Data processing method and device, electronic equipment and storage medium

Abstract

The invention provides a data processing method and device, electronic equipment and a storage medium, and the data processing method comprises the steps: obtaining at least one target text associated with at least one target event in a first time period, extracting target feature data of at least two dimensions from each target text, the dimensions are used for representing different attributes contained in the target text; based on the target feature data of each target text, first influence indexes of the target events are determined through a preset rule, the first influence indexes are mapped to the same value interval, and the first influence indexes are used for representing the influence of the target events; and based on the mapped first influence index in the first time period, predicting the change degree of the first influence index of each target event in a second time period, the second time period being later than the first time period.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a data processing method, device, electronic device and storage medium. Background Art

[0002] In the current information age, the impact assessment of public events has become extremely important. However, the current product index of public platforms lacks customized solutions for the analysis of the dissemination index of specific events, and cannot predict the changes in influence. At the same time, the index of social platform products often focuses on the data source and dimensions of the product, which limits the development of cross-platform, multi-dimensional comprehensive evaluation capabilities. This makes it difficult to quickly and comprehensively compare the dissemination of various events across the entire network. Summary of the invention

[0003] In view of the above problems, the present disclosure provides a data processing method, device, electronic device and storage medium.

[0004] According to a first aspect of the present disclosure, a data processing method is provided, including: obtaining at least one target text associated with at least one target event within a first time period, extracting target feature data of at least two dimensions from each target text, the dimensions being used to characterize different attributes contained in the target text; based on the target feature data of each target text, determining a first influence index of each target event by a preset rule, and mapping each first influence index to the same value range, the first influence index being used to characterize the influence of the target event; based on the first influence index within the first time period after mapping, predicting the degree of change of the first influence index of each target event in a second time period, the second time period being later than the first time period.

[0005] According to an embodiment of the present disclosure, at least one target text associated with at least one target event within a first time period is obtained, and target feature data of at least two dimensions are extracted from each target text, including: extracting multiple target phrases in each target text; and classifying each target phrase based on at least two predefined dimensions to obtain at least one target feature data for each dimension.

[0006] According to an embodiment of the present disclosure, based on the target feature data of each target text, determining the first influence index of each target event through preset rules includes: based on the target feature data of the target text in each dimension, calculating the second influence index of the target text in each dimension through preset rules, the second influence index is used to characterize the influence of the target text in a certain dimension; aggregating the second influence indexes of all dimensions of the target text to obtain the third influence index of the target text, the third influence index is used to characterize the influence of the target text; aggregating the third influence index of each target text to obtain the first influence index of the target event.

[0007] According to an embodiment of the present disclosure, a dimension includes at least a degree of relevance between a target text and a target object, and based on the target feature data of the target text in each dimension, a second influence index of the target text in each dimension is calculated by preset rules, including: obtaining target feature data with different weights in the relevance dimension based on at least the mention position, number of mentions and mention depth of the target object in the target text; calculating the mention rate of the target object in the target text by using preset rules based on the target feature data with different weights in the relevance dimension, normalizing the mention rate, and obtaining the second influence index in the relevance dimension.

[0008] According to an embodiment of the present disclosure, the dimension includes at least a target text sentiment dimension, and based on the target feature data of the target text in each dimension, the second influence index of the target text in each dimension is calculated by preset rules, including: based on the target feature data of the target text sentiment dimension, using a sentiment classification model to obtain a probability value of the sentiment tendency of the target text; normalizing the probability value to obtain the second influence index in the target text sentiment dimension.

[0009] According to an embodiment of the present disclosure, a dimension includes at least a target text characteristic value, and based on the target feature data of the target text in each dimension, calculating the second influence index of the target text in each dimension by preset rules includes: obtaining a target identifier of the target text, the target identifier is at least used to characterize the dissemination medium and the creator of the target text; based on the target identifier, performing differentiated processing on the target feature data of the target text characteristic value dimension by using matching preset rules to obtain the second influence index of the target text in the target text characteristic value dimension.

[0010] According to an embodiment of the present disclosure, the method also includes: monitoring the target feature data of the target dimension in each dimension; when the value of the target feature data of the target dimension is greater than or equal to a preset threshold, obtaining a weighting factor of the target event, the weighting factor being used to characterize the degree of influence of the target feature data of the target dimension on the target event; based on the weighting factor, updating the first influence index.

