Information technology consultation management method based on big data

By analyzing social media data, identifying key influencers, and optimizing resource allocation, the problem of lagging market sentiment capture in IT consulting management was solved, enabling enterprises to respond agilely and optimize resources in a rapidly changing market.

CN119048157AInactive Publication Date: 2024-11-29LINYI CITY CLOUD BIG DATA CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411538854.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-31
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing IT consulting management lacks real-time data support in a rapidly changing market environment, making it difficult to capture subtle changes in market sentiment. This leads to delayed corporate decision-making and an inability to adjust strategies in a timely manner to maximize resource efficiency.

Method used

By collecting user-generated text from social media platforms, analyzing the sentiment and interaction data within the text, identifying key influencers, and generating resource allocation plans to optimize marketing strategies.

Benefits of technology

It enables enterprises to identify and accurately predict market trends in a timely manner, optimize resource allocation, and improve market competitiveness and brand awareness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048157A_ABST
    Figure CN119048157A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of resource planning, in particular to an information technology consultation management method based on big data, and the method comprises the steps: collecting a user generated text on a social media platform, recording a timestamp of the text, extracting the number of likes in the text, extracting the number of sharing in the text, and extracting the number of comments in the text. And summarizing the subject popularity data based on the text contents and the associated tags to obtain a social media data set. According to the invention, the enterprise can identify the market trend more accurately by recording the user interaction and the emotional response, so that the resource configuration and the strategy planning are optimized. For example, through classification of emotion scores and monitoring of emotion changes, enterprises can adjust market strategies in time so as to cope with emotion fluctuation of consumers and solve possible market crisis or grasp sudden market opportunities in advance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of resource planning, and in particular to an information technology consulting management method based on big data. Background Art

[0002] Information technology consulting management focuses on how to support and improve the planning and management of enterprise resources through information technology. This includes evaluating existing IT resources, designing and implementing solutions to optimize the use of these resources, and providing strategic advice to help enterprises achieve their business goals. The purpose of information technology consulting management is mainly to provide professional knowledge and technical support to help enterprises make reasonable decisions in a complex technological environment, thereby improving operational efficiency and market competitiveness.

[0003] However, existing information technology consulting management relies on internal data and traditional market research in operation, which limits the ability of enterprises to respond swiftly in a rapidly changing market environment. At the same time, the lack of real-time data support makes it difficult to capture subtle changes in market sentiment, resulting in enterprises not being able to predict market trends in a timely or accurate manner, thus missing the best time to adjust strategies. For example, without immediate and extensive market sentiment analysis, enterprises may not be able to effectively identify or utilize market dynamics driven by social media. In this case, the decision-making process of the enterprise may be too conservative or lagging, unable to maximize the benefits of resources, thus affecting overall market competitiveness and profitability. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an information technology consulting management method based on big data.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: An information technology consulting management method based on big data comprises: Collect user-generated texts on social media platforms, record the timestamps of the texts, extract the number of likes in the texts, extract the number of shares in the texts, extract the number of comments in the texts, summarize the topic popularity data based on the text content and associated tags, and obtain a social media dataset; According to the social media dataset, user-generated text is extracted, the text is decomposed into words, sentiment values ​​are assigned to words matching sentiment dictionary entries, the sentiment values ​​of the words in the text are summarized to calculate the sentiment score, the text is classified as positive when the score exceeds a threshold, is classified as negative when the score is below the threshold, and is classified as neutral when the score is within the threshold range, and the number of texts in each sentiment category in each time period is counted to obtain a market sentiment index; According to the market sentiment index, the number of positive, negative and neutral texts in each time period is recorded, the rate of change of the number between adjacent time periods is calculated, the rate of change is compared to identify the upward or downward trend, the keywords in the text with sentiment changes are extracted, the keywords corresponding to the growth or decrease trend are determined, and the market trend insight results are obtained; Based on the market trend insight results, identify users who frequently use trend keywords, retrieve the number of user interactions, measure network relationships by counting followers and connections, compare users based on the number of interactions and network measurements, select users with the highest scores, and obtain key influencer profiles; Based on the key influencer profile, available resource allocation options are matched with the identified key influencers, resources are allocated to channels frequently used by the key influencers, resource planning parameters are adjusted to reflect the new allocation, and a resource allocation plan is generated.

