from the selected social media platforms.

[0161] According to a specific implementation of an embodiment of the present invention, the method of collecting user-generated social media data from n social media platforms M={M1, M2, ...Mn} in real time, cleaning, formatting and sentiment analysis the collected social media data to obtain preprocessed data D also includes:

[0162] Adjust the data from different sources in social media data to a consistent format and fill in default values ​​for fields with missing values;

[0163] Normalize the data attributes in social media to keep the value range consistent;

[0164] Set up a sentiment analysis algorithm to perform sentiment analysis on each text content in the social media data, obtain positive, negative or neutral sentiment scores, and form the final preprocessed data D.

[0165] According to a specific implementation of an embodiment of the present invention, the feature calculation function Gf is used to calculate the features in the preprocessed data D to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, including:

[0166] Calculate the total number of times Tn the target feature appears in the preprocessed data D:

[0167]

[0168] N is the total number of samples in the data set, di is the i-th sample, Is an indicator function, if the di sample contains the target feature, the value is 1, otherwise it is 0;

[0169] Setting the sentiment score function , which returns the calculated value representing the sentiment score Eq of a single sample di:

[0170]

[0171] Calculate user interaction volume Ud:

[0172]

[0173] represents the number of interactions of the i-th sample;

[0174] Calculate the propagation range Tf:

[0175]

[0176] Indicates the number of users covered by the i-th sample.

[0177] According to a specific implementation of an embodiment of the present invention, the historical data on n social media platforms M={M1, M2, ...Mn} is calculated by the weight calculation function Gs to obtain a platform weight vector KM=[KM1, KM2, ...KMn], including:

[0178] Set multiple evaluation indicators X for historical data, and the evaluation indicators X correspond to the evaluation values , i represents the i-th platform, j represents the j-th indicator of the evaluation indicator X;

[0179] Calculate the evaluation value The corresponding conversion value :

[0180]

[0181] and are the minimum and maximum values ​​of all platforms on the jth indicator respectively;

[0182] Assign a weight coefficient to each indicator , for each platform Mi, calculate its comprehensive score S i :

[0183]

[0184] The comprehensive score S of each platform i Normalize and get the platform weight KMi:

[0185] .

[0186] According to a specific implementation of an embodiment of the present invention, the drift feature calculation is performed on the historical data of the target feature on n social media platforms through the drift calculation function Gp to obtain a drift weight vector KP=[KP1, KP2, ...KPn], including:

[0187] Get the data at two time points t1 and t2, and for each feature F and each platform Mi, define the drift value for:

[0188]

[0189] represents the platform at time t The value of the upper feature F;

[0190] For each platform Mi, calculate its comprehensive drift score Pi:

[0191]

[0192] w F is the weight coefficient;

[0193] Quantify the comprehensive drift score Pi of each platform to obtain the drift weight KPi:

[0194] .

[0195] According to a specific implementation of an embodiment of the present invention, the calculation of the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP includes:

[0196] For each platform Mi, define weighted feature data :

[0197]

[0198] in is the value of the jth feature in the feature data set Kt on the platform Mi, is the number of days from the time of the feature data to the current time, α is the time decay factor, 0<α<1;

[0199] Set the feature interaction coefficient matrix I, which is a 4×4 matrix. Represents the interaction coefficient between feature p and feature q, weighted feature data for:

[0200] .

[0201] According to a specific implementation of an embodiment of the present invention, the calculation of the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP also includes:

[0202] For each platform Mi, calculate its comprehensive feature score Si:

[0203]

[0204] Add up the comprehensive feature scores of all platforms to get the evaluation value E of the target feature:

[0205] .

[0206] Corresponding to the above method embodiment, see Figure 3 The embodiment of the present invention further discloses a target feature extraction device 30 based on social media data, comprising:

[0207] A collection module 301 collects user-generated social media data from n social media platforms M={M1, M2, ...Mn} in real time, cleans, formats and performs sentiment analysis on the collected social media data to obtain pre-processed data D={D1, D2, ...Dn};

[0208] The feature module 302 calculates the features in the preprocessed data D through the feature calculation function Gf to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, where Tn represents the number of occurrences of the target feature, Eq represents the emotional tendency, Ud represents the user interaction amount, and Tf represents the spread range;

[0209] The weight module 303 calculates the historical data on n social media platforms M={M1,M2,…Mn} through the weight calculation function Gs to obtain the platform weight vector KM=[KM1,KM2,…KMn];

[0210] The drift module 304 calculates the drift feature of the target feature on the historical data of n social media platforms through the drift calculation function Gp to obtain a drift weight vector KP=[KP1, KP2, ...KPn];

[0211] The calculation module 305 calculates the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP.

[0212] See also Figure 4 The embodiment of the present invention further provides an electronic device 60, the electronic device comprising:

[0213] at least one processor; and,

[0214] a memory communicatively connected to the at least one processor; wherein,

[0215] 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 target feature extraction method based on social media data in the aforementioned method embodiment.

[0216] 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.

[0217] 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 target feature extraction method based on social media data in the aforementioned method embodiment.

[0218] 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.

[0219] 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.

[0220] 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.

[0221] 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.