[0011] The second aspect of the present disclosure provides a data processing device, including: an extraction module, configured to obtain at least one target text associated with at least one target event within a first time period, and extract target feature data of at least two dimensions from each target text, where the dimensions are used to characterize different attributes contained in the target text; an evaluation module, configured to determine the first influence index of each target event based on the target feature data of each target text by preset rules, and map each first influence index to the same value range, where the first influence index is used to characterize the influence of the target event; a determination module, configured to predict the degree of change of the first influence index of each target event in a second time period based on the first influence index in the first time period after mapping, where the second time period is later than the first time period.

[0012] A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute to obtain at least one target text associated with at least one target event within a first time period, extract target feature data of at least two dimensions from each target text, and the dimensions are used to characterize different attributes contained in the target text; based on the target feature data of each target text, determine the first influence index of each target event through a preset rule, and map each first influence index to the same value range, and the first influence index is used to characterize the influence of the target event; based on the first influence index in the first time period after mapping, predict the degree of change of the first influence index of each target event in a second time period, and the second time period is later than the first time period.

[0013] A fourth aspect of the present disclosure provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the following steps: obtaining at least one target text associated with at least one target event within a first time period, extracting target feature data of at least two dimensions from each target text, wherein the dimensions are used to characterize different attributes contained in the target text; determining a first influence index of each target event based on the target feature data of each target text by a preset rule, and mapping each first influence index to the same value range, wherein the first influence index is used to characterize the influence of the target event; and predicting the degree of change of the first influence index of each target event in a second time period based on the first influence index in the first time period after mapping, wherein the second time period is later than the first time period.

[0014] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0016] Figure 1 A diagram schematically shows an application scenario of a data processing method and a processing device according to an embodiment of the present disclosure;

[0017] Figure 2 A flowchart schematically shows a data processing method according to an embodiment of the present disclosure;

[0018] Figure 3 A schematic diagram schematically shows a change degree of a first impact index of a target event in a second time period according to an embodiment of the present disclosure;

[0019] Figure 4A block diagram schematically shows a data processing device according to an embodiment of the present disclosure; and

[0020] Figure 5 A schematic block diagram of an electronic device according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0021] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0022] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0023] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0024] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0025] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0026] The disclosed embodiment discloses a data processing method, including: obtaining at least one target text associated with at least one target event within a first time period, extracting target feature data of at least two dimensions from each target text, the dimensions being used to characterize different attributes contained in the target text; based on the target feature data of each target text, determining the first influence index of each target event by a preset rule, and mapping each first influence index to the same value interval, the first influence index being used to characterize the influence of the target event; based on the first influence index within the first time period after mapping, predicting the degree of change of the first influence index of each target event within a second time period, the second time period being later than the first time period. By comprehensively considering the multi-dimensional conditions of the target texts related to the public event (target event), integrating the multi-dimensional target feature data, and using preset rules to quantify the influence of the target event, a comparable first influence index is formed, thereby making the assessment of the influence more comprehensive and accurate.

[0027] Figure 1 An application scenario diagram 100 of a data processing method and a processing device according to an embodiment of the present disclosure is schematically shown.

[0028] It should be noted that Figure 1 The examples shown are merely scenarios in which the embodiments of the present disclosure can be applied, to help those skilled in the art understand the technical content of the present disclosure, but do not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0029] like Figure 1 As shown, it shows a schematic diagram of an application program running environment 100 provided by an embodiment of the present disclosure. The application program running environment may include: terminal devices 101, 102, 103, a network 104 and a server 105. The terminal devices 101, 102, 103 include but are not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle terminals, game consoles, e-book readers, multimedia playback devices, wearable devices and other electronic devices. The client of the application program can be installed in the terminal devices 101, 102, 103.

[0030] In the disclosed embodiment, the above-mentioned application can be any application that can provide target feature data of public events. Typically, the application can be a social media application. Of course, in addition to social media applications, other types of applications can also collect and provide target feature data of public events. For example, news and information applications, announcement platforms, forum community applications, corporate public relations management systems, market intelligence analysis software, online comment aggregation tools, etc., which are not specifically limited in the disclosed embodiment. Optionally, the client of the above-mentioned application is running in the terminal devices 101, 102, and 103, which is used to collect, process and display the target feature data of public events in real time.

[0031] The server 105 is used to provide background services for the clients of the applications in the terminal devices 101, 102, and 103. For example, the server 105 can be the background server of the above-mentioned application. The server 105 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Optionally, the server 105 provides background services for the applications in multiple terminal devices 101, 102, and 103 at the same time.