[0006] Preferably, the steps of acquiring the social media dataset are: Collecting user-generated text content from multiple social media platforms and recording the timestamp, number of likes, number of shares, and number of comments on the text, where the timestamp, number of likes, number of shares, and number of comments data are recorded as social media activity records; Based on the social media activity records, summarize and analyze text content and associated tags, extract and classify topic information, and generate topic classification records; According to the topic classification records, the popularity value of each topic is counted and calculated, including the cumulative number of likes, shares and comments of the text under each topic, using the formula: in Respectively represent the number of likes, shares and comments on a topic within a period of time. are the weights of social indicators, The number of texts representing the topic, generating the popularity data for each topic; The popularity data of each topic is aggregated to generate a social media dataset.

[0007] Preferably, the steps of obtaining the market sentiment index are: Extracting user-generated text from the social media dataset, breaking the text into words, comparing each word with a sentiment dictionary, identifying words that match sentiment dictionary entries, assigning sentiment values ​​to the words, and generating sentiment word analysis records; According to the sentiment word analysis record, the sentiment values ​​of the words in the text are summarized, and the sentiment score of each text is calculated. The score calculation adopts the weighted average method, in which the importance and frequency of the sentiment words determine the weight, and the sentiment score record of each text is generated; Using the sentiment score record of each text, a threshold is set, and texts with sentiment scores exceeding the threshold are classified as positive, those below the threshold are classified as negative, and those within the threshold range are classified as neutral, thereby generating a text sentiment classification record; According to the text sentiment classification records, the number of positive, negative and neutral texts in each time period is counted, and the rate of change of the number between adjacent time periods is calculated to obtain the market sentiment index.

[0008] Preferably, the specific steps of calculating the rate of change of the quantity between adjacent time periods and comparing the rate of change to identify the upward or downward trend are: Extracting the number of positive, negative and neutral texts in each time period from the market sentiment index to generate sentiment text statistical records; Using the sentiment text statistics records, calculate the rate of change of the number of positive, negative, and neutral texts between adjacent time periods, applying the formula: Get the change rate of each emotion category and generate the change rate record, where: represents the rate of change, Represents the amount of text in the current time period. Represents the number of texts in the previous time period; According to the change rate record, compare the change rate of each time period, identify the rise or fall of the sentiment trend, and generate a trend analysis record; The trend analysis records are summarized to obtain the market trend change rate.

[0009] Preferably, the steps for obtaining the market trend insight results are: Using the market trend change rate, determine the time period when significant changes occur, and generate a significant change time record; Extract keywords from the text in the significant change time record and apply the formula: Calculate the importance score of the keyword and generate a keyword importance record, where: Represents the importance score of the keyword. Represents the frequency of occurrence of keywords. Represents the number of documents containing the keyword; Determine keywords corresponding to an increasing or decreasing trend according to the keyword importance record, and generate a trend keyword record; The trend keywords are aggregated to obtain market trend insight results.

[0010] Preferably, the steps of obtaining the key influencer profile are: Using the market trend insight results, identifying users who frequently use trend keywords on social media through keyword tracking, and generating user keyword usage records; According to the user keyword usage record, retrieve the user's interaction quantity, including the number of comments, reposts and likes, count the interaction data, and generate the user interaction statistics record; From the user interaction statistics, the number of followers and the number of connections are statistically analyzed to measure the network relationship strength of the user, using the formula: Determine the user's influence in the social network and generate user network measurement records, where: Represents the number of followers, Represents the number of connections. represents the total number of interactions; According to the user network measurement records, users with the highest network relationship strength are screened to generate key influencer profiles.

[0011] Preferably, the steps of obtaining the resource allocation scheme are: Using the key influencer profiles, identify and match available resource allocation options, including advertising budgets, promotional materials, and access to social media tools, and generate resource matching records; Allocate resources to the channels most frequently used by key influencers, including Weibo, WeChat and Tik Tok. Determine the method of resource allocation based on the activity frequency and fan interaction data of key influencers and generate a resource allocation plan.