[0222] 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 target feature extraction method based on social media data, characterized in that: include: Collect user-generated social media data from n social media platforms M={M1,M2,…Mn} in real time, clean, format and perform sentiment analysis on the collected social media data to obtain preprocessed data D={D1,D2,…Dn}; The features in the preprocessed data D are calculated by the feature calculation function Gf to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, where Tn represents the number of occurrences of the target feature, Eq represents the emotional tendency, Ud represents the user interaction amount, and Tf represents the dissemination range; The historical data on n social media platforms M={M1,M2,…Mn} are calculated through the weight calculation function Gs to obtain the platform weight vector KM=[KM1,KM2,…KMn]; The drift calculation function Gp is ​​used to calculate the drift characteristics of the historical data of the target feature on n social media platforms, and the drift weight vector KP=[KP1, KP2, …KPn] is obtained; Based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP, the evaluation value of the target feature is calculated.

2. The method according to claim 1, characterized in that: The method collects user-generated social media data from n social media platforms M={M1, M2, …Mn} in real time, cleans, formats and performs sentiment analysis on the collected social media data to obtain preprocessed data D, including: Obtain API access to the social media platforms where data collection is required; Define the keywords, hashtags, and account names for data collection to obtain social media data from the selected social media platforms.

3. The method according to claim 2, characterized in that The method collects user-generated social media data from n social media platforms M={M1, M2, ...Mn} in real time, cleans, formats and performs sentiment analysis on the collected social media data to obtain preprocessed data D, and further includes: Adjust the data from different sources in social media data to a consistent format and fill in default values ​​for fields with missing values; Normalize the data attributes in social media to keep the value range consistent; Set up a sentiment analysis algorithm to perform sentiment analysis on each text content in the social media data, obtain positive, negative or neutral sentiment scores, and form the final preprocessed data D.

4. The method according to claim 3, characterized in that The feature calculation function Gf is used to calculate the features in the preprocessed data D to obtain a feature data set Kt={Tn,Eq,Ud,Tf}, including: Calculate the total number of times Tn the target feature appears in the preprocessed data D: N is the total number of samples in the data set, di is the i-th sample, Is an indicator function, if the di sample contains the target feature, the value is 1, otherwise it is 0; Setting the sentiment score function , which returns the calculated value representing the sentiment score Eq of a single sample di: Calculate user interaction volume Ud: represents the number of interactions of the i-th sample; Calculate the propagation range Tf: Indicates the number of users covered by the i-th sample.

5. The method according to claim 4, characterized in that The weight calculation function Gs is used to calculate the historical data on n social media platforms M={M1,M2,…Mn} to obtain the platform weight vector KM=[KM1,KM2,…KMn], including: Set multiple evaluation indicators X for historical data, and the evaluation indicators X correspond to the evaluation values , i represents the i-th platform, j represents the j-th indicator of the evaluation indicator X; Calculate the evaluation value The corresponding conversion value : and are the minimum and maximum values ​​of all platforms on the jth indicator respectively; Assign a weight coefficient to each indicator , for each platform Mi, calculate its comprehensive score S i : The comprehensive score S of each platform i Normalize and get the platform weight KMi: 。 6. The method according to claim 5, characterized in that The drift calculation function Gp is ​​used to calculate the drift feature of the historical data of the target feature on n social media platforms to obtain a drift weight vector KP=[KP1, KP2, ...KPn], including: Get the data at two time points t1 and t2, and for each feature F and each platform Mi, define the drift value for: represents the platform at time t The value of the upper feature F; For each platform Mi, calculate its comprehensive drift score Pi: w F is the weight coefficient; Quantify the comprehensive drift score Pi of each platform to obtain the drift weight KPi: 。 7. The method according to claim 6, characterized in that The step of calculating the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP includes: For each platform Mi, define weighted feature data : in is the value of the jth feature in the feature data set Kt on the platform Mi, is the number of days from the time of the feature data to the current time, α is the time decay factor, 0<α<1; Set the feature interaction coefficient matrix I, which is a 4×4 matrix. Represents the interaction coefficient between feature p and feature q, weighted feature data for: 。 8. The method according to claim 7, characterized in that The step of calculating the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP further includes: For each platform Mi, calculate its comprehensive feature score Si: Add up the comprehensive feature scores of all platforms to get the evaluation value E of the target feature: 。 9. A target feature extraction device based on social media data, characterized in that: include: The collection module collects user-generated social media data from n social media platforms M={M1,M2,…Mn} in real time, cleans, formats and performs sentiment analysis on the collected social media data to obtain pre-processed data D={D1,D2,…Dn}; The feature module calculates the features in the preprocessed data D through the feature calculation function Gf to obtain a feature data set Kt={Tn, Eq, Ud, Tf}, where Tn represents the number of occurrences of the target feature, Eq represents the emotional tendency, Ud represents the user interaction amount, and Tf represents the dissemination range; The weight module calculates the historical data on n social media platforms M={M1,M2,…Mn} through the weight calculation function Gs to obtain the platform weight vector KM=[KM1,KM2,…KMn]; The drift module calculates the drift characteristics of the target feature on the historical data of n social media platforms through the drift calculation function Gp, and obtains the drift weight vector KP=[KP1, KP2, …KPn]; The calculation module calculates the evaluation value of the target feature based on the feature data set Kt, the platform weight vector KM and the drift weight vector KP.

10. 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 target feature extraction method based on social media data as described in any one of claims 1 to 8.

Images