[0032] Optionally, the terminal devices 101, 102, 103 and the server 105 can communicate with each other via the network 104. The terminal devices 101, 102, 103 and the server 105 can be directly or indirectly connected via wired or wireless communication, which is not limited in the present disclosure.

[0033] It should be noted that the data processing method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the data processing device provided in the embodiment of the present disclosure can generally be set in the server 105. The data processing method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the target device 101, 102, 103 and / or the server 105. Accordingly, the data processing device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the target device 101, 102, 103 and / or the server 105.

[0034] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0035] The following will be based on Figure 1 The scene described by Figure 2~Figure 3 The data processing method of the embodiment of the present disclosure is described in detail.

[0036] Figure 2 The flowchart of the data processing method according to the embodiment of the present disclosure is schematically shown.

[0037] like Figure 2As shown, it shows a flow chart of the data processing method provided by the embodiment of the present disclosure. The method can be applied to a computer device, which refers to an electronic device with data calculation and processing capabilities. For example, the execution subject of each step can be Figure 1 The server 105 in the application running environment shown. The method includes S210-S230.

[0038] In S210, at least one target text associated with at least one target event in a first time period is obtained, and target feature data of at least two dimensions are extracted from each target text, where the dimensions are used to characterize different attributes contained in the target text.

[0039] The first time period is the time range in which the target text can be collected. It can be set according to specific needs, such as the past week, month, or a specific date interval. In addition, the target event can be any event that attracts attention or needs to be analyzed, such as product launches, market fluctuations, and social hot spots. Generally speaking, the spread of articles related to the target event (i.e., target text) constitutes and promotes the spread of the entire event. Therefore, text information associated with these target events is collected through channels such as search engines, social media platforms, and news reports. Then, at least two dimensions of target feature data are extracted from the text information (target text). These dimensions can be relevance, sentiment, author information, and interactive data (such as likes, comments), etc. For example, the target feature data of the relevance dimension can be the number of keywords, etc., to measure the relevance of the text to the target object.

[0040] For example, suppose you need to analyze the release event of a new mobile phone in the past week. First, set the first time period to the past week (for example, from last Monday to last Sunday). Then, collect text information such as news reports, user comments, and social media posts related to the release event of this new mobile phone through search engines and social media platforms. From the collected target text, extract feature data of the relevance dimension, that is, the number of keywords (such as "mobile phone name"), and positive comments (such as "this phone is great!"), negative comments (such as "the price is too expensive!") or neutral comments (such as "the function introduction of this phone is very detailed").

[0041] In S220, based on the target feature data of each target text, a first influence index of each target event is determined by a preset rule, and each first influence index is mapped to the same value range. The first influence index is used to characterize the influence of the target event.

[0042] Apply preset rules to deeply process the target feature data of different dimensions extracted from each target text, and calculate the first impact index of the target event. Among them, the preset rules are mainly formulated according to business needs and the characteristics of the target event. When processing feature data of different dimensions, the preset rules also have corresponding differences. At the same time, in order to facilitate comparison and analysis, the calculated first impact index is mapped to a unified value range (such as 0-100), so that the influence of different target events can be compared on the same scale.

[0043] For example, after extracting the feature data of the relevance dimension and the sentiment dimension of the target text, the ratio of the number of "mobile phone names" to the number of mobile phone names of other brands in the text is calculated. At the same time, according to the evaluation of the target text, the sentiment tendency is judged and the corresponding probability is calculated, and then the calculation results of the relevance dimension and the sentiment dimension are normalized and weighted summed, and finally mapped to a unified value interval (such as 0-100) to obtain the first impact index of the target event when the text is associated.

[0044] It should be noted that when a target event is associated with multiple target texts at the same time, for the first influence index of the target event, the influence of all target texts needs to be counted to obtain the overall influence score of the event.

[0045] At S230 , based on the mapped first impact index in the first time period, a change degree of the first impact index of each target event in a second time period is predicted, where the second time period is later than the first time period.

[0046] After the target event occurs, as time goes by, the event continues to ferment and its influence continues to increase. Therefore, based on the mapped first impact index, we can infer the possible changes in the first impact index of each target event in the second time period (which is later than the first time period and usually represents the stage when the event further ferments and its influence expands).

[0047] In some exemplary embodiments, an autoregressive integrated moving average (ARIMA) time series model is used to predict the degree of change of the first impact index of the target event in the second time period, that is, the rate of change and the growth amount of the impact index, so as to obtain the diffusion trend of the event in the future time period.