[0012] Preferably, the step of acquiring the resource allocation scheme further comprises: adjusting the duration, frequency and positioning of the target group of advertisement delivery according to the resource delivery plan to obtain the resource allocation scheme.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by collecting and analyzing social media data, enterprises are allowed to understand and predict market dynamics in detail. Specifically, by recording user interactions and emotional responses, enterprises can more accurately identify market trends, thereby optimizing resource allocation and strategic planning. For example, by categorizing emotional scores and monitoring emotional changes, enterprises can adjust market strategies in a timely manner to respond to fluctuations in consumer emotions, resolve possible market crises in advance, or seize sudden market opportunities. In addition, by identifying and utilizing key influencers, enterprises can not only allocate marketing resources more effectively, but also expand market influence and brand awareness through these key nodes, enabling enterprises to stay ahead in a highly competitive market. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0016] See also Figure 1 The present invention provides a technical solution: an information technology consulting management method based on big data includes: Collect user-generated texts on social media platforms, record the timestamps of the texts, extract the number of likes in the texts, extract the number of shares in the texts, extract the number of comments in the texts, summarize the topic popularity data based on the text content and associated tags, and obtain a social media dataset; Based on the social media dataset, we extract user-generated texts, decompose the texts into words, assign sentiment values ​​to words that match sentiment dictionary entries, summarize the sentiment values ​​of the words in the text and calculate the sentiment score. When the score exceeds the threshold, the text is classified as positive, when the score is below the threshold, it is classified as negative, and when the score is within the threshold range, it is classified as neutral. The number of texts in each sentiment category in each time period is counted to obtain the market sentiment index. According to the market sentiment index, record the number of positive, negative and neutral texts in each time period, calculate the rate of change between adjacent time periods, compare the rate of change to identify the upward or downward trend, extract keywords from the texts with sentiment changes, determine the keywords corresponding to the growth or decrease trend, and obtain market trend insights; Based on the market trend insights, identify users who frequently use trending keywords, retrieve the number of user interactions, measure network relationships by counting followers and connections, compare users based on the number of interactions and network measurements, select the users with the highest scores, and obtain key influencer profiles; Based on the key influencer profiles, match available resource allocation options with identified key influencers, allocate resources to channels frequently used by key influencers, adjust resource planning parameters to reflect the new allocation, and generate a resource allocation plan.

[0017] The steps to obtain the social media dataset are: Collecting user-generated text content from multiple social media platforms and recording the timestamp, number of likes, number of shares, and number of comments on the text, where the timestamp, number of likes, number of shares, and number of comments data are recorded as social media activity records; Based on social media activity records, summarize and analyze text content and related tags, extract and classify topic information, and generate topic classification records; According to the topic classification records, count and calculate the popularity value of each topic, including calculating the cumulative number of likes, shares and comments of the text under each topic, using the formula: in Respectively represent the number of likes, shares and comments on a topic within a period of time. are the weights of social indicators, The number of texts representing the topic, generating the popularity data for each topic; The popularity data of each topic is aggregated to generate a social media dataset.

[0018] Specifically, we collect text content generated by users and obtain user interaction data from social media platforms, including but not limited to the timestamp of the text, the number of likes, the number of shares, and the number of comments. The collection of these data forms a preliminary record of social media activity. This record not only contains the direct amount of user interaction, but also provides a temporal analysis basis for the data through the timestamp, thereby tracking the popularity trend and changes of specific content. In this process, various data are automatically collected through database queries and API calls. Each data is indexed by timestamp to ensure the efficiency and accuracy of data processing. The generated social media activity record serves as the basic data for subsequent analysis.

[0019] Analyze text content and associated tags, and use topic modeling technology, such as the Latent Dirichlet Allocation (LDA) model, to classify social media data into topics. The data collected in this process is converted into a bag-of-words model, and each text is regarded as an unordered collection of word frequencies. Through the statistical analysis model, frequently occurring keywords and their associated tags are identified. These keywords and tags are used to define topics, and then the distribution probability of each topic in different texts is calculated, so as to classify the text into the most likely topic. The generated topic classification records provide specific text and user behavior indicators for each topic, providing preparation for further heat analysis.

[0020] formula The calculation process: Set each weight parameter , which indicates the relative importance of likes, shares, and comments in the total popularity calculation. Collect actual data for a period of time, the number of likes under a certain topic , number of shares , number of comments , the number of texts related to the topic , put these values ​​into the formula for calculation:

[0021]

[0022]

[0023] The result shows that, considering the different weights of each interactive behavior, the topic heat value obtained is 0.1245. This value can be used to evaluate the comprehensive influence of the topic on social media, and then analyze the user participation and popularity of the topic based on this heat value.

[0024] Generate an overall social media dataset, which summarizes the popularity data of all topics, reflecting the popularity trends and user engagement of different social media topics. The entire dataset is obtained by aggregating the popularity values ​​of each topic. During the aggregation process, the popularity data of each topic is weighted averaged. Considering the number of users and time factors, the time-weighted average method is used to ensure that the latest data has a higher influence. The obtained dataset is used to formulate marketing strategies and content promotion, providing companies with targeted market insights.

[0025] The steps to obtain the market sentiment indicator are: Extract user-generated text from social media datasets and decompose the text into words, compare each word with the sentiment dictionary, identify words that match sentiment dictionary entries, assign sentiment values ​​to the words, and generate sentiment word analysis records; According to the sentiment word analysis records, the sentiment values ​​of the words in the text are summarized and the sentiment score of each text is calculated. The score calculation adopts the weighted average method, in which the importance and frequency of the sentiment words determine the weight, and the sentiment score record of each text is generated; Using the sentiment score record of each text, a threshold is set, and texts with sentiment scores exceeding the threshold are classified as positive, those below the threshold are classified as negative, and those within the threshold range are classified as neutral, generating a text sentiment classification record; According to the text sentiment classification records, the number of positive, negative and neutral texts in each time period is counted, and the rate of change of the number between adjacent time periods is calculated to obtain the market sentiment index.

[0026] Specifically, user-generated texts are extracted from the aforementioned social media datasets and decomposed into words. Key information in the text, such as timestamps, number of likes, number of shares, and number of comments, is used as the basic data for analysis. Natural language processing technology is used to decompose the text content into individual words. The comparison between words and sentiment dictionaries uses standard dictionary matching technology to identify words that match sentiment dictionary entries and assign sentiment values ​​to these words. The assignment of sentiment values ​​is based on the frequency of occurrence and sentiment intensity of each word. The measurement of sentiment intensity uses a convolutional neural network model. The execution of these steps ensures the accuracy and effectiveness of sentiment analysis and generates sentiment word analysis records.

[0027] According to the sentiment word analysis records, the sentiment values ​​of the words in each text are summarized. The sentiment score is calculated using the weighted average method, which combines the sentiment value of each word with its frequency of occurrence in the text to ensure the accuracy of the score. The importance is determined based on the classification of the word in the sentiment dictionary and the frequency of its use in the context. The automation of this process ensures the efficiency of the analysis and generates a sentiment score record for each text.

[0028] Using the sentiment score record of each text, the set threshold is determined through data analysis to reflect the sentiment differences of different types of text. The sentiment classification method of the text adopts the convolutional neural network model. The convolutional neural network model distinguishes positive, negative and neutral emotions through the set threshold. This classification process is based on the sentiment score. The calculation of the score reflects the diversity of user emotions in detail and generates a text sentiment classification record.

[0029] According to the text sentiment classification records, the number of texts in each time period is counted, including positive, negative and neutral. The quantity statistics use database query technology to ensure the real-time and accuracy of the data. By calculating the rate of change of quantity between adjacent time periods, the standard ratio calculation method is used. This statistical method can effectively capture the changes in sentiment trends, thereby accurately reflecting the sentiment trends on social media and generating market sentiment indicators.