[0048] The general form of ARIMA is:

[0049]

[0050] in, represents the slope value at time point t, p represents the autoregressive order, d represents the difference order, q represents the moving average order, represents the autoregressive coefficient, represents the moving average coefficient, B represents the lag operator, Represents the error term at time point t, which is usually assumed to be white noise.

[0051] It should be noted that the rate of change of the influence score (first influence index) reflects the fermentation and diffusion speed of the event, until the final slope approaches 0, indicating that the event propagation is nearing its end. The score at this time can be used as the score for evaluating the final influence of the event. When the predicted slope continues to increase in the next time period, it means that the event is still in a rapid fermentation stage and needs to be paid special attention to. When the predicted slope becomes smaller, it means that the fermentation speed of the event has slowed down. Similarly, when the growth rate continues to remain at a high level, it means that the influence of the event is still steadily increasing. When the growth rate begins to decline significantly, it may mean that the influence of the event is close to saturation, or that the attention to the event has declined significantly.

[0052] Figure 3 A schematic diagram schematically shows the degree of change of the first impact index of a target event in a second time period according to an embodiment of the present disclosure.

[0053] like Figure 3 As shown in the figure, the influence score shows an upward trend between the predicted 0 and 12 days. As time goes by, the curve tends to stabilize around the twelfth day, indicating that the influence of the event has basically stabilized or ended. On the other hand, the change rate curve has the largest slope at the beginning of the prediction, indicating the climax stage of the event. Then, the slope begins to decrease, indicating that the growth of influence is slowing down. Finally, on the twelfth day, the slope approaches 0, indicating that the influence of the event has basically stopped changing and the spread of the event is nearing its end.

[0054] It is understandable that by comprehensively considering the multi-dimensional target feature data, adapting different preset rules according to the data characteristics of each dimension, and mapping them to the same value range, not only a quantifiable and comparable first impact index is obtained, but also the impact assessment is more comprehensive and accurate, which helps relevant departments to monitor the trend of events, compare the impact of different events, and take public relations measures in a timely manner. In addition, by predicting the degree of change of the first impact index of each target event in the second time period, it is possible to provide early warnings for relevant departments, making the decision-making of relevant departments more data-driven and scientific.

[0055] On the basis of the above embodiments, in this embodiment, based on the target feature data of each target text, determining the first influence index of each target event through preset rules includes: based on the target feature data of the target text in each dimension, calculating the second influence index of the target text in each dimension through preset rules, the second influence index is used to characterize the influence of the target text in a certain dimension; aggregating the second influence indexes of all dimensions of the target text to obtain the third influence index of the target text, the third influence index is used to characterize the influence of the target text; aggregating the third influence index of each target text to obtain the first influence index of the target event.

[0056] For each target text, according to its target feature data in each dimension, first calculate the second influence index of each target text in each dimension through preset rules to obtain the influence of the target text in each dimension. Then, aggregate the second influence index of each target text in all dimensions to obtain the third influence index of the target text. The aggregation method may include weighted average, summation and maximum value selection, etc., depending on the degree of emphasis on the influence of different dimensions and the design of preset rules, to obtain the overall influence of the target text. Finally, aggregate the third influence index of all target texts associated with the target event to obtain the first influence index of the target event.

[0057] For example, when the target event is all the relevant tweets released by a brand in a certain day, it is hoped to determine the reputation influence of the brand in this day by analyzing these tweets. In this process, first collect all the relevant tweets released by the brand in the target time period (i.e., one day) as the target text. For each target text, extract its target feature data in dimensions such as the number of brand mentions, sentiment tendency (positive, negative, neutral), and interactive data. Assume that different weights (such as 0.3, 0.2, and 0.5) are set for the number of brand mentions, sentiment tendency, and interactive data (such as forwarding volume and number of comments), and the sentiment tendency is standardized (such as 1 for positive, -1 for negative, and 0 for neutral). Then, for each target text, calculate its second influence index in each dimension based on these weights and standardized data. Then, the second influence index of each target text in all dimensions is weighted averaged to obtain the third influence index of the target text. Finally, sum the third influence indexes of all target texts to obtain the first influence index of the target event (i.e., all the relevant tweets released by the brand in one day).

[0058] It should be noted that for details not included in this embodiment, please refer to Figure 2 The described embodiment.

[0059] It can be understood that through the setting of the second impact index and the third impact index, not only the performance of the target text in various dimensions is taken into consideration, but also the influence of each dimension is effectively integrated through reasonable aggregation rules, thereby achieving a comprehensive evaluation of the overall influence of the target text, thereby obtaining a more representative influence indicator of the target event.

[0060] Based on the above embodiments, in this embodiment, at least one target text associated with at least one target event within a first time period is obtained, and target feature data of at least two dimensions are extracted from each target text, including: extracting multiple target phrases in each target text; and classifying each target phrase based on at least two predefined dimensions to obtain at least one target feature data for each dimension.

[0061] In the first time period, at least one target text associated with at least one target event can be obtained from multiple channels through crawler technology and API interface, and then each target text can be preprocessed, such as removing stop words, punctuation marks and HTML tags, and then using natural language processing technology (such as word segmentation, part-of-speech tagging, etc.) to extract multiple target phrases in the text, such as keywords, phrases and entity names, to reflect the theme and content of the text. Furthermore, based on at least two predefined dimensions (such as relevance dimension, sentiment dimension, etc.), machine learning algorithms or rules can be used to classify the extracted target phrases so that at least one target feature data is obtained under each dimension. Finally, the target feature data under each dimension can be quantified according to specific needs. For example, the weight of the phrase can be calculated using technologies such as TF-IDF and word embedding, or the sentiment tendency of the text can be calculated using sentiment analysis algorithms.

[0062] For example, in a "news report about the brand's new product launch", the word segmentation technology is used to extract phrases such as "brand name", "new product launch" and "features". Then, these phrases are classified into the "relevance" dimension, and the weight of each phrase is calculated. Finally, based on the weights of these phrases and their positions and number of occurrences in the text, a comprehensive score is calculated as the target feature data under this dimension. Then, these target feature data are calculated according to preset rules, and the calculation results are mapped to the same value range (such as 0-20) to obtain the first impact index of the relevance dimension.

[0063] It should be noted that for details not included in this embodiment, please refer to the implementation details of the aforementioned embodiment.

[0064] It is understandable that extracting target phrases from the target text and classifying and quantifying them can comprehensively and meticulously parse the text content, while helping to deeply understand all aspects of the target event and provide a more comprehensive basis for subsequent data analysis and event prediction.

[0065] Based on the above embodiments, in this embodiment, the method also includes: monitoring the target feature data of the target dimension in each dimension; when the value of the target feature data of the target dimension is greater than or equal to a preset threshold, obtaining a weighting factor of the target event, the weighting factor is used to characterize the degree of influence of the target feature data of the target dimension on the target event; based on the weighting factor, updating the first influence index.

[0066] When evaluating the influence of the target event, the target feature data of the target dimension in each dimension can also be continuously monitored, where the target dimension can be the communication media, creator, content quality, and communication interaction, etc. The target feature data of the communication interaction dimension can be the number of hot searches. By collecting the data of the target dimension in real time or regularly, and judging whether the target feature data reaches or exceeds the preset threshold (such as the number of hot searches is greater than or equal to 1), it can be known whether the influence of this dimension on the target event is significant, so as to further consider its degree of influence. When the feature data of the target dimension reaches a significant level, the weighting factor of the target event can be calculated, and then, based on the calculated weighting factor, the initial first impact index can be updated and adjusted.

[0067] For example, if the target event appears on the hot search list in the communication interaction dimension, it means that the communication coverage is wider and has a greater promoting effect on the overall influence. At this time, the event hot search weighting factor can be determined.

[0068]

[0069] Among them, hot_search_factor represents the weighting factor of the event hot search, n represents the number of hot searches for the event, and r highest Indicates the highest ranking of the event in the hot search list, r avg represents the average ranking of the event on the hot search list, wci represents the average popularity of each platform (if any), α, , and Represent the corresponding weights respectively.

[0070] After determining the weighting factor, the second impact index of the target dimension can be adjusted, and then the adjusted second impact index can be re-aggregated to obtain the third impact index of the target text, and finally the first impact index of the target event can be obtained. When the weighting factor has an impact on the overall target event, the determined first impact index can also be directly adjusted. Finally, based on the adjusted first impact index, the degree of change of the first impact index in the future time period can be predicted.

[0071] It can be understood that by introducing the weighting factor, the overall impact of this change on the target event can be more accurately reflected, thereby making the first impact index more dynamic and accurate.

[0072] In an embodiment of the present disclosure, the dimension includes at least the degree of relevance between the target text and the target object. Based on the target feature data of the target text in each dimension, the second influence index of the target text in each dimension is calculated by preset rules, including: based on the mention position, number of mentions and mention depth of the target object at least in the target text, obtaining target feature data with different weights in the relevance dimension; based on the target feature data with different weights in the relevance dimension, calculating the mention rate of the target object in the target text by using preset rules, normalizing the mention rate, and obtaining the second influence index in the relevance dimension.