[0030] The specific steps to calculate the rate of change of quantity between adjacent time periods and compare the rate of change to identify the upward or downward trend are: Extract the number of positive, negative and neutral texts in each time period from the market sentiment index to generate sentiment text statistics records; Using sentiment text statistics records, calculate the rate of change of the number of positive, negative, and neutral texts between adjacent time periods, applying the formula: Get the change rate of each emotion category and generate the change rate record, where: represents the rate of change, Represents the amount of text in the current time period. Represents the number of texts in the previous time period; According to the change rate records, compare the change rate of each time period, identify the rise or fall of sentiment trends, and generate trend analysis records; Summarize the trend analysis records to obtain the market trend change rate.

[0031] Specifically, the number of positive, negative and neutral texts in each time period is extracted from the market sentiment index. First, according to the data interface provided by the social media platform, the text data posted by users in each time period is obtained through API calls, including the sentiment tendency of the text. The text data includes three categories: positive, negative and neutral. Each type of text is marked with a specific quantity. Then, the database query language is used to screen the data, aggregate the data by time period, calculate the total number of each type of sentiment text, and record the data including the time label and the number of corresponding sentiment texts. This step ensures the accuracy and completeness of the data. Finally, a sentiment text statistical record is generated, which lists in detail the number of each type of sentiment text in each time period.

[0032] formula The calculation process: Using sentiment text statistics, calculate the rate of change of the number of positive, negative, and neutral texts between adjacent time periods. To this end, first determine the total number of texts in each time period. For example, suppose that in time period t, the number of positive texts is 150, and in the previous time period t-1, the number is 100. According to the formula: The calculated rate of change is 0.5, which means that the number of positive emotional texts has increased by 50%. This numerical result shows that from time period t-1 to time period t, the market's acceptance of positive content has increased significantly. The identification of this trend helps to understand changes in market sentiment.

[0033] According to the change rate records, the change rates of different time periods are compared. First, the calculated sentiment category change rates of each time period are extracted from the database, including the change rates of three types of texts: positive, negative, and neutral. Through data comparison and analysis, determine which time periods show significant sentiment changes. For example, if the change rate of positive sentiment texts continues to rise for several consecutive time periods, it indicates that market sentiment is tending to be optimistic. Otherwise, it may indicate that market sentiment is tending to be pessimistic or stable. This step ensures accurate identification and analysis of trends through sophisticated data analysis, and generates trend analysis records.

[0034] Summarize the trend analysis records to obtain the market trend change rate. First, integrate the sentiment trend data of all time periods, including the change rate of sentiment categories in each time period. Calculate the overall market sentiment trend through statistical methods, such as weighted average, etc. This integration process involves data comparison and aggregation of multiple time periods, ensuring the comprehensiveness and representativeness of the results. Ultimately, the generated market trend change rate reflects the overall trend of market sentiment and provides a basis for the formulation of market strategies.

[0035] The steps to obtain market trend insights are as follows: Use the market trend change rate to determine the time period when significant changes occur and generate a record of significant change times; Extract keywords from the text in the significant change time record and apply the formula: Calculate the importance score of the keyword and generate a keyword importance record, where: Represents the importance score of the keyword. Represents the frequency of occurrence of keywords. Represents the number of documents containing the keyword; According to the keyword importance records, determine the keywords corresponding to the growth or decrease trend, and generate trend keyword records; Summarize trend keywords to obtain market trend insights.

[0036] Specifically, the market trend change rate is used to determine the time period with significant changes. The sentiment change rate data is analyzed to locate the moments with large sentiment fluctuations within the time period. The labels of these time periods are recorded in the database to ensure that each data record contains a clear time mark, thereby achieving accurate screening and classification of data. Through further data query and conditional filtering, the set of time periods that meet the significant changes are screened out to generate significant change time records.

[0037] formula The calculation process: in is the weighted importance score of the keyword in the specified time period. represents the frequency of the keyword appearing in the text during that time period, and is the document frequency, which indicates the total number of documents containing the keyword; the text frequency of a keyword in this time period is 50, and the document frequency is 10. Substituting these two values ​​into the formula yields: The result shows that the keyword has a high importance in the text of this time period. The size of this value reflects the market relevance of the keyword, and a keyword importance record is generated for subsequent analysis.