[0073] When evaluating the relevance between the target text and the target object (such as a brand, person, or event), the three key factors of the target object's mention position, number of mentions, and mention depth in the target text can be comprehensively considered. Each factor has a different weight to reflect its importance in evaluating the relevance. Among them, the mention position refers to the position where the target object appears in the target text. Generally, the content that appears in the title, introduction, or conclusion is often more important than other parts in the text. The number of mentions refers to the number of times the target object is mentioned in the target text. The more times it is mentioned, the higher the relevance of the target object to the text content and the greater its importance in the text. The mention depth refers to the level of detail and contextual relevance when the target object is mentioned in the text. For example, whether a brand is mentioned with a detailed description of its products, services, or image, or whether an event is mentioned with detailed information such as its background, process, and results.

[0074] In some exemplary embodiments, when it is necessary to measure the relevance of the target text to the brand (target object), the number of core keywords of each brand in the article body and title is first counted. Different weights can be given to these keywords according to the location where the keywords appear (such as the title, the beginning of the text, the middle of the text, and the end of the text), whether they are related to the product, and the high-level level mentioned (such as the mention of senior management personnel such as CEO and chairman). Then, the final discussion degree of the brand that the article mainly talks about in the whole text is calculated, that is, the number of keywords of the brand in the article divided by the total number of keywords of each brand in the article. If the article is related to the industry and involves multiple brands, the median can be used as the benchmark value of the brand discussion degree to eliminate the influence of extreme values. If only a single brand appears in the full text, but the overall frequency of the brand word in the full text is low, the brand's discussion score can be reduced by de-weighting to avoid evaluation bias caused by excessive mention of a single brand. Finally, the calculated brand discussion degree is normalized to obtain the second impact index in the relevance dimension.

[0075] It should be noted that for details not included in this embodiment, please refer to the implementation details of the aforementioned embodiment.

[0076] It can be understood that by considering the mention position, number of mentions and mention depth of the target feature data in the dimension of the relevance between the target text and the target object, and calculating the second influence index according to preset rules, it is possible to more accurately locate the scope of influence of the target object in the target text, which helps to identify key information points.

[0077] In an embodiment of the present disclosure, the dimension includes at least a target text sentiment dimension, and based on the target feature data of the target text in each dimension, the second influence index of the target text in each dimension is calculated by preset rules, including: based on the target feature data of the target text sentiment dimension, using a sentiment classification model to obtain a probability value of the sentiment tendency of the target text; normalizing the probability value to obtain the second influence index in the target text sentiment dimension.

[0078] On the basis of the above embodiments, in this embodiment, the emotional dimension of the target text can be considered. Specifically, based on the target feature data of the emotional dimension of the target text, a pre-trained sentiment classification model can be used to capture the emotional tendency of the text, and give probability values ​​p1, p2 and p3 belonging to positive, neutral or negative emotions. Among them, p1 represents the probability of positive emotion, p2 represents the probability of neutral emotion, and p3 represents the probability of negative emotion. Then, in order to convert the emotional tendency into a comparable second impact index, these probability values ​​can be normalized. Furthermore, the second impact index of other dimensions of the target text can be calculated, and then the second impact index of all dimensions can be aggregated to obtain the third impact index of the target text. Finally, the third impact index of all target texts is aggregated to obtain the first impact index of the target event.

[0079] It should be noted that when the emotional tendency of the target text is positive, the probability value p1 of the positive emotion is used for normalization as the second influence index of the text in the emotional dimension; when the emotional tendency is negative, the probability value p3 of the negative emotion is used for normalization as the second influence index. For neutral emotions, since it does not have obvious positive or negative tendencies, in practical applications, it is chosen to ignore its influence or normalize the positive probability value according to specific needs. It can be understood that by introducing the sentiment classification model and the normalization processing strategy, the influence of the target text in the emotional dimension can be effectively quantified and the accuracy of sentiment analysis can be improved.

[0080] In an embodiment of the present disclosure, a dimension includes at least a target text characteristic value, and based on the target feature data of the target text in each dimension, calculating the second influence index of the target text in each dimension by preset rules includes: obtaining a target identifier of the target text, the target identifier is at least used to characterize the dissemination medium and the creator of the target text; based on the target identifier, performing differentiated processing on the target feature data of the target text characteristic value dimension by using matching preset rules to obtain the second influence index of the target text in the target text characteristic value dimension.

[0081] When calculating the second impact index of the target text characteristic value dimension, the target identifier of the target text can be identified first. When the target identifier reveals a specific communication medium (communication platform), the target feature data is the number of interactions (likes, collections, comments, etc.), and then the target feature data can be calculated for different communication media through different preset rules. For example, for video platforms, the number of views, likes, collections, and comments can be weighted and summed; for chat software platforms, the number of readings and likes can be weighted and summed; for social media platforms, the number of reposts, comments, and likes can be weighted and summed; for other platforms, the number of views, likes, and comments can be weighted and summed. When the target identifier represents the creator, the creator can be regarded as the target feature data, and the media and self-media accounts of various mainstream social media platforms can be evaluated, and rating libraries can be established for core media and influencers respectively. Then, according to the record and quality of the account's past posts, each account can be given a corresponding priority, and then the priority can be mapped to convert the account priority into the second impact index.

[0082] It should be noted that for details not included in this embodiment, please refer to the implementation details of the aforementioned embodiment.

[0083] On the basis of the above embodiment, in this embodiment, the content quality dimension of the target text can be considered to measure the content richness and writing quality of the target text. Specifically, first, the target feature data of the target text can be calculated, such as the number of words, the number of sentences, the average number of words in a single sentence, the number of named entities (such as names, place names, organization names, etc.), the number of new words, and the text information entropy. Then, due to the differences in user groups, content styles, and publishing rules of different social media platforms, the performance of the same target text on different platforms may also be different. In order to weaken the difference between such platforms, the above calculation results can be smoothed, that is, a certain amount of text data is randomly extracted from each platform as a sample, and then the sample data is processed to calculate the mean of each indicator (such as the number of words, the number of sentences, etc.), and then the actual calculation results of the target text are compared and adjusted with the mean of the corresponding platform to weaken the impact of platform differences on content quality scores. Finally, reasonable weights can be assigned to each indicator based on experience or data analysis results, and the actual value of each indicator can be multiplied by the corresponding weight to obtain the weighted indicator value. At the same time, the weighted index values ​​can be added to obtain a total score, and then the total score can be normalized (such as mapping the total score to between 0 and 1) to obtain a second impact index of the final content quality dimension. Based on the above data processing method, the present disclosure also provides a data processing device. The data processing device is described in detail below.

[0084] Figure 4 The block diagram schematically shows a data processing device according to an embodiment of the present disclosure.

[0085] like Figure 4 As shown, the data processing device 400 includes an extraction module 410 , an evaluation module 420 and a determination module 430 .

[0086] According to some embodiments of the present disclosure, the data processing device 400 may be used to implement the reference Figure 2 and Figure 3 The data processing method according to the embodiment of the present disclosure is described.

[0087] The extraction module 410 can execute, for example, S210, which is configured to obtain at least one target text associated with at least one target event within a first time period, and extract target feature data of at least two dimensions from each target text, where the dimensions are used to characterize different attributes contained in the target text.

[0088] The evaluation module 420 may execute, for example, S220, which is configured to determine the first influence index of each target event based on the target feature data of each target text by a preset rule, and map each first influence index to the same value interval, wherein the first influence index is used to characterize the influence of the target event.

[0089] The determination module 430 may execute, for example, S230, configured to predict the degree of change of the first influence index of each target event in a second time period based on the first influence index in the first time period after mapping, the second time period being later than the first time period.

[0090] For example, any multiple of the extraction module 410, the evaluation module 420, and the determination module 430 can be combined in one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the extraction module 410, the evaluation module 420, and the determination module 430 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware or in any appropriate combination of any of them. Alternatively, at least one of the extraction module 410, the evaluation module 420, and the determination module 430 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0091] It should be noted that the collection, storage, use, processing, transmission, provision, disclosure and application of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals. In the technical solution of this disclosure, the user's authorization or consent is obtained before obtaining or collecting user personal information.

[0092] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0093] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement the method of an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0094] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0095] A number of components in the electronic device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0096] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as data processing methods. For example, in some embodiments, the data processing method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the data processing method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the data processing method in any other appropriate manner (e.g., by means of firmware).

[0097] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0098] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0099] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0100] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0101] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0102] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. Among them, the server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0103] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0104] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A data processing method, comprising: Acquire at least one target text associated with at least one target event within a first time period, and extract target feature data of at least two dimensions from each of the target texts, where the dimensions are used to characterize different attributes contained in the target text; Based on the target feature data of each target text, determine the first influence index of each target event through a preset rule, and map each first influence index to the same value range, wherein the first influence index is used to characterize the influence of the target event; Based on the mapped first impact index in the first time period, a change degree of the first impact index of each target event in a second time period is predicted, and the second time period is later than the first time period.