[0038] Based on the keyword importance records, determine the keywords with corresponding growth or decrease trends, analyze the frequency of keyword occurrence in the records and its changes over time, and identify keywords with significant fluctuations by comparing the keyword importance scores of this time period with the previous and next periods. Summarize and mark the data of these keywords with significant growth or decrease trends to generate trend keyword records.

[0039] Summarize trend keywords to obtain market trend insights. First, arrange all records marked as trend keywords in chronological order. According to the time when the keywords appear and the distribution of the text, perform layered group analysis on the keywords with trend characteristics. Comprehensively compare the change rates of keyword scores in different time periods. Finally, classify and summarize the market trend insights to provide a reference for subsequent market analysis.

[0040] The steps to obtain key influencer profiles are: Use market trend insights to identify users who frequently use trending keywords on social media through keyword tracking and generate user keyword usage records; According to the user's keyword usage records, retrieve the user's interaction number, including comments, reposts and likes, count the interaction data, and generate user interaction statistics records; From the user interaction statistics, the number of followers and connections are statistically analyzed to measure the user's network relationship strength using the formula: Determine the user's influence in the social network and generate user network measurement records, where: Represents the number of followers, Represents the number of connections. represents the total number of interactions; Based on the user network measurement records, users with the highest network relationship strength are screened to generate key influencer profiles.

[0041] Specifically, based on the results of market trend insights, by tracking and analyzing user behavior and content on social media, the system will identify users who frequently use specific trend keywords. These users reflect specific market dynamics or public sentiment through their speeches and interactions. The identification process includes data crawling and keyword matching, which requires regular collection and updating of data on social media platforms, and the use of natural language processing technology, especially keyword extraction and frequency analysis, to determine which users dominate the discussion. In this way, a keyword-user database can be constructed, which contains information such as the number of times each user speaks, the frequency of using specific keywords, etc., thereby providing basic data for subsequent influence analysis and user rankings.

[0042] Through the user keyword usage records, the number of user interactions is retrieved, including comments, reposts and likes. This process involves in-depth analysis of social media data and requires statistics on each user's interactive behavior within a specific time frame. This includes crawling data from social media platforms, using APIs to obtain user activity data, and then removing invalid or erroneous data records through data cleaning and preprocessing steps. After that, the database management system is used to store and query this data for fast data retrieval and calculation. By performing statistical analysis on the number of user interactions, the user's activity level and influence in social networks can be effectively revealed. This information will be used to evaluate the user's market influence and provide a basis for the next strategic decision.

[0043] formula The calculation process: Represents the number of followers of the user, Represents the number of connections for the user. Represents the total number of interactions of a user. The formula is used to evaluate the influence of a user in a social network by analyzing the interaction data on social media, such as the number of comments, reposts, and likes, which are collected from the user's public activities; User A has 120 followers, 200 connections, and 500 interactions (comments, reposts, and likes combined) in a time period. The calculation process is as follows: This result shows that every time user A conducts a social interaction, on average, he will influence approximately 0.64 network connections or followers. This value reflects that user A has a relatively high network influence, thus revealing the size of the user's influence in the social network.

[0044] Screening out users with the highest network relationship strength through user network measurement records involves a sorting and screening process. First, based on the network relationship strength values ​​of each user calculated previously, these values ​​are stored in a list, and then the list is sorted in descending order. By selecting a sort or quick sort algorithm, it is determined which users have the highest influence among all users. This process requires efficient data structures and algorithms to ensure processing speed and accuracy. After selecting the users with the highest influence, the detailed information of these users (such as user name, contact information, main social platform, etc.) is compiled into key influencer profiles. These profiles will be used for subsequent marketing strategies and brand cooperation, providing key data support for marketers to ensure that marketing activities can target the most influential social media users.

[0045] The steps to obtain the resource allocation plan are: Using key influencer profiles, identify and match available resource allocation options, including advertising budget, promotional materials and access to social media tools, and generate resource matching records; Allocate resources to the channels most frequently used by key influencers, including Weibo, WeChat, and TikTok. Determine the method of resource allocation based on the activity frequency and fan interaction data of key influencers and generate a resource allocation plan; According to the resource delivery plan, adjust the duration, frequency and positioning of the target group of the advertisement to obtain the resource allocation plan.

[0046] Specifically, in the process of identifying key influencers, it is first necessary to conduct an in-depth analysis of the data on social media platforms, including advertising budgets, promotional materials, and access to specific social media tools. These resource allocation options are matched with the specific needs of key influencers. To this end, it is necessary to collect and analyze various social media data, such as the frequency of posts, the number of interactions (likes, comments, and shares), and the use of keywords. Furthermore, based on the results of these data analyses, the most appropriate resource combination is determined. This process involves clustering analysis and priority sorting of data to ensure that each resource can be used most effectively in the right channel and time, thereby generating a resource matching record.

[0047] How resources are allocated is crucial based on the characteristics and activity frequency of key influencers. This process includes considering which social media channels (Weibo, WeChat and Tik Tok) are most suitable for delivery and which time periods are most likely to attract the attention of the target audience. For example, the user interaction data (comments, likes and reposts) of key influencers in different time periods, as well as the frequency of their content releases, are analyzed. Based on these data, a detailed resource delivery strategy is developed. This requires the use of data analysis technology to predict user behavior and analyze content trends, and ultimately determine the most effective advertising time and method to generate a resource delivery plan.

[0048] Based on the resource allocation plan, adjust the duration, frequency and target group positioning of the advertisement. This step involves real-time monitoring and analysis of market feedback data, including but not limited to changes in user click-through rate, conversion rate and interactive data. Through these data, optimize the content and form of the advertisement to ensure that each placement can achieve the highest market response. In addition, the adjustment process also needs to consider the characteristics of different platforms, such as the visual effects of video platforms and the information density of text platforms, so as to maximize the input-output ratio and obtain a resource allocation plan.

Claims

1. An information technology consulting management method based on big data, characterized in that: The method comprises: Collect user-generated texts on social media platforms, record the timestamps of the texts, extract the number of likes in the texts, extract the number of shares in the texts, extract the number of comments in the texts, summarize the topic popularity data based on the text content and associated tags, and obtain a social media dataset; According to the social media dataset, user-generated text is extracted, the text is decomposed into words, sentiment values ​​are assigned to words matching sentiment dictionary entries, the sentiment values ​​of the words in the text are summarized to calculate the sentiment score, the text is classified as positive when the score exceeds a threshold, is classified as negative when the score is below the threshold, and is classified as neutral when the score is within the threshold range, and the number of texts in each sentiment category in each time period is counted to obtain a market sentiment index; According to the market sentiment index, the number of positive, negative and neutral texts in each time period is recorded, the rate of change of the number between adjacent time periods is calculated, the rate of change is compared to identify the upward or downward trend, the keywords in the text with sentiment changes are extracted, the keywords corresponding to the growth or decrease trend are determined, and the market trend insight results are obtained; Based on the market trend insight results, identify users who frequently use trend keywords, retrieve the number of user interactions, measure network relationships by counting followers and connections, compare users based on the number of interactions and network measurements, select users with the highest scores, and obtain key influencer profiles; Based on the key influencer profile, available resource allocation options are matched with the identified key influencers, resources are allocated to channels frequently used by the key influencers, resource planning parameters are adjusted to reflect the new allocation, and a resource allocation plan is generated.

2. The information technology consulting management method based on big data according to claim 1 is characterized in that: The steps for obtaining the social media dataset are as follows: Collecting user-generated text content from multiple social media platforms and recording the timestamp, number of likes, number of shares, and number of comments on the text, where the timestamp, number of likes, number of shares, and number of comments data are recorded as social media activity records; Based on the social media activity records, summarize and analyze text content and associated tags, extract and classify topic information, and generate topic classification records; According to the topic classification records, the popularity value of each topic is counted and calculated, including the cumulative number of likes, shares and comments of the text under each topic, using the formula: in Respectively represent the number of likes, shares and comments on a topic within a period of time. are the weights of social indicators, The number of texts representing the topic, generating the popularity data for each topic; The popularity data of each topic is aggregated to generate a social media dataset.