2. According to the method of claim 1, the step of obtaining at least one target text associated with at least one target event within the first time period, and extracting target feature data of at least two dimensions from each target text comprises: extracting a plurality of target phrases from each of the target texts; Based on at least two predefined dimensions, each target phrase is classified to obtain at least one target feature data of each dimension.

3. According to the method of claim 1, the determining the first impact index of each target event by a preset rule based on the target feature data of each target text comprises: Based on the target feature data of the target text in each dimension, a second influence index of the target text in each dimension is calculated by the preset rule, wherein the second influence index is used to characterize the influence of the target text in a certain dimension; Aggregating the second influence indexes of all dimensions of the target text to obtain a third influence index of the target text, wherein the third influence index is used to characterize the influence of the target text; The third impact indexes of the target texts are aggregated to obtain the first impact index of the target event.

4. The method according to claim 3, wherein the dimension at least includes the relevance between the target text and the target object, and the second impact index of the target text in each dimension is calculated by the preset rule based on the target feature data of the target text in each dimension, comprising: Based on at least the mention position, the number of mentions and the mention depth of the target object in the target text, obtaining target feature data with different weights in the relevance dimension; Based on the target feature data with different weights in the relevance dimension, the mention rate of the target object in the target text is calculated using the preset rule, and the mention rate is normalized to obtain a second influence index in the relevance dimension.

5. According to the method of claim 3, the dimension at least includes a target text sentiment dimension, and the second impact index of the target text in each dimension is calculated by the preset rule based on the target feature data of the target text in each dimension, comprising: Based on the target feature data of the target text sentiment dimension, using a sentiment classification model, obtaining a probability value of the sentiment tendency of the target text; The probability value is normalized to obtain a second influence index in the sentiment dimension of the target text.

6. The method according to claim 3, wherein the dimension at least includes the target text characteristic value, and the calculation of the second impact index of the target text in each dimension by the preset rule based on the target feature data of the target text in each dimension comprises: Acquire a target identifier of the target text, wherein the target identifier is used to at least characterize a dissemination medium and a creator of the target text; Based on the target identifier, the target feature data of the target text characteristic value dimension is differentiated by using matching preset rules to obtain a second influence index of the target text in the target text characteristic value dimension.

7. The method according to claim 1, further comprising: Monitoring target characteristic data of a target dimension in each of the dimensions; When the value of the target feature data of the target dimension is greater than or equal to a preset threshold, a weighting factor of the target event is obtained, where the weighting factor is used to characterize the degree of influence of the target feature data of the target dimension on the target event; Based on the weighting factor, the first influence index is updated.

8. A data processing device, comprising: An extraction module is configured to obtain at least one target text associated with at least one target event within a first time period, and extract target feature data of at least two dimensions from each of the target texts, wherein the dimensions are used to characterize different attributes contained in the target text; An evaluation module is configured to determine a first influence index of each target event based on the target feature data of each target text by a preset rule, and map each first influence index to the same value interval, wherein the first influence index is used to characterize the influence of the target event; The determination module is configured to predict the degree of change of the first influence index of each target event in a second time period based on the first influence index in the first time period after mapping, and the second time period is later than the first time period.

9. An electronic device, comprising: one or more processors; a storage device for storing one or more programs, Wherein, when the one or more programs are executed by the one or more processors, the one or more processors are enabled to execute the acquisition of at least one target text associated with at least one target event within a first time period, and to extract target feature data of at least two dimensions from each of the target texts, wherein the dimensions are used to characterize different attributes contained in the target texts; based on the target feature data of each of the target texts, a first influence index of each of the target events is determined by a preset rule, and each of the first influence indexes is mapped to the same value range, wherein the first influence index is used to characterize the influence of the target event; Based on the mapped first impact index in the first time period, a change degree of the first impact index of each target event in a second time period is predicted, and the second time period is later than the first time period.

10. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to obtain at least one target text associated with at least one target event within a first time period, extract target feature data of at least two dimensions from each of the target texts, wherein the dimensions are used to characterize different attributes contained in the target texts; based on the target feature data of each of the target texts, determine a first influence index of each of the target events by a preset rule, and map each of the first influence indexes to the same value range, wherein the first influence index is used to characterize the influence of the target event; based on the mapped first influence index within the first time period, predict the degree of change of the first influence index of each of the target events in a second time period, wherein the second time period is later than the first time period.

Images