3. The information technology consulting management method based on big data according to claim 1 is characterized in that: The steps for obtaining the market sentiment index are: Extracting user-generated text from the social media dataset, breaking the text into words, comparing each word with a sentiment dictionary, identifying words that match sentiment dictionary entries, assigning sentiment values ​​to the words, and generating sentiment word analysis records; According to the sentiment word analysis record, the sentiment values ​​of the words in the text are summarized, and the sentiment score of each text is calculated. The score calculation adopts the weighted average method, in which the importance and frequency of the sentiment words determine the weight, and the sentiment score record of each text is generated; Using the sentiment score record of each text, a threshold is set, and texts with sentiment scores exceeding the threshold are classified as positive, those below the threshold are classified as negative, and those within the threshold range are classified as neutral, thereby generating a text sentiment classification record; According to the text sentiment classification records, the number of positive, negative and neutral texts in each time period is counted, and the rate of change of the number between adjacent time periods is calculated to obtain the market sentiment index.

4. The information technology consulting management method based on big data according to claim 1 is characterized in that: The specific steps to calculate the rate of change of quantity between adjacent time periods and compare the rate of change to identify the upward or downward trend are: Extracting the number of positive, negative and neutral texts in each time period from the market sentiment index to generate sentiment text statistical records; Using the sentiment text statistics records, calculate the rate of change of the number of positive, negative, and neutral texts between adjacent time periods, applying the formula: Get the change rate of each emotion category and generate the change rate record, where: represents the rate of change, Represents the amount of text in the current time period. Represents the number of texts in the previous time period; According to the change rate record, compare the change rate of each time period, identify the rise or fall of the sentiment trend, and generate a trend analysis record; The trend analysis records are summarized to obtain the market trend change rate.

5. The information technology consulting management method based on big data according to claim 4 is characterized in that: The steps for obtaining the market trend insight results are as follows: Using the market trend change rate, determine the time period when significant changes occur, and generate a significant change time record; Extract keywords from the text in the significant change time record and apply the formula: Calculate the importance score of the keyword and generate a keyword importance record, where: Represents the importance score of the keyword. Represents the frequency of occurrence of keywords. Represents the number of documents containing the keyword; Determine keywords corresponding to an increasing or decreasing trend according to the keyword importance record, and generate a trend keyword record; The trend keywords are aggregated to obtain market trend insight results.

6. The information technology consulting management method based on big data according to claim 1 is characterized in that: The steps for obtaining the key influencer profile are as follows: Using the market trend insight results, identifying users who frequently use trend keywords on social media through keyword tracking, and generating user keyword usage records; According to the user keyword usage record, retrieve the user's interaction quantity, including the number of comments, reposts and likes, count the interaction data, and generate the user interaction statistics record; From the user interaction statistics, the number of followers and the number of connections are statistically analyzed to measure the network relationship strength of the user, using the formula: Determine the user's influence in the social network and generate user network measurement records, where: Represents the number of followers, Represents the number of connections. represents the total number of interactions; According to the user network measurement records, users with the highest network relationship strength are screened to generate key influencer profiles.

7. The information technology consulting management method based on big data according to claim 1 is characterized in that: The steps for obtaining the resource allocation scheme are: Using the key influencer profiles, identify and match available resource allocation options, including advertising budgets, promotional materials, and access to social media tools, and generate resource matching records; Allocate resources to the channels most frequently used by key influencers, including Weibo, WeChat and Tik Tok. Determine the method of resource allocation based on the activity frequency and fan interaction data of key influencers and generate a resource allocation plan.

8. The information technology consulting management method based on big data according to claim 7 is characterized in that: The step of acquiring the resource allocation scheme also includes: adjusting the duration, frequency and positioning of the target group of advertisement delivery according to the resource delivery plan to obtain the resource allocation scheme.

Citation Information

Cited By

  • Pig farm data intelligent monitoring and tracing system based on big data analysis

    CN121391